Union Pacific Resources Group, Inc. v. Rhône-Poulenc, Inc.

IN THE UNITED STATES COURT OF APPEALS
FOR THE FIFTH CIRCUIT

__________________________

No. 99-10326
__________________________

UNION PACIFIC RESOURCES GROUP, INC.;
BIG ISLAND TRONA CO.,
Plaintiffs-Appellants,

versus

RHÔNE-POULENC, INC.,
Defendant-Appellee.

___________________________________________________

Appeal from the United States District Court
for the Northern District of Texas

___________________________________________________

April 5, 2001

Before GARWOOD, WIENER, and DENNIS, Circuit Judges.

WIENER, Circuit Judge:

Union Pacific Resources Group, Inc. (“UPRG”) and its wholly

owned subsidiary, Big Island Trona Co. (“Big Island”),1 brought

this diversity action against Rhône-Poulenc, Inc. (“RPI”),

asserting claims of conversion, securities fraud, negligent

misrepresentation, and fraud. The district court granted summary

judgment for RPI, dismissing all claims, and UPRG timely appealed.

We conclude that summary judgment was properly granted on the

1
Hereafter, UPRG and Big Island will be referred to
collectively as “UPRG” unless otherwise noted.
conversion, negligent misrepresentation, and securities fraud

claims and affirm for essentially the reasons assigned in the

district court's well-reasoned opinion.2 Concluding that UPRG has

presented sufficient evidence to defeat RPI’s summary judgment

motion on the fraud claim, however, we reverse the district court’s

dismissal of that claim and remand for proceedings consistent with

this opinion.

I.

Facts and Proceedings

This action arose from RPI’s sale of all of its stock in two

of its subsidiaries, Rhône-Poulenc of Wyoming Holding Company

(“Holding”) and Rhône-Poulenc of Wyoming Company (“RPW”). The

purchaser was OCI America, Inc. (“OCI”). Prior to that sale, which

closed effective February 29, 1996, RPI owned 100 percent of the

outstanding stock of Holding and 80 percent of the outstanding

stock of RPW. UPRG owned the remaining 20 percent of the

outstanding stock of RPW.

Just over five years earlier, in an amended and restated

agreement dated December 5, 1991 (the “partnership agreement”),

UPRG, acting through Big Island, and RPI, acting through Holding,

had (along with RPW) formed Rhône-Poulenc of Wyoming Limited

Partnership (the “Wyoming partnership” or “Green River”). RPI’s

wholly owned subsidiary, Holding, was an initial general partner

2
See Union Pac. Resources Group, Inc. v. Rhône-Poulenc, Inc.,
45 F. Supp.2d 544 (S.D. Tex. 1999).

2
with a 50.49 percent majority interest in the Wyoming partnership;

UPRG’s wholly owned subsidiary, Big Island, was an initial general

partner with a 48.51 percent minority interest in the Wyoming

partnership. The remaining 1 percent was owned by RPW, as a

limited partner.3

The principal business of the Wyoming partnership was the

operation of a trona mine and manufacturing plant at Green River,

Wyoming.4 Under the partnership agreement, RPI’s subsidiary,

Holding, assumed the obligation to provide day-to-day management,

staffing, and technological support for Green River, as well as the

responsibility to prepare budgets and capital expenditure programs

for approval by the other partners. Through a series of

transactions, these obligations were indirectly assumed by RPI,

which performed the functions of managing partner for the Wyoming

partnership.5

RPI also agreed to make participation in two pension plans

3
By virtue of their respective ownership interests, RPI had a
51.29 percent equity interest in the Wyoming partnership (50.49
percent through Holding and 80 percent of 1 percent through RPW),
and UPRG had a 48.71 percent equity interest therein (48.51 percent
through Big Island and 20 percent of 1 percent through RPW). RPI’s
controlling interest in the Wyoming partnership was 51.49 percent
by virtue of the 50.49 percent ownership through Holding and its 80
percent control of RPW’s 1 percent.
4
Trona is a naturally occurring gray or white mineral used as
a source of sodium compounds, frequently in the manufacture of
glass.
5
Hereafter, RPI and Holding will be referred to collectively
as “RPI” unless otherwise noted.

3
(“the plans”) sponsored and administered by RPI available to all

eligible employees of Green River. One of those RPI plans, Plan

1674, was for salaried employees; the other, Plan 1679, was for

hourly employees. RPI did not maintain these plans exclusively for

Green River employees; on the contrary, a substantial number of

employees of RPI and others of its subsidiaries unrelated to Green

River were participants in the plans. Neither UPRG nor Big Island

was a sponsor of the plans: Neither company had legal control over

either plan, and neither company had the legal right of direct

access to any plan records or to any books, records, or reports

relating to them.

The plans were ERISA-qualified defined benefit plans,

entitling participating employees to fixed periodic payments at

retirement. In addition, as defined benefit plans, they owned a

common pool of assets rather than segregated accounts with separate

assets for individual participants or groups of individual

participants. Typically, all plan assets were held in trust for

the exclusive benefit of the plans’ participants and their

beneficiaries, and to defray RPI’s costs of administering the

plans.

To fund the plans for the participating Green River employees,

RPI (1) made contributions of its 51 percent ratable share directly

to the Wyoming partnership, and (2) charged UPRG’s 49 percent

4
ratable share to its partner’s account in the Wyoming partnership.6

Then, as managing partner, RPI caused a sum equal to the total of

those two amounts to be transferred in cash directly from the

Wyoming partnership to the trustees of the plans. According to

audited financial statements prepared for the Wyoming partnership

on a yearly basis, the plans were funded to “provide for benefits

attributable for services rendered to date, as well as for those

expected to be earned in the future” (emphasis added). This

indicates that RPI intended for the plans to be funded at the

higher “projected benefit obligation” (“PBO”) level, rather than at

the minimum “accrued benefit obligation” (“ABO”) level.7

In September, 1995 RPI contacted its actuaries, Coopers &

Lybrand (“Coopers”), by mail relative to what it referred to as a

“potential spinoff” of its interest in Green River, inquiring about

6
UPRG’s partnership contributions were apparently charged by
RPI on a lump-sum basis, with no details provided as to what
amounts, if any, were allocated to funding the plans.
7
“PBO”(Projected Benefit Obligation) and “ABO” (Accrued
Benefit Obligation) are actuarial terms that describe two different
levels at which the liability of pension plans can be calculated.
PBO is the actuarial present value of all benefits attributed to
past employee service, based on assumed future compensation levels.
ABO differs from PBO in not taking into account the effect of
actuarial projections of future salary. Thus, PBO is always larger
than ABO, but neither is necessarily an actual funding figure.
Based on the summary judgment record as supplemented by post-
argument filings ordered by this court, it is now clear that, at
all relevant times, the actual market value of the assets held by
the plans ranged somewhere between ABO and PBO, and on occasion
even exceeded PBO. An affidavit of Matt Sicking, an independent
actuary for UPRG, concluded that the Wyoming partnership’s
allocated pension cost was determined with PBO, not ABO.

5
the funding status of the plans. In a responding letter dated

October 13, 1995, Coopers provided RPI with the estimated value

of assets and liabilities of the plans as of August 1, 1995. These

estimates included calculations of the PBO and ABO of each plan.

According to Coopers’s estimates, the then-present market value of

assets in both plans exceeded the plans’ ABO levels substantially.

(Post-argument memoranda reveal that the actual funding level of

Plan 1679 was in excess of PBO on January 1, 1996 and at other

relevant times.)

Ten days later, on October 23, 1995, RPI formally notified

UPRG that it intended to sell all of its stock in Holding and RPW

and thereby transfer its entire interest in the Wyoming partnership

to OCI. As expressly required by the partnership agreement, RPI

officially informed UPRG of the interest that RPI proposed to sell

(100 percent of its interest in the Wyoming partnership), to whom

the interest would be sold (OCI), and the purchase price ($150

million).8 Under the partnership agreement, UPRG had a right of

8
Section 10.4(a) of the partnership agreement provides:

Should any Partner or Interest Holder (the “Seller”)
propose to transfer any of its Interest in the
Partnership, such Seller shall first obtain a bona fide
written offer (the “Purchase Offer”) for the purchase
thereof from a responsible purchaser (the “Purchaser”)
who is ready, willing, and able to purchase the same,
which offer shall provide for a purchase price payable in
cash of the lawful money of the United States of America
or other “hard” currency. The Seller shall give the
other Partners (the “Offerees”) written notice by
registered or certified mail, the notice to be
accompanied by a photostatic copy of such Purchase Offer.

6
first refusal (“ROFR”) to acquire the interest in the partnership

on the same terms for which RPI proposed to sell that interest.

The partnership agreement also required that the consent of the

non-transferring partners be obtained before the transferee of a

general partner could be admitted to the Wyoming partnership as a

general partner.

In addition, a shareholders’ agreement governing RPI’s and

UPRG’s stock ownership in RPW granted a ROFR to each shareholder in

the event that the another shareholder wished to sell some or all

of its stock in RPW. Under the terms of both the Wyoming

partnership agreement and the RPW shareholders’ agreement, UPRG had

thirty days following receipt of written notice of RPI’s intent to

sell within which to decide whether to exercise these ROFRs.

On the same day that RPI formally notified UPRG of the

proposed transfer of interest to OCI, it also sent to UPRG, under

separate cover, a copy of the 113-page draft stock purchase

agreement (the “purchase agreement”) between RPI and OCI. In the

letter of transmittal, the president of RPI assured UPRG that “[w]e

have emphasized to [OCI] the openness of our working relationship

with you, and as a result, the excellent quality of our partnership

meetings and discussion” (emphasis added). In correspondence with

UPRG approximately one week later, RPI stressed the fact that the

Such notice shall specify [1] the Percentage Interest to
be transferred (the “Offered Interest”), [2] the name of
the proposed purchaser, and [3] the price for which such
Offered Interest is proposed to be transferred.

7
partnership agreement did not “require that we deliver the Stock

Purchase Agreement since [the partnership agreement] specifies only

that we have to notify you of the interest to be transferred and

the identity of the purchaser.” RPI expressly reassured UPRG that

RPI had nevertheless “provided additional perspective and

information in the spirit of openness” (emphasis added). In both

letters, the word “openness” appears to operate as the functional

equivalent of the term “full disclosure.”

Three provisions of the extensive RPI/OCI purchase agreement

are directly relevant to this appeal. First, section 1.02 provides

that the cash purchase price for RPI’s interests in Holding and RPW

would be $150 million, “as such cash amount may be adjusted from

time to time, both prior to and after the Closing, pursuant to

section 1.03.” Second, section 1.03 provides that the purchase

price “shall be reduced” in an amount equal to 51 percent of the

difference between the ABO and PBO of the pension plans to the

extent allocable to coverage of Green River employees.

Third, section 8.04 of the purchase agreement required OCI to

ensure that Green River employees who were participants in Plans

1679 and 1674 immediately prior to closing would be covered by

“substantially similar” pension plans after the sale. As the

potential purchaser, OCI agreed to assume liability for the accrued

benefits of the Green River employees as those benefits came due

from and after closing. In section 8.04(b), RPI agreed that, as

soon as practicable after the closing, it would cause assets to be

8
transferred from its pension trust to OCI “in an amount equal to

the [ABO]” of the plans to the extent allocable to the

participating Green River employees. Although RPI had already

received from Coopers its determination of the ABO, PBO, and actual

market value of the plans as of August 1, 1995, RPI elected not to

share that information or subsequent updates of those data with

UPRG.

UPRG’s receipt of the draft RPI/OCI purchase agreement touched

off a flurry of correspondence between UPRG and RPI, concerned

primarily with the tolling of the thirty-day ROFR period,9 the

consent provisions of the partnership agreement, UPRG’s objections

to several particular provisions related to a $70 million note, and

UPRG’s requests for additional information about the purchase

agreement, specifically some exhibits and schedules referenced in

the purchase agreement that RPI had failed to attach.

Finally, on November 27, 1995, after fourteen letters

concerning various aspects of the transaction had passed among the

parties and OCI, UPRG confirmed to RPI that it would not exercise

9
After receiving the purchase agreement, UPRG informed RPI
that it had failed to attach many of the exhibits and schedules
referenced in the purchase agreement. UPRG also informed RPI that
“it is our view that the 30-day period for election . . . has not
commenced and will not commence until we receive a complete set of
documents that allows us to complete our review of the offer.”
Although RPI complied with UPRG’s request and provided the missing
documents, RPI insisted that the 30-day ROFR period had begun on
October 23, when formal notice of the divestiture had been sent to
UPRG, and would conclude on November 22.

9
its ROFRs. By letter dated November 28, 1995, UPRG formally

granted its consent to the proposed transaction, but expressly

conditioned its approval “on our review of the final executed Stock

Purchase Agreement, with complete exhibits and attachments, and a

determination that [the Stock Purchase Agreement] contains no terms

or provisions which would directly or indirectly affect any

existing rights or obligations of UPRG or impose any additional

liabilities” (emphasis added). On November 29, 1995, RPI and OCI

executed the purchase agreement. Two months later, RPI complied

with filing requirements of the Internal Revenue Service (“IRS”) by

giving notice of the intended transfer of assets from its pension

trusts to OCI’s new plans. Attached to each form was an actuarial

statement of valuation certifying that the transfers were in

accordance with IRS regulations.

Just before the February 29, 1996 closing, RPI received a

letter from Coopers dated February 22, warning RPI that, as

structured, the sale would result in the new OCI plans being

underfunded by an estimated $3-4 million. Coopers further

cautioned RPI that the Pension Benefit Guaranty Corporation

(“PBGC”), which regulates such matters, could require RPI to

contribute an additional $3-4 million into the new OCI plans.10

10
The February 22, 1996 letter from Coopers to RPI stated: “The
PBGC is focused on transactions that increase the PBGC’s financial
risk. In particular they are concerned with situations where
strong companies spin off weak subsidiaries along with underfunded
pension plans. This may be the case with the transaction (we
estimate that the spun off plans will be underfunded by $3 million

10
Coopers advised RPI on ways to avoid the underfunding problem: One

way would be to infuse at least $3-4 million into the OCI plans

after the transfer; another would be to agree to provide “credit”

to OCI should OCI be unable to meet its future funding commitments

to the plans.11 None of this information was furnished currently

to UPRG.

On February 28, 1996, the day before the effective date of the

closing, RPI and OCI updated the purchase agreement to specify the

amount of the reduction in purchase price provided for in section

1.03. The agreed amount of the reduction was $3,832,000, based on

actuarial calculations of ABO and PBO as of the closing date.12

Also, the purchase agreement was amended to acknowledge that “100

[percent] of the cost accrued by [RPI] for the provision of . . .

pension plan coverage for Wyoming Employees and Former Employees .

. . has been separately and specifically charged to, and has been

fully reimbursed by, the Wyoming Partnership.”13 None of this

information was furnished currently to UPRG.

to $4 million).”
11
RPI did not follow either of these suggestions from Coopers.
12
RPI’s actuaries arrived at this figure by first calculating
PBO for the Green River employees to be $20,464,087 and ABO to be
$13,070,205. The difference of $7,393,882 between PBO and ABO was
then multiplied by 51%, in accordance with section 1.03 of the
purchase agreement.
13
This provision was not in the draft purchase agreement that
RPI had furnished to UPRG, which classifies this as another example
of RPI’s pattern of selective disclosure calculated to avoid
alerting UPRG to potential problems for it.

11
 The sale from RPI to OCI closed effective February 29, 1996.

Weeks after the closing, a letter from RPI to Coopers dated March

18, 1996 confirmed that RPI had engaged a different actuarial firm,

Watson Wyatt Worldwide (“Watson”), to represent RPI “with respect

to the spin-off of assets from the [RPI] plans to the OCI plans.”14

Watson thus assumed responsibility for the final calculations of

the value of assets to be transferred from the old RPI plans to the

new OCI plans. The district court noted that

[o]n or about March 26, 1996, defendant gave
notice to the Internal Revenue Service that it
had appointed a new actuary, Watson Wyatt
Worldwide, and was filing new actuarial
statements of valuation for plans 1674 and
1679. Watson Wyatt Worldwide issued its
report on asset transfer, calculating the
amount to be transferred from defendant’s
retirement trust to OCI’s pension plans as of
March 28, 1996 to be $13,070,205.15

On March 28, 1996, RPI instructed the trustee of its pension plans

to transfer $13,070,205 to OCI’s plans. On March 29, 1996, RPI

filed notices of reportable events with the PBGC, reporting the

transfer of pension liabilities and benefits. On May 15, 1996,

Coopers, this time acting as actuaries for OCI, independently

recalculated the ABO and PBO figures and agreed that Watson’s

conclusions accurately reflected the plans’ formulae and agreed-to

14
The representation by counsel for RPI at oral argument that
this eleventh-hour switch was prompted solely by differences in
discount rate interpretation belies the fact that Coopers and
Watson used precisely the same rate of interest in their respective
calculations. Pretext inevitably raises the specter of mendacity.
15
See Union Pac. Resources Group, 45 F. Supp.2d at 550.

12
actuarial assumptions. None of this information was furnished

currently to UPRG.

The other shoe finally hit the floor when, approximately one

year after the transfer, OCI notified UPRG that it must make

additional contributions to the new OCI plans in the amount of $3-4

million —— not so coincidentally, 49 percent of the difference

between (1) ABO and (2) the market value of RPI’s plans’ assets

attributable to the Green River employees as of February 29, 1996.

Stunned by this wholly unexpected assessment, UPRG immediately

contacted RPI, OCI, and Coopers to identify the reason for the

shortfall. With the assistance of OCI and Coopers, UPRG was able

to determine that the current level of funding of the new plans

was, in the aggregate, approximately $4.6 million below the actual

funding levels of the old RPI plans, as of closing, of which UPRG’s

49 percent came to roughly $2.3 million. (Variation between

closing date levels and those over a year later were likely

attributable at least in part to the performances of the plans’

portfolios, the fluctuations in employment and benefits, and

employer contributions made during the period of more than one year

following closing.) UPRG also learned for the first time that RPI

had caused its own plans to retain, untransferred, 100 percent of

the amount by which the market value of the portion of pension plan

assets allocable to the Green River employees exceeded such share

of ABO at the time those employees were transferred to the new OCI

13
plans on February 29, 1996.

UPRG thus contends that when the dust finally settled on this

complex transaction, the wins, losses, and ties chalked up by the

three players as a result of the pension-plan maneuvering was as

follows:

! OCI essentially broke even: It received as
a credit the $3.8 million purchase-price
reduction (51 percent of the difference
between ABO and PBO), which offset its
predicted liability for 51 percent of the
anticipated funding deficiency in its own new
plans, a deficiency directly attributable to
RPI’s under-transfer of plan assets.

! RPI won: It ended up with an excess of
assets over the pre-closing funding levels in
its own pension plans, an excess directly
attributable to RPI’s causing its own plans to
retain 100 percent of the differential between
ABO and actual value of plan assets allocable
to the transferred Green River employees,16
while “paying” for that 100 percent retainage
with a credit to OCI of only 51 percent of the
ABO/PBO differential, thereby indirectly
pocketing UPRG’s 49 percent of the excess
funding for which UPRG had paid over the years
as RPI’s partner;

! UPRG lost: It ended up with a multi-
million-dollar liability for its 49 percent
share of the funding shortfall in the new OCI
plans, a liability directly attributable to
RPI’s retention in its plans of 100% of the
excess of asset value over ABO without
providing an offsetting benefit to UPRG for 49
percent of its share of that excess.

In essence, UPRG’s position is that, as a result of the acts of

16
Assuming a straight, pro-rata ratio of asset value to ABO,
Plan 1674 was funded 10.693 percent in excess of ABO before the
transfer, and Plan 1679 was funded 19.06 percent in excess of ABO.

14
RPI, it was required to pay the same 49 percent twice, first in

contributing that percentage through the Wyoming Partnership to

fund the old plans for the participating employees of Green River,

and again, a year after closing, when notified by OCI of the need

to remit 49 percent of the then-current shortfall in the new plans’

level of funding.

In 1998, UPRG filed suit against RPI, alleging breach of

contract, conversion, negligent misrepresentation, fraud, and

securities fraud in connection with UPRG’s relinquishment of its

ROFRs and RPI’s subsequent transfer of its partnership interest to

OCI. RPI moved for summary judgment on all claims. The district

court granted RPI’s motion for summary judgment and dismissed all

of UPRG’s claims with prejudice. This appeal followed.

II.

Analysis

A. Standard of Review

We review a grant of summary judgment de novo, applying the

same standard as the district court.17 A motion for summary

judgment is properly granted only if there is no genuine issue as

to any material fact.18 An issue is material if its resolution

17
Morris v. Covan World Wide Moving, Inc., 144 F.3d 377, 380
(5th Cir. 1998).
18
Fed.R.Civ.P. 56(c); Celotex Corp. v. Catrett, 477 U.S. 317,
322 (1986).

15
could affect the outcome of the action.19 In deciding whether a

fact issue has been created, we must view the facts and the

inferences to be drawn therefrom in the light most favorable to the

nonmoving party.20

The standard for summary judgment mirrors that for judgment as

a matter of law.21 Thus, the court must review all of the evidence

in the record, but make no credibility determinations or weigh any

evidence.22 In reviewing all the evidence, the court must disregard

all evidence favorable to the moving party that the jury is not

required to believe, and should give credence to the evidence

favoring the nonmoving party as well as that evidence supporting

the moving party that is uncontradicted and unimpeached.23

B. UPRG’s Fraud Claim: Voluntary Partial Disclosure

Applying this standard of review to the instant case, we

conclude that RPI was not entitled to summary judgment on UPRG’s

fraud claim. We disagree with the district court’s holding that

UPRG had not put forward sufficient evidence to support an

essential element of its claim; specifically, summary judgment

19
Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986).
20
See Olabisiomotosho v. City of Houston, 185 F.3d 521, 525
(5th Cir. 1999).
21
Celotex Corp., 477 U.S. at 323.
22
Reeves v. Sanderson Plumbing Products, Inc., 530 U.S. 133,
135, 150 (2000) (citation omitted).
23
Id. at 151 (citations omitted).

16
evidence that RPI was under a duty as a matter of law to make full

disclosure to UPRG and failed to do so. The district court appears

to have proceeded on the legal assumption that a duty of disclosure

arises only when one must correct one's own prior false or

misleading statements. But a well-established tenet of tort law

proclaims that one who voluntarily elects to make a partial

disclosure is deemed to have assumed a duty to tell the whole

truth, i.e., to make full disclosure, even though the speaker was

under no duty to make the partial disclosure in the first place.24

UPRG asserts that when RPI elected to make partial disclosures

about the stock sale over and above the disclosures that it was

required to make by the partnership agreement, RPI assumed the duty

of full disclosure. In so doing, UPRG continues, RPI obligated

itself to make known the whole truth, i.e., all material facts,

about the transaction with OCI. More particularly, UPRG contends

that RPI committed fraud by failing to disclose its scheme to

retain in its own pension trusts all partnership-funded plan assets

in excess of ABO, transferring to OCI’s new plans only enough asset

value to equal ABO, a figure millions of dollars below the funding

level for the Green River employees in the old plans.25 UPRG

24
See Restatement (Second) of Torts § 551(2)(b) (1976), with
which Texas law is in accord. See, e.g., Boggan v. Data Sys.
Network Corp., 969 F.2d 149, 154 (5th Cir. 1992) (citation
omitted); Southeastern Financial Corp. v. United Merchants & Mfrs.,
Inc., 701 F.2d 565, 566 (5th Cir. 1983) (citation omitted).
25
RPI’s protestations that it “took no money” from its pension
trust except for the amount transferred to OCI are fatuous. UPRG

17
insists that full disclosure was necessary to correct the false and

misleading impression that “there was nothing about the proposed

transaction that would harm [UPRG],”26 an impression created ——

deliberately, according to UPRG —— by RPI’s voluntary partial

disclosures that (1) after the transfer, the new pension plans

would be “substantially similar” to the old ones, (2) funds equal

to only the ABO amount of the old RPI pension plans would be

transferred to OCI for the new plans, and (3) OCI would receive a

purchase price credit that, unbeknownst to UPRG, would “cover”

OCI’s 51 percent share —— but not UPRG’s 49 percent share —— of

the resulting differential between ABO and the actual funding level

for Green River employees participating in the old plans

immediately before the transfer.

At the heart of UPRG’s fraud claim is the contention that it

was led —— or misled —— by RPI’s partial disclosure and

does not claim that RPI somehow removed assets held in trust for
pension plan participants and their beneficiaries. Rather, UPRG
argues that RPI, as plan administrator, fraudulently caused funds
previously contributed by the Wyoming partnership — only 51 percent
of which had been contributed by RPI, the remaining 49 percent
having been contributed by UPRG — to be retained in RPI’s own
pension plans. After the stock sale to OCI, RPI’s pension plans
for its own non-Green River employees who continued to be
participants in those plans were overfunded; RPI therefore
benefitted by avoiding the need to make routine contributions to
meet its pension obligations to its own employees; or, if the plans
remain overfunded, RPI can take the excess if the plans are ever
terminated.
26
According to the affidavit of Gilbert H. Kuhnhausen, Director
of Minerals with UPRG and Vice President with Big Island, RPI
offered these oral assurances to UPRG during late October and
November of 1995.

18
representation of “openness” to believe that the actual funding

level of the plans was at or below ABO, rather than several

millions of dollars above ABO as was eventually discovered —— more

than a year after closing —— to have been the case.

Here’s the significance: In its selective disclosure, RPI let

its partner, UPRG, know that (1) assets worth only the low-level

ABO funding would be transferred from RPI’s plans to OCI’s new

plans, and (2) the new plans for the transferred Green River

employees would have to be “substantially similar” to the old

plans, (mis)leading any reasonable person in UPRG’s partially

informed position to conclude that the old plans must be funded at

(or below) ABO. Then, as though that were not sufficient to lull

UPRG to such a deduction, its partner of five years standing

steadfastly contended that such was in fact the case, that the

funding level of the old plans was at or below ABO. Indeed, all

the way through oral argument before this panel and in post-

argument submittals as well, RPI insisted that such was the case

even though RPI was in the unique position among all parties at

interest to know the market value of its plans’ portfolios and the

ABO of those plans.27 That, at closing, the old plans’ assets were

27
The dissent’s claim that “there is no evidence that RPI ever
failed to furnish UPRG any information which UPRG requested
respecting the value of the assets of the plans in reference to ABO
or PBO” is wide of the mark: RPI has (misleadingly) insisted
throughout this litigation that the actual value of the plan assets
was at all relevant times at or even below ABO. We reiterate that
ABO and PBO are not actual funding figures but actuarial terms that
describe two different levels at which the liability of pension

19
worth many millions more than ABO was a well-kept and well-

protected secret which inured to RPI’s benefit and UPRG’s detriment

and which UPRG cannot reasonably be held at fault for failing to

ferret out in the thirty days it had in which to do so. We

conclude that there is sufficient summary evidence to establish a

genuine issue of material fact —— whether the plans, which were in

the exclusive control of RPI, were funded substantially above the

ABO level that RPI had disclosed to be the transfer level —— to

require the fraud claim to go to the fact finders.

We begin with basic legal principles of the Texas tort law of

fraud.28 Texas defines fraud as “a material representation, which

was false, and which was either known to be false when made or was

asserted without knowledge of its truth, which was intended to be

acted upon, which was relied upon, and which caused injury.”29

Failure to disclose a material fact is fraudulent only if the

plans can be calculated. Thus, even a plan funded at PBO will not
necessarily (or likely) have assets in that amount; accordingly,
knowledge of the plans’ ABO or PBO would not have alerted RPI to
the relevant discrepancy between ABO and the actual market value of
the plans at closing, and it is this differential —— not the
difference between ABO and PBO —— that lies at the heart of UPRG’s
claim.
28
We assume, as did the parties and the district court, that
UPRG’s common-law fraud claim arises under Texas state law.
29
Formosa Plastics Corp. v. Presidio Engineers and Contractors,
960 S.W.2d 41, 47 (Tex. 1998)(internal quotation marks and citation
omitted).

20
defendant has a duty to disclose that fact.30 A duty to speak can

arise by operation of law or by agreement of the parties.31

A duty to speak arises by operation of law when (1) a

confidential or fiduciary relationship exists between the parties;

or (2) one party learns later that his previous affirmative

statement was false or misleading; or (3) one party knows that the

other party is relying on a concealed fact, provided that the

concealing party also knows that the relying party is ignorant of

the concealed fact and does not have an equal opportunity to

discover the truth; or (4) one party voluntarily discloses some but

less than all material facts, so that he must disclose the whole

truth, i.e., all material facts, lest his partial disclosure convey

a false impression.32 Here, when we view the facts and all

reasonable inferences from the summary judgment evidence favorably

to UPRG as the non-movant, we conclude that RPI assumed the

affirmative duty to make full disclosure when it volunteered some

(but not all) material information about the transaction. It

thereby obligated itself to speak the whole truth; it could not

remain silent after merely making partial disclosures that conveyed

30
Spoljaric v. Percival Tours, Inc., 708 S.W.2d 432, 435 (Tex.
1986).
31
Trustees of the Northwest Laundry & Dry Cleaners Health &
Welfare Trust Fund v. Burzynski, 27 F.3d 153, 157 (5th Cir. 1994).
32
World Help v. Leisure Lifestyles, Inc., 977 S.W.2d 662, 670
(Tex. App.–Forth Worth 1998, pet. denied); see also Trustees, 27
F.3d at 157.

21
a false impression.33

As we have already noted, the partnership agreement obligated

RPI to disclose nothing more than (1) its intention to transfer an

interest in the partnership and the amount of the interest that it

intended to transfer —— here, 100 percent, (2) the identity of the

proposed transferee —— here, OCI, and (3) the purchase price ——

here, $150 million in cash (subject to adjustments). The

partnership agreement also obligated RPI to provide UPRG with a

“photostatic copy” of a written “purchase offer”; however, all that

the partnership agreement required the purchase offer to include

was the purchase price. RPI nevertheless opted to disclose not

33
UPRG also argues that a confidential or fiduciary
relationship existed between the parties that created a duty to
disclose. RPI counters that, as it was not a party to the
partnership agreement, any dealings between RPI and UPRG must be
characterized as arms-length. In support of its contention, RPI
notes that UPRG does not advance any claims alleging breach of
fiduciary duty. Although our holding does not depend on the
existence of a relationship of trust or confidence between RPI and
UPRG, we note that a strict fiduciary relationship is not required
for a duty to disclose to arise. See Trustees, 27 F.3d at 157
n.16. Rather, the relationship must only be one of “trust and
confidence.” See Lang v. Lee, 777 S.W.2d 158, 164
(Tex.Civ.App.–Dallas 1989, no writ).
More to the point, when examining claims of fraud, courts can
ignore technicalities such as multiple layers of business entities,
and look directly to the true parties in interest and control, such
as parent corporations owning and controlling — and directly acting
through — their wholly owned subsidiaries. Here, Holding was
obviously created as RPI’s alter ego for the Wyoming partnership
and served as such; the same could be said of Big Island and UPRG.
In particular, the summary judgment evidence shows that RPI made no
bones about being the de facto managing partner of the Wyoming
partnership even though Holding was technically the named partner.
RPI cannot here hide behind such legal niceties and claim to be a
stranger to the Wyoming partnership.

22
merely the purchase price and the identity of the purchaser, but,

more substantially, the myriad terms and conditions of a complex

“Stock Purchase Agreement” that extended over 100 pages.

Moreover, RPI went out of its way to make UPRG aware that, by

sharing the full purchase agreement with it, RPI had voluntarily

supplied information beyond its contractual obligation. In a

letter responding to UPRG’s first request for more information

about the proposed transfer, the president of RPI stated,

we are forwarding the additional documents you
requested even though the information you
received on October 23 is sufficient for
formal notification of a definitive offer.
Because we believe these documents will not
impact your analysis of Union Pacific’s first
refusal rights, we add the following points.
Section 10.4 does not require that we deliver
the Stock Purchase Agreement since it
specifies only that we have to notify you of
the interest to be transferred and the
identity of the purchaser. We have done all
of this formally on October 23 and provided
additional perspective and information in the
spirit of openness (emphasis added).

Clearly RPI elected unilaterally to go beyond the minimum formal

disclosures required by the partnership agreement; just as clearly,

when it did so, RPI assumed the affirmative duty to disclose the

whole truth, i.e., all material facts, about the proposed

transaction. A trier of fact could reasonably conclude that RPI

disclosed significantly more than the minimum required by law to

trigger this duty when it chose to reveal some, but less than all,

material facts about the proposed sale to OCI.

23
 A fact finder could also conclude that RPI intentionally

lulled UPRG into a false sense of security by embellishing its

charade with a constant extolling of the “spirit of openness” in

which, RPI intimated, it had always conducted its dealings with

UPRG. Further, a trier of fact could find that the false sense of

security that RPI engendered in UPRG by this ploy was reinforced by

the several instances when RPI’s cajolery was closely preceded or

followed by exhortations for UPRG to make a decision on the ROFRs

within the 30-day period specified in the contracts (despite RPI’s

having omitted several schedules that were supposed to be attached

to the purchase agreement).

Inasmuch as RPI was under an affirmative duty of full

disclosure to UPRG as a matter of law, a trier of fact could make

the ultimate finding that RPI committed fraud against UPRG by

making partial disclosures that conveyed a false impression;

specifically, that the funding levels of the pension plans would be

“substantially similar” to those of the old RPI plans, all the

while lulling UPRG into sufficient complacency and trust that it

would not conduct a more extensive independent search. RPI invoked

“spirit of openness” as a synonym for “full disclosure” on more

than one occasion when corresponding with UPRG. Surely, RPI cannot

now disclaim its duty to UPRG to disclose the whole truth of the

transaction, especially when RPI contemporaneously accompanied its

“sweet talk” to UPRG with reminders that UPRG had only 30 days in

which to decide whether to exercise its ROFRs, despite RPI’s

24
omission of schedules and exhibits from the copy of the purchase

agreement that it voluntarily sent to UPRG.

C. RPI’s Affirmative Defenses

1. UPRG Knew What It Did Not Know

RPI counters by asserting that even if it were under a duty of

full disclosure, it fulfilled that obligation by disclosing to

UPRG, through the purchase agreement, that (1) OCI would receive a

reduction in the purchase price equivalent to 51 percent of any

difference between the plans’ ABOs and PBOs, and (2) plan assets

worth only ABO would be transferred to OCI. Therefore, claims RPI,

the only piece of information ostensibly “hidden” from UPRG was the

precise amount of the difference between ABO or PBO. It follows,

according to RPI, that because UPRG “knew that it did not know”

this figure, RPI’s failure to disclose it, in the absence of an

explicit request to do so, cannot constitute fraud. In sum, RPI

argues that no material fact was “hidden” from UPRG because it

“knew what it didn’t know” —— the difference between the ABO and

PBO valuations of the pension plans —— and never explicitly

requested that information. We disagree.

First, this is a quintessential red herring. Neither the fact

that the purchase price would be adjusted for the difference

between ABO and PBO, nor the quantum of that differential, is what

is at issue. What is at issue, and what RPI appears to be

intentionally obfuscating by harping on the ABO/PBO purchase-price

credit, is not the differential between ABO and PBO but the

25
differential between ABO and the fair market value of the plan

assets, i.e., the value of the plan assets over and above ABO.34

That difference is the quantum that UPRG insists RPI

surreptitiously caused its pension plans not to transfer to the OCI

plans35; that is the “hidden ball” alleged by UPRG.

The undisclosed existence of a multi-million-dollar

differential between ABO and the plans’ actual funding levels is

what allegedly forced UPRG to pay twice to fund the plan; more

precisely, to pay 49 percent of that differential twice. RPI knew,

at all times from and after October 19, 1995, that (1) the value of

the plan assets always exceeded ABO, and (2) at most times the

excess of PBO over ABO would likely be greater than the excess of

actual funding over ABO. Thus the 51 percent of the ABO/PBO

differential that RPI “paid” to OCI in the form of a purchase-price

34
The dissent’s contention that “[a]ll UPRG didn’t know was the
amount by which the PBO level would exceed the ABO level” is thus
beside the point: The critical fact that UPRG did not know —— and
without information from RPI could not know —— was the difference
between the actual market value of the plan assets and ABO. It is
this difference which is the gravamen of UPRG’s fraud claim, and
that makes the PBO/ABO discussion a “red herring.” See supra n.27.
35
At oral argument before this court, counsel for RPI insisted
that the ABO/PBO purchase price adjustment was “just part of the
negotiation.” His representation is undercut by the fact that
prior to the drafting of the purchase agreement, Coopers advised
RPI that if it should ever assume an obligation to transfer plan
assets at the higher PBO level, the resulting underfunding in its
own pension fund accounts “could be reflected as an adjustment to
the purchase price.” More importantly, this information tends to
show that RPI, despite all of its diversionary invocations of ABO
and PBO, was and is quite aware that actual funding of the plans,
rather than either ABO or PBO, is the critical figure vis-à-vis
UPRG’s fraud claim.

26
reduction assured OCI of an adequate financial offset to the

pension plan liability that it was certain to incur at closing as

accurately forecast by the actuaries; specifically, liability for

51 percent of the shortfall between the value of assets transferred

at ABO and the actual level of plan funding.

The thrust of UPRG’s fraud claim is that RPI made no

concomitant provision for such an offsetting credit to cover UPRG’s

49 percent share of the employer’s liability for the new plans’

funding shortfall. Even more central to that fraud claim is the

allegation that, given RPI’s exclusive sponsorship and

administration of the old plans, there was no hint that anything

was amiss and no way for UPRG to discover what RPI was doing, at

least not within the 30-day ROFR window, short of rejecting RPI’s

“spirit of openness” and instituting litigation to compel

disclosure through protracted discovery. Significantly, RPI (and

OCI, for that matter) had to have been aware that, after closing,

an additional infusion of assets would be needed to bring the new

plans up to the pre-closing funding level of the old ones: That

would be required to ensure that the Department of Labor, the IRS,

and the PBGC would be satisfied that the Green River employees had

been transferred to pension plans that were essentially identical

to the old ones —— in level of funding as well as in substantive

provisions.

RPI’s argument also fails to confront the fact that even if

UPRG had asked RPI for the exact amount of the ABO/PBO differential

27
in the context of the purchase price adjustment, disclosure of that

figure still would not have alerted UPRG to either the existence or

extent of the amount by which the actual value of the pension plan

assets exceeded ABO, the latter being the value of assets actually

transferred to OCI. Not only is there no correlation between these

two differentials, but it is also undisputed that only RPI, as the

sole sponsor of the plans, had control over and access to plan

records and reports that would have made the gap between actual

funding and ABO apparent or even determinable.36 As UPRG correctly

notes, it is hornbook law that a fraud feasor cannot defeat a claim

for damages by complaining that the defrauded party might have

discovered the truth by exercising more diligence.37

RPI’s heavy reliance on Koch Indus., Inc. v. Sun Co.,38 which

involved breach of contract and not fraud, is thus misplaced. In

Koch, we held that specific performance is not available as a

remedy for breach of an ROFR agreement when the seller has made

reasonable disclosure of an offer’s terms, but the rightholder has

36
Given the short, 30-day fuse on the ROFRs and the minimal
disclosure requirements of the partnership agreement, it is
questionable whether UPRG would have been furnished that
information even if it had requested it. RPI’s apparent culture of
deceptively selective disclosure seems to have infected its
appellate counsel who, in oral argument and in post-argument
submittals, insisted that the plans’ funding levels were barely at
or even below ABO at all significant times, a representation
eventually shown to have been false.
37
Southeastern Finanacial Corp., 701 F.2d at 567.
38
918 F.2d 1203 (5th Cir. 1990).

28
failed to undertake a reasonable investigation of any terms unclear

to him.39 The claim in Koch, that particular terms in the offer

were unclear, thus sounded in contract; the claim in this case,

that UPRG fraudulently concealed key information about the proposed

transaction, sounds in tort. RPI’s contention that UPRG was

somehow obligated to ferret out, in 30 days, each and every

potentially damaging fact somehow alluded to in a 113-page document

that RPI had voluntarily provided to UPRG “in the spirit of

openness” thus rings hollow, to say the least.40

Finding no support for its novel “due diligence” defense in

binding precedent, RPI reaches out to the Tenth Circuit case of

Jensen v. Kimble for the proposition that when a fraud plaintiff

“knew what he didn’t know,” such knowledge belies a claim of

fraud.41 But, when the plaintiff stock seller in Jensen

specifically asked the defendant purchaser to reveal the names of

the other parties involved in the transaction, the purchaser

informed the plaintiff precisely what information he was refusing

39
Id. at 1212.
40
We also note that RPI and UPRG were, at a bare minimum,
“coadventurers, subject to fiduciary duties akin to those of
partners.” See Meinhard v. Salmon, 164 N.E. 545, 546 (N.Y. 1928).
When a relationship of this kind exists, the law does not exact
diligence on the part of the defrauded party as prompt and as
searching as that required into the conduct of a stranger. See
Lang, 777 S.W.2d at 163-64.
41
1 F.3d 1073, 1078 (10th Cir. 1993).

29
to disclose.42 Jensen is thus far different from the case at bar,

in which UPRG alleges — and supports with summary judgment evidence

—— that RPI selectively volunteered information only in part, all

the while deceptively proclaiming “openness” so as to create the

illusion of full disclosure.

Finally, in Jensen, the defendant’s omissions were not

misleading with regard to any of the purchaser’s other statements.43

That is not so in the instant case, in which UPRG alleges that

RPI’s material omissions aided and abetted its partial disclosures

by misleading UPRG about crucial aspects of the transaction. If

UPRG can prove this by a preponderance of the evidence, it will

have established the existence of a stereotypical primrose path

down which it was cunningly led by RPI’s (1) partial disclosures,

(2) repeated reminders about the 30-day ROFR deadline, (3) omission

of key schedules and exhibits from the purchase agreement, and (4)

assurances of “openness.”

2. The Tort/Contract Distinction

As an alternative defense, RPI argues that UPRG’s fraud claim

is barred by the general rule that tort actions are not cognizable

in a suit on a contract.44 RPI maintains that its conduct would not

give rise to liability unless such acts breached its contractual

42
Id.
43
Id.
44
See Southwestern Bell Telephone Co. v. DeLanney, 809 S.W.2d
493, 494 (Tex. 1991).

30
agreements with UPRG.

The flaw in RPI’s argument is that it is based on the false

premise that its obligation to disclose information about the

prospective sale to OCI arose solely from the ROFR requirements of

the agreements. To the contrary, as we have already noted, RPI’s

duty of full disclosure arose from its voluntary partial

disclosures of information beyond that required by the partnership

agreement, not to mention RPI’s incomplete responses to the

inquiries that UPRG did make and RPI’s own insistence that it was

making the additional disclosures in the “spirit of openness.” In

conducting itself that way, RPI assumed an obligation to make known

to UPRG whatever further information might be necessary to correct

any false impressions conveyed by RPI’s partial disclosures.

Whether RPI met its contractual obligations under the ROFR

agreements is irrelevant because its duty to disclose arose

independently of those agreements. As even RPI must acknowledge,

the law makes clear that if particular conduct would give rise to

liability independent of the fact that a contract exists between

the parties, the plaintiff’s claim sounds in tort.45

It is true that RPI’s initial obligation to make contractually

45
See DeLanney, 809 S.W.2d at 494. Whether the Texas Supreme
Court’s subsequent holding in Formosa that a fraudulent inducement
claim can be brought in addition to a breach of contract claim
applies to all fraud claims or only to claims of fraudulent
inducement is thus irrelevant, because DeLanney’s stricter
“independent duty” requirement is satisfied here in either case.
See Formosa, 960 S.W.2d at 47.

31
specified disclosures —— its intention to transfer its partnership

interest, the identity of the purchaser, and the purchase price ——

did come from the agreement between the parties, so that any claim

arising out of alleged violations of the ROFRs would indeed sound

in contract. But when, as here, the fraud claim is based on an

obligation that arose, by operation of law, out of one party’s

voluntary disclosure of information beyond that required by its

agreements with the other party, the general rule against fraud

claims based on contractual obligations is no barrier to liability.

III.

Conclusion

We affirm the district court’s summary judgment to the extent

it dismissed UPRG’s claims grounded in conversion, securities

fraud, and negligent misrepresentation. For the reasons discussed

above, however, we reverse the district court to the extent that it

dismissed UPRG’s fraud claim, and we remand that claim for further

proceedings consistent with this opinion. In so doing we stress

that we imply nothing about the likely outcome of a trial on this

claim, only that the theory of UPRG’s fraud claim and the summary

judgment evidence proffered in support of that theory, viewed in

the light most favorable to UPRG as the non-movant, are sufficient

to defeat summary judgment and require a determination by the trier

of fact.

AFFIRMED in part; REVERSED in part; and REMANDED.

32
33
GARWOOD, Circuit Judge, dissenting.

I concur in the majority opinion to the extent that it affirms

the dismissal of UPRG’s conversion, securities fraud and negligent

misrepresentation claims. However, I respectfully dissent from the

reversal and remand of the judgment dismissing UPRG’s fraud claim.

The majority, correctly, does not contend that there is any

evidence that RPI made any false or misleading factual

representation to UPRG or that RPI failed to disclose to UPRG any

fact necessary in order to make the facts which it did represent to

UPRG not misleading. Rather, the majority asserts that there is

evidence that RPI committed fraud on UPRG by breaching a duty to

make full disclosure to UPRG. I disagree.

It has always been unclear to me just what UPRG claims should

have been but was not disclosed. UPRG’s ultimate complaint is

that, about a year after RPI’s sale to OCI, UPRG was called upon by

OCI to come up with UPRG’s forty-nine percent share of the then

some 4.6 million dollar difference between the actual funding of

the new pension plans for partnership employees and the appropriate

PBO level for those plans. This wholly resulted from the

facts–disclosed by RPI and known to UPRG–that RPI was transferring

from its pension plans to the OCI plans assets sufficient only to

meet ABO levels (not the higher PBO levels) for the transferring
employees, that under the RPI plans’ “contributions are intended to

provide for benefits attributable for service rendered to date, as

well as for those expected to be earned in the future” (i.e., PBO),

that OCI was going to continue the business (which would render PBO

funding appropriate), agreed to have the transferring employees

covered by “substantially similar” plans as they were before, and

assumed liability for those employees’ accrued benefits as they

thereafter became due, and that OCI would receive a credit on its

purchase price obligation equal to fifty-one percent (RPI’s

partnership percentage) of the difference between the ABO and PBO

funding levels for the plans for the transferred employees. RPI

disclosed and UPRG knew all these facts. Moreover, UPRG plainly

knew, and certainly RPI could reasonably assume UPRG knew, that the

PBO level would exceed the ABO level. All UPRG didn’t know was the

amount by which the PBO level would exceed the ABO level, and

clearly UPRG knew it didn’t know this. Nothing RPI said can be

construed either as any sort of implied representation as to the

extent of this PBO/ABO difference or as any sort of implied

characterization of the extent of that difference as being minor or

the like.

The majority labels all this as a “quintessential red

herring.” This characterization is hard to understand, as the

extent of the PBO/ABO difference is the entire reason for and the

measure of the “loss” of which UPRG complains. The majority claims

35
that “what is at issue” is that at the time of the RPI-OCI

agreement, and when RPI made disclosures to UPRG respecting it, the

value of the assets of the plans was substantially above ABO, and

that this was not disclosed to UPRG. Nothing, however, is pointed

to as constituting any even implied representation by RPI to UPRG

that either plan was then not funded (or did not have assets with

a then market value) materially above ABO. Nor is there any

evidence that UPRG believed, or construed anything RPI had said as

even impliedly representing, that the plans were funded (or that

their assets had a then market value) only at or below, or not

materially above, ABO. To the contrary, the partnership audited

financial statements, furnished annually to UPRG state respecting

these plans that “contributions are intended to provide benefits

attributable to services rendered to date, as well as for those

expected to be earned in the future.” At oral argument, counsel

for UPRG read this language and characterized it as stating “That’s

PBO” and as constituting evidence that the plans were then funded

“above ABO.”

I also note that there is no evidence that RPI ever failed to

furnish UPRG any information which UPRG requested respecting the

value of the assets of the plans in reference to ABO or PBO, or the

extent of the difference between ABO and PBO funding of the plans.

Apropos here are the remarks of the Tenth Circuit in Jensen v.

Kimble, 1 F.3d 1073 (10th Cir. 1993), upholding summary judgment

36
for the defendant Kimble in a securities fraud case, viz:

“. . . [B]y virtue of the disclosures that Kimble did
make, Jensen knew what he didn’t know. Under these
circumstances, even assuming arguendo that a special
relationship of trust existed between Jensen and Kimble,
we do not believe it can be said that Kimble’s omissions
misled Jensen with respect to any of Kimble’s other
remarks. Accordingly, even viewing the evidence in the
light most favorable to the plaintiffs, we conclude that
Kimble’s omissions were neither manipulative nor
deceptive within the meaning of Rule 10b-5 and thus are
not actionable under this rule.” Id. at 1078.

If the concept of “fraud” is so protean and malleable as to

allow UPRG recovery on this record, then the value of written

contracts between even the most sophisticated businesses is

substantially undercut and we have given an unfortunate boost to

the rule of ad hocracy and a concomitant disservice to the rule of

law.

37