NCMIC Insurance Company v. Allied Professionals Insurance Company, a Risk Retention Group, Inc.

United States Court of Appeals
For the Eighth Circuit
___________________________
No. 23-3261
___________________________
NCMIC Insurance Company
Plaintiff - Appellee
v.
Allied Professionals Insurance Company, a Risk Retention Group, Inc.
Defendant - Appellant
Charlotte Erdmann; Kristin M. Schantzen; Jay J. Schantzen; Valley Chiropractic
Clinic, Ltd.
Defendants
____________
Appeal from United States District Court
for the District of Minnesota
____________
Submitted: May 8, 2024
Filed: August 5, 2024
____________
Before COLLOTON, Chief Judge, SHEPHERD and STRAS, Circuit Judges.
____________
SHEPHERD, Circuit Judge.
In this insurance coverage dispute, Allied Professionals Insurance Company
(APIC) seeks to compel arbitration of NCMIC Insurance Company’s (NCMIC)

-- 1 of 10 --

-2-
declaratory judgment action. Proceeding under a theory of direct-benefits estoppel,
APIC argued that NCMIC could be compelled to arbitrate because of a clause in
APIC’s insured’s policy. Concluding that Minnesota law embraced only a limited
version of direct-benefits estoppel under which APIC could not prevail, the district
court1 dismissed APIC’s motion to compel arbitration. Having jurisdiction under 9
U.S.C. § 16(a), we affirm.
I.
Charlotte Erdmann is a massage therapist who was sued by a patient, Kristin
Schantzen, and her husband, Jay, in Minnesota state court for injuries allegedly
sustained by Kristin during a massage-therapy session. Erdmann was individually
insured by APIC. Her employer, Valley Chiropractic Clinic (Valley), was separately
insured by NCMIC. APIC and Erdmann tendered the patient’s claims to NCMIC,
but NCMIC declined to defend or indemnify Erdmann. Instead, NCMIC filed a
declaratory judgment action against APIC, the Schantzens, and Erdmann. NCMIC
sought, first, a declaration that it was not obligated to defend or indemnify Erdmann
and, second, a declaration that if NCMIC’s policy were found to cover Erdmann, the
coverage was in excess of the primary coverage already provided by APIC’s policy.
While the declaratory judgment action was pending, the Schantzens settled with
Erdmann and Valley. Erdmann agreed to pay $1.6 million, while Valley agreed to
pay $250,000. It is undisputed that NCMIC must pay the $250,000 portion of the
settlement. What remains in dispute is whether NCMIC or APIC must pay
Erdmann’s $1.6 million.
APIC moved for the district court to compel arbitration of that dispute,
seeking enforcement of the following provision contained in its policy covering
Erdmann:
1 The Honorable Patrick J. Schiltz, Chief Judge, United States District Court
for the District of Minnesota.

-- 2 of 10 --

-3-
All disputes or claims involving [APIC] shall be resolved by binding
arbitration, whether such dispute or claim arises between the parties to
this Policy, or between [APIC] and any person or entity who is not a
party to the Policy but is claiming rights either under the Policy or
against [APIC].
APIC argued that, although NCMIC was not a party to the APIC-Erdmann policy, it
could enforce the arbitration clause contained therein under a theory of
direct-benefits estoppel, which typically “applies when a nonsignatory knowingly
exploits the agreement containing the arbitration clause.” Reid v. Doe Run Res.
Corp., 701 F.3d 840, 846 (8th Cir. 2012) (citation omitted).
The district court denied the motion. It reasoned that Minnesota law, while
never having expressly addressed the doctrine, would likely embrace a version of
direct-benefits estoppel previously articulated by this Court in Reid: “Nonsignatories
can be bound to an arbitration agreement . . . ‘(1) by knowingly seeking and
obtaining “direct benefits” from that contract; or (2) by seeking to enforce the terms
of that contract or asserting claims that must be determined by reference to that
contract.’” Id. (citation omitted).
Looking to the first prong under Reid, the district court observed that NCMIC
never claimed to be a third-party beneficiary to the APIC-Erdmann policy, that the
policy itself said that there are no intended third-party beneficiaries, and that NCMIC
never argued that it was entitled to rights or benefits under the APIC-Erdmann
policy. It therefore concluded that NCMIC did not seek to obtain a direct benefit
under the policy.
Turning to the second prong, the district court noted that, since NCMIC sought
a declaration that APIC’s coverage is primary and that NCMIC’s is excess, “it is
literally true that the Court cannot determine the priority of coverage without
referring to the terms of the APIC policy.” Still, after explaining the concerning
policy implications—NCMIC would be forced to arbitrate a claim against a
defendant based on a clause in a contract it never signed—the district court held that

-- 3 of 10 --

-4-
direct-benefits estoppel did not apply. The district court considered it significant
that neither the Eighth Circuit nor the Minnesota Supreme Court had ever applied
direct-benefits estoppel to a fact pattern such as this, wherein a defendant-signatory
sought to enforce an arbitration clause in a contract against a plaintiff-nonsignatory.
APIC now appeals the denial of its motion.
II.
Direct-benefits estoppel is a kind of equitable estoppel. See Int’l Paper Co. v.
Schwabedissen Maschinen & Anlagen GMBH, 206 F.3d 411, 417-18 (4th Cir.
2000). The “application of equitable estoppel presents at least mixed questions of
law and fact,” which we “review[] de novo.” Donaldson Co. v. Burroughs Diesel,
Inc., 581 F.3d 726, 731 (8th Cir. 2009). The use of equitable estoppel to compel
arbitration is a matter of state contract law. Id. at 732 (citing Arthur Andersen LLP
v. Carlisle, 556 U.S. 624, 631 (2009)). Therefore, the threshold issue for us upon de
novo review is whether Minnesota law embraces a doctrine of estoppel that allows
a defendant-signatory such as APIC to compel an unwilling plaintiff-nonsignatory
such as NCMIC to arbitrate its claims.
Minnesota courts have never expressly adopted the doctrine of direct-benefits
estoppel. Accordingly, we must predict how (or indeed if) the Minnesota Supreme
Court would apply the doctrine. Progressive N. Ins. Co. v. McDonough, 608 F.3d
388, 390 (8th Cir. 2010) (“This court is bound by decisions of the highest state court
when interpreting state law. If the highest state court has not decided an issue we
must attempt to predict how the highest court would resolve the issue, with decisions
of intermediate state courts being persuasive authority.” (citation omitted)).
The closest the Minnesota Supreme Court has come in addressing the doctrine
is in Onvoy, Inc. v. SHAL, LLC, 669 N.W.2d 344 (Minn. 2003). There, the court
explained that, “[g]enerally, arbitration clauses are contractual and cannot be
enforced by persons who are not parties to the contract.” Id. at 356. It noted,
however, that “[t]here are exceptions to the rule” and that “[f]ederal cases have set

-- 4 of 10 --

-5-
out at least three principles on which a nonsignatory to a contract can compel
arbitration” including “equitable estoppel.” Id. That doctrine, the court explained,
“prevents a signatory from relying on the underlying contract to make his or her
claim against the nonsignatory.” Id. But the court stopped short of applying
equitable estoppel and remanded to the district court, instructing it to consider the
doctrine if it first found that SHAL could compel arbitration and other individual
defendants were not agents of SHAL. Id. at 357. Significantly, however, the court
cited two authorities for the foregoing exposition: MS Dealer Service Corp. v.
Franklin, 177 F.3d 942 (11th Cir. 1999), abrogated on other grounds by Carlisle, 556
U.S. at 631, and Gabriel M. Wilner, Domke on Commercial Arbitration § 10 (2d ed.
1999).2
In MS Dealer, the Eleventh Circuit explained the circumstances in which
“equitable estoppel allows a nonsignatory to compel arbitration.” 177 F.3d at 947
(citation omitted). It did not address the fact pattern before us, however, wherein a
signatory attempts to compel a nonsignatory to arbitrate its claims. But the Wilner
treatise did, explaining that “even though arbitration is contractual by nature, a
nonsignatory to an arbitration agreement may nonetheless be bound by the
agreement under an accepted theory of agency or contract law.” Wilner, supra,
§ 10.00. For that proposition, the Wilner treatise cited Thomson-CSF, S.A. v.
American Arbitration Ass’n, 64 F.3d 773 (2d Cir. 1995). There, the Second Circuit
explained that the court had previously “bound nonsignatories to arbitration
agreements under an estoppel theory” when the nonsignatory “directly benefitted
from the” agreement containing the arbitration clause—i.e., direct-benefits estoppel.
Id. at 778-79.
2 We note that, strictly speaking, the Minnesota Supreme Court cited “Gabriel
M. Wilner, Domke on Commercial Arbitration . . . (1983).” Because the second
edition was current at the time of Onvoy, and because the pagination of that edition
aligns with the Minnesota Supreme Court’s citations, we refer to the second edition
throughout our opinion.

-- 5 of 10 --

-6-
The Second Circuit also identified “an alternative estoppel theory”—virtually
identical to Reid’s second prong—whereby
a signatory was bound to arbitrate with a nonsignatory at the
nonsignatory’s insistence because of “the close relationship between
the entities involved, as well as the relationship of the alleged wrongs
to the nonsignatory’s obligations and duties in the contract . . . and [the
fact that] the claims were ‘intimately founded in and intertwined with
the underlying contract obligations.’”
Id. at 779 (alteration in original) (citations omitted). However, the Second Circuit
explained that this alternative theory of estoppel was only applicable in cases where
a signatory was being estopped from avoiding arbitration. Id. In other words,
according to Thomson-CSF, a nonsignatory can only be compelled to arbitrate by a
signatory under an estoppel theory based on the nonsignatory directly benefitting
from the agreement containing the arbitration clause, and not the “alternative
estoppel theory” premised on the claims being “intimately founded in and
intertwined with the underlying contract obligations.” Id. (citation omitted).
We therefore predict that, when faced with a defendant-signatory’s attempt to
compel a plaintiff-nonsignatory to arbitrate its claims, the Minnesota Supreme Court
would embrace, at most, the version of direct-benefits estoppel articulated in
Thomson-CSF: APIC can compel NCMIC to arbitrate only if NCMIC “directly
benefitted” from the APIC-Erdmann policy. 64 F.3d at 779. This is, in effect, to
ask a similar question as posed in the first prong of Reid, namely, whether the
nonsignatory “knowingly s[ought] and obtain[ed] ‘direct benefits’ from” the
contract. Reid, 701 F.3d at 846 (citation omitted). We therefore conclude that the
Minnesota Supreme Court likely would not embrace the version of direct benefits
estoppel announced in the second prong of Reid, which binds nonsignatories when
they “seek[] to enforce the terms of th[e] contract or assert[] claims that must be
determined by reference to th[e] contract.” Id. (citation omitted).

-- 6 of 10 --

-7-
This prediction is bolstered by nonbinding but persuasive Minnesota
authorities. First, we note that two concurring Justices of the Minnesota Supreme
Court expressed skepticism in Onvoy when it came to the enforcement of arbitration
clauses against unwilling nonsignatories. See Onvoy, 669 N.W.2d at 358-59
(Anderson, J., concurring) (“[S]ound[ing] a note of caution to courts that when a
right as fundamental as the right of access to the courts and trial by jury is at stake,
waiver of that right is not to be lightly presumed.”); id. at 361-62 (Gilbert, J.,
concurring in part and dissenting in part) (“To enforce the arbitration clause in our
case, the majority cites several federal circuit court decisions relating to equitable
estoppel . . . . In doing so, the majority constructs a line of reasoning that effectively
nullifies the constitutional rights of citizens by restricting their access to our courts
to adjudicate claims.”).
Second, the Minnesota Court of Appeals, considering a near-identical
fact pattern to that underlying this appeal, and applying Onvoy and the Wilner
treatise, rejected the application of direct-benefits estoppel. St. Paul Fire & Marine
Ins. Co. v. API, Inc., No. A04-204, 2004 WL 2161181, at *4 (Minn. Ct. App. Sept.
28, 2004) (holding that the trial court did not err in denying a signatory-insurer’s
motion to compel arbitration of an insured’s claims because the insured was not a
signatory to the contract containing the arbitration clause). Although St. Paul Fire
might suggest that the Minnesota Supreme Court would not adopt any version of
direct-benefits estoppel on these facts, see Progressive, 608 F.3d at 390 (decisions
of intermediate state appellate courts are persuasive authority absent a decision of
the state’s highest court), it was designated as unpublished and is therefore not
precedential, Dynamic Air, Inc. v. Bloch, 502 N.W.2d 796, 800 (Minn. Ct. App.
1993) (“Unpublished opinions of the Court of Appeals are not precedential. At best,
these opinions can be of persuasive value.”). We therefore consider this decision as
bolstering our limited conclusion, rather than dictating a stronger one.
Because we predict that the Minnesota Supreme Court would embrace, at
most, the version of direct-benefits estoppel articulated in Thomson-CSF, this case
turns on whether NCMIC directly benefitted from the APIC-Erdmann policy. Recall

-- 7 of 10 --

-8-
that the district court held that NCMIC did not seek a direct benefit: NCMIC never
claimed to be a third-party beneficiary to the APIC-Erdmann policy; the policy
stated that there are no intended third-party beneficiaries; and NCMIC had not
argued that it was entitled to rights or benefits under the policy.
APIC does not dispute these conclusions on appeal. Rather, it argues that,
“[w]hile NCMIC is not seeking a direct payment from APIC, it is nonetheless
seeking to directly benefit from the application of” the APIC-Erdmann
policy—without which “NCMIC would be responsible to indemnify Erdmann.” But
this argument contorts the meaning of a “direct benefit” and APIC concedes as much
when it states that this benefit is “not ‘direct’ in the traditional sense.” Moreover,
we have rejected this argument in similar circumstances. In Nitro Distributing, Inc.
v. Alticor, Inc., we analyzed whether plaintiff-nonsignatory businesses could be
bound to arbitrate their claims by the defendant-signatory multinational company
under a theory of estoppel. 453 F.3d 995, 998 (8th Cir. 2006). We rejected the
defendant-signatories’ contention that “benefitting in general” from a distribution
agreement is the same as “directly benefitting” from it. Id. APIC asks us to draw
the same equivalence here, and we decline to do so. For the foregoing reasons, and
because APIC offers us no other reason to disturb the district court’s conclusions on
this point, we conclude that NCMIC did not directly benefit from the APIC-Erdmann
policy. Accordingly, we conclude that, under Minnesota law, APIC cannot compel
NCMIC to arbitrate its claims under a theory of direct-benefits estoppel.
In so holding, we need not reach the district court’s analysis and APIC’s
arguments concerning the second Reid prong. As explained, we predict that the
Minnesota Supreme Court would not embrace that version of direct-benefits
estoppel. Granted, we have previously stated that “Minnesota appears to follow
federal law regarding equitable estoppel,” but we did so when faced with the reverse
fact pattern, wherein a defendant-nonsignatory sought to compel arbitration of a
plaintiff-signatory’s claims. In re Wholesale Grocery Prods. Antitrust Litig., 707
F.3d 917, 922-24 (8th Cir. 2013) (explaining that plaintiff-signatory retailers sought
to bring claims against defendant-nonsignatory wholesaler and determining whether

-- 8 of 10 --

-9-
“equitable estoppel permits the non-signatory Wholesaler to compel arbitration”).
Faced with a novel fact pattern and correspondingly novel legal issue, we are not
bound by this language from Wholesale Grocery. See Sanzone v. Mercy Health,
954 F.3d 1031, 1039-40 (8th Cir. 2020) (explaining that “[t]ypically, stare decisis
requires this court to follow the opinions of prior panels” but that “when an issue is
not squarely addressed in prior case law, we are not bound by precedent through
stare decisis” (citation omitted)).
III.
For the foregoing reasons, we affirm the judgment of the district court.
STRAS, Circuit Judge, concurring in the judgment.
I agree that the Minnesota Supreme Court would not force parties into
arbitration just because their claims “refer[]” to contracts they never signed. Ante at
6 (quoting Reid v. Doe Run Res. Corp., 701 F.3d 840, 846 (8th Cir. 2012)). But it is
not clear to me that direct-benefits estoppel is even a legitimate equitable doctrine,
much less that any court in Minnesota would adopt it.
For one thing, the Minnesota Supreme Court has already expressed its
skepticism. Observing that it comes from “[f]ederal cases,” not “Minnesota courts,”
and that the trial court may not need to address it on remand, was hardly a ringing
endorsement. Onvoy, Inc. v. SHAL, LLC, 669 N.W.2d 344, 356–57, 356 n.12 (Minn.
2003). For another, this sort of “estoppel” has nothing to do with the traditional
equitable doctrine that Minnesota courts apply. Compare, e.g., Bridas S.A.P.I.C. v.
Gov’t of Turkm., 345 F.3d 347, 360–61 (5th Cir. 2003) (justifying so-called
“equitable estoppel” based on what “makes sense” and policy concerns about parties
trying “to have it both ways” (citation omitted)), with N. Petrochemical Co. v. U.S.
Fire Ins. Co., 277 N.W.2d 408, 410 (Minn. 1979) (listing the classic elements:
“representations or inducements,” “reasonabl[e] reli[ance],” and “harm[] if the claim
of estoppel is not allowed”). Equity has a rich history, including in Minnesota, and

-- 9 of 10 --

-10-
it is telling that direct-benefits estoppel is a recent federal invention that bears little
resemblance to longstanding equitable doctrines. My best (educated) Erie guess is
that Minnesota courts will want nothing to do with it.
______________________________

-- 10 of 10 --