Estate of Robert Cornell, Jr. v. Bayview Loan Servicing

RECOMMENDED FOR FULL-TEXT PUBLICATION
Pursuant to Sixth Circuit I.O.P. 32.1(b)
File Name: 18a0251p.06

UNITED STATES COURT OF APPEALS
FOR THE SIXTH CIRCUIT

ESTATE OF ROBERT CORNELL, JR.; AUDREY D. ┐
BANTOM, as Personal Representative of the Estate of │
Robert L. Cornell, Jr.; ANTHONY CORNELL, │
Plaintiffs-Appellants, │
│
> No. 18-1245
v. │
│
│
BAYVIEW LOAN SERVICING, LLC; THIEN HOANG │
TRAN, │
Defendants-Appellees. │
┘

Appeal from the United States District Court
for the Eastern District of Michigan at Detroit.
No. 2:17-cv-12121—Gershwin A. Drain, District Judge.

Decided and Filed: November 13, 2018

Before: SUHRHEINRICH, MOORE, and BUSH, Circuit Judges.

_________________

COUNSEL

ON BRIEF: Vanessa G. Fluker, VANESSA G. FLUKER, ESQ., PLLC, Detroit, Michigan, for
Appellants. Deborah S. Lapin, Martin S. Frenkel, MADDIN, HAUSER, ROTH & HELLER,
P.C., Southfield, Michigan, for Appellee Bayview Loan Servicing. Joseph J. Bernardi,
BERNARDI, RONAYNE & GLUSAC, P.C., Plymouth, Michigan, for Appellee Thien Hoang
Tran.

SUHRHEINRICH, J., delivered the opinion of the court in which BUSH, J., joined.
MOORE, J. (pp. 12–28), delivered a separate dissenting opinion.
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 2

_________________

OPINION
_________________

SUHRHEINRICH, Circuit Judge. This appeal concerns a non-judicial foreclosure under
Michigan law. After reviewing the pleadings, we conclude that the district court lacked subject
matter jurisdiction to hear the case. Thus, we VACATE the judgment of the district court with
instructions to REMAND to Michigan state court.

I. FACTS

Robert Cornell, Jr. (“Robert”) died on July 29, 2015, owing an outstanding mortgage
amount of $113,358.12 on his home at 8615 Wisconsin Street in Detroit, Michigan. At the time
of Robert’s death, the monthly mortgage payments on the Wisconsin Street home were up to
date. Yet in the five months following his death, the mortgage went unpaid. Defendant Bayview
Loan Servicing, LLC (“Bayview”), the mortgage holder, sent a delinquency notice to the home
on December 16, 2015, showing an unpaid balance of $5,813.95. On November 3, 2016,
Bayview foreclosed on the mortgage and purchased the home by sheriff’s deed at public auction.
Bayview later sold the home to Defendant Thien Hoang Tran (“Tran”).

II. PROCEDURAL HISTORY

On May 25, 2017, Plaintiffs-Appellants Estate of Robert L. Cornell, Jr. (“Estate”), by and
through Personal Representative Audrey D. Bantom and Anthony Cornell (collectively,
“Plaintiffs”) filed a complaint in Michigan state court alleging four causes of action against
Bayview, including most notably a lack of standing to foreclose under the Garn-St. Germain
Depository Institutions Act of 1982, codified at 12 U.S.C. § 1701j-3 (“Garn-St. Germain Act” or
“Act”) and MICH. COMP. LAWS § 445.1626. Bayview timely removed to federal court on the
basis of federal question jurisdiction under 28 U.S.C. § 1331, citing the Garn-St. Germain Act.
Plaintiffs did not object to removal or seek remand. Instead, Plaintiffs filed an amended
complaint asserting an additional claim of quiet title against Tran (Count V). Defendants moved
for judgment on the pleadings in part on the argument that the Garn-St. Germain Act does not
authorize a private right of action. The district court agreed, ruling that the Garn-St. Germain
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 3

Act does not authorize a private right of action, the Garn-St. Germain Act did not apply to
Plaintiffs’ claims, or both. The district court granted Defendants’ motion on all counts and
entered a judgment in their favor. The district court denied Plaintiffs’ motion for
reconsideration, and Plaintiffs filed this timely appeal.

III. ANALYSIS

“Federal courts are courts of limited jurisdiction.” Kokkonen v. Guardian Life Ins. Co. of
Am., 511 U.S. 375, 377 (1994). The district courts “have original jurisdiction of all civil actions
arising under the Constitution, laws, or treaties of the United States.” 28 U.S.C. § 1331.
A defendant may remove a case only if the claim could have been brought in federal court.
28 U.S.C. § 1441(a). Removal jurisdiction is determined from the “well-pleaded complaint.”
Merrell Dow Pharm. Inc. v. Thompson, 478 U.S. 804, 808 (1986).

Although no one has specifically addressed subject matter jurisdiction to this point, we
have an independent obligation to consider it and may do so sua sponte. Answers in Genesis of
Kentucky, Inc. v. Creation Ministries Int’l, Ltd., 556 F.3d 459, 465 (6th Cir. 2009); see also
United States v. Cotton, 535 U.S. 625, 630 (2002) (Subject matter jurisdiction “can never be
forfeited or waived.”). We must correct any defect in subject matter jurisdiction regardless of
whether the district court considered it, Cotton, 535 U.S. at 630, even if “many months of work
on the part of the attorneys and the court may be wasted,” Henderson ex rel. Henderson v.
Shinseki, 562 U.S. 428, 435 (2011); see also Hampton v. R.J. Corman R.R. Switching Co.,
683 F.3d 708, 711–12 (6th Cir. 2012) (vacating district court’s grant of summary judgment after
determining that Federal Railroad Safety Act did not create a private cause of action).

The face of the complaint references a federal statute, the Garn-St. Germain Act,
12 U.S.C. § 1701j-3, which was the sole basis for federal question jurisdiction removal from
state court. Before Congress passed the Garn-St. Germain Act, many states had laws restricting
the enforcement of due-on-sale clauses.1 Dupuis v. Yorkville Fed. Sav. & Loan Ass’n, 589 F.

1A due-on-sale clause is a “contract provision which authorizes a lender, at its option, to declare due and
payable sums secured by the lender’s security instrument if all or any part of the property, or an interest therein,
securing the real property loan is sold or transferred without the lender's prior written consent.” 12 U.S.C. § 1701j-
3(a)(1).
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 4

Supp. 820, 822 (S.D.N.Y. 1984). The Garn-St. Germain Act prohibits states from banning due-
on-sale clauses, providing in principal part that “[n]otwithstanding any provision of the
constitution or laws (including judicial decisions) of any State to the contrary, a lender may,
subject to subsection (c) of this section, enter into or enforce a contract containing a due-on-sale
clause with respect to a real property loan.” 12 U.S.C. § 1701j-3(b)(1). That means a due-on-
sale clause is presumptively valid unless it qualifies as one of nine exceptions listed in § 1701j-
3(d):

With respect to a real property loan secured by a lien on residential real property
containing less than five dwelling units, including a lien on the stock allocated to
a dwelling unit in a cooperative housing corporation, or on a residential
manufactured home, a lender may not exercise its option pursuant to a due-on-
sale clause upon—
(1) the creation of a lien or other encumbrance subordinate to the
lender’s security instrument which does not relate to a transfer of rights of
occupancy in the property;
(2) the creation of a purchase money security interest for household
appliances;
(3) a transfer by devise, descent, or operation of law on the death of a
joint tenant or tenant by the entirety;
(4) the granting of a leasehold interest of three years or less not
containing an option to purchase;
(5) a transfer to a relative resulting from the death of a borrower;
(6) a transfer where the spouse or children of the borrower become an
owner of the property;
(7) a transfer resulting from a decree of a dissolution of marriage, legal
separation agreement, or from an incidental property settlement
agreement, by which the spouse of the borrower becomes an owner of the
property;
(8) a transfer into an inter vivos trust in which the borrower is and
remains a beneficiary and which does not relate to a transfer of rights of
occupancy in the property; or
(9) any other transfer or disposition described in regulations prescribed
by the Federal Home Loan Bank Board.

In other words, after the Garn-St. Germain Act, states can only regulate nine types of
due-on-sale clauses. In response to the Garn-St. Germain Act, Michigan created its own cause of
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 5

action for lendees harmed by one of those nine banned due-on-sale clauses. See MICH. COMP.
LAWS § 445.1626 (“A lender shall not enforce a due-on-sale clause in a residential real property
loan in any circumstances under which enforcement is prohibited under section 341(d) of the
Garn-St. Germain depository institutions act of 1982, 12 U.S.C. 1701j-3, as currently in force.”);
MICH. COMP. LAWS § 445.1628 (creating a private cause of action for a violation of § 445.1626).

To fulfill our obligation of ascertaining subject matter jurisdiction, we must determine
whether a private cause of action “arises under” the statute sufficient to confer federal subject
matter jurisdiction. The “arising under” gateway into federal court in fact has two distinct paths:
1) “litigants whose causes of action are created by federal law,” and 2) “state-law claims that
implicate significant federal issues.” Hampton, 683 F.3d at 711 (quoting Eastman v. Marine
Mech. Corp., 438 F.3d 544, 550 (6th Cir. 2006)). Because the Garn-St. Germain Act does not
meet this first test, we join those courts, including this one, that have concluded 12 U.S.C.
§ 1701j-3 does not establish subject matter jurisdiction based on a federal cause of action.
Turman v. Wells Fargo Bank, N.A., No. 16-6546, 2018 WL 1840199, at *2 (6th Cir. Mar. 21,
2018) (order) (dismissing because “section 1701j–3 . . . does not provide a right of action”);
Dupuis, 589 F. Supp. at 823 (concluding “§ 1701j–3(d)(1) does not create a cause of action for
damages”); Nelson v. Nationstar Mortg. LLC, No. 7:16-CV-00307-BR, 2017 WL 1167230, at *2
(E.D.N.C. Mar. 28, 2017) (dismissing because “the Garn-St. Germain Act does not create a
cause of action for damages”). As we explain below, subject matter jurisdiction is also not
established under the second test.

A. Causes of Action Created by Federal Law

“[T]he vast majority of cases brought under the general federal-question jurisdiction of
the federal courts are those in which federal law creates the cause of action.” Hampton, 683 F.3d
at 711 (quoting Merrell Dow, 478 U.S. at 808). To determine whether a private cause of action
exists, we must begin with the text of the statute. Touche Ross & Co. v. Redington, 442 U.S.
560, 568 (1979). The cause of action may be express, Ohlendorf v. United Food & Commercial
Workers Int’l Union, Local 876, 883 F.3d 636, 640 (6th Cir. 2018), or implied, California v.
Sierra Club, 451 U.S. 287, 292–93 (1981).
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 6

1. Express Cause of Action

The Garn-St. Germain Act does not create an express cause of action because it does not
state, “in so many words, that the law permits a claimant to bring a claim in federal court.”
Ohlendorf, 883 F.3d at 640 (quoting Traverse Bay Area Intermediate Sch. Dist. v. Mich. Dep’t of
Educ., 615 F.3d 622, 627 (6th Cir. 2010)). Section 1701j-3(b)(1) tells states that they cannot
pass laws that restrict the use of due-on-sale clauses, subject to nine exceptions. And while
§ 1701j-3(d) lists the nine things a lender may not do, it does not offer an aggrieved lendee
recourse in federal court.

2. Implied Cause of Action

Nor does the Garn-St. Germain Act create an implied cause of action. “[T]he fact that a
federal statute has been violated and some person harmed does not automatically give rise to a
private cause of action in favor of that person.” Touche Ross, 442 U.S. at 568 (quoting Cannon
v. Univ. of Chicago, 441 U.S. 677, 688 (1979)). Instead, we must discern congressional intent
before implying a remedy. Thompson v. Thompson, 484 U.S. 174, 179 (1988). Absent
congressional intent, we may not imply a remedy, no matter how desirable it may be. Bowling
Green v. Martin Land Dev. Co., 561 F.3d 556, 559 (6th Cir. 2009).

When Congress wishes to create new rights—even implied rights of action—“it must do
so in clear and unambiguous terms.” Gonzaga Univ. v. Doe, 536 U.S. 273, 290 (2002).
Congress implies a right of action when its rights-creating language is “clear and unambiguous.”
Ohlendorf, 883 F.3d at 641 (quoting McCready v. White, 417 F.3d 700, 703 (7th Cir. 2005)).
Thus, the statute must specify the right and identify the beneficiary.2 Id.; see also Alexander v.
Sandoval, 532 U.S. 275, 289 (2001) (“Statutes that ban conduct but do not identify specific
beneficiaries do not suffice.”). This Act does not do both. Section 1701j-3(a)(1) bans states
from passing laws restricting due-on-sale clauses, and § 1701j-3(d) bans lenders from exercising

2For example, a statute that merely prohibits the funding of “any educational agency or institution which
has a policy or practice of permitting the release of education records” does not create any rights, even though
students may benefit from the statute and have an interest in enforcing it. Ohlendorf, 883 F.3d at 641 (quoting
Gonzaga, 536 U.S. at 287–88). Similarly, “a statute that authorizes federal agencies ‘to effectuate the provisions of
[a ban on intentional race discrimination] . . . by issuing rules, regulations, or orders of general applicability’” does
not create any rights. Id. (quoting Alexander v. Sandoval, 532 U.S. 275, 288–89 (2001)).
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 7

due-on-sale clauses in certain scenarios. Neither section identifies specific beneficiaries.
Although mortgagors may benefit, because § 1701j-3 “focus[es] on the person[s] regulated rather
than the individuals protected,” Sandoval, 532 U.S. at 289, and does not unambiguously specify
a beneficiary, no right of action can be implied.

This conclusion is consistent with the legislative history, assuming it is even relevant, 3
and the traditional areas of state regulation. The legislative history is silent regarding a remedy;
at most, it references the general goal of the statute to preempt state laws restricting due-on-sale
clauses. “The Senate Bill provides a federal preemption of state laws and judicial decisions
which restrict the enforcement of due-on-sale clauses in real property loans . . .” S. REP. NO. 97-
641, at 89 (1982) (Conf. Rep.). Nothing in the legislative history of the Garn-St. Germain Act
suggests that Congress intended to displace all state mortgage laws. Instead, all it does is draw a
line for state regulation of due-on-sale clauses—with certain exceptions—and leaves states in
control of the quintessentially state areas of contract and mortgage law. See § 1701j-3(b)(2)
(“the exercise by the lender of its option pursuant to such a clause shall be exclusively governed
by the terms of the loan contract, and all rights and remedies of the lender and the borrower
shall be fixed and governed by the contract” (emphasis added)); see also 12 C.F.R. § 591.4
(same). Thus, in states that allow judicial foreclosure actions, a mortgagor can raise as a defense
an exception to the Garn-St. Germain Act in state court.4 Dupuis, 589 F. Supp. at 822. And in
states that allow non-judicial foreclosures, like Michigan, the state can create its own cause of
action that allows a mortgagor to preemptively to raise the defense. See, e.g., MICH. COMP.
LAWS §§ 445.1626, 445.1628. But there is nothing implicit in the Act granting immediate access
to federal court.

3See Gonzaga, 536 U.S. at 286 (“[W]here the text and structure of a statute provide no indication that
Congress intends to create new individual rights, there is no basis for a private suit, whether under § 1983 or under
an implied right of action.” (emphasis added)); Sandoval, 532 U.S. at 288 (“We therefore begin (and find that we
can end) our search for Congress’s intent with the text and structure of Title VI.” (emphasis added)).
4Butthis, as we know, is not a free ticket for removal into federal court because the cause of action must be
determined from the face of the complaint. Hampton v. R.J. Corman R.R. Switching Co., 683 F.3d 708, 711 (6th
Cir. 2012).
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 8

B. State Law Implicates Substantial Question of Federal Law

State laws that implicate a “substantial question of federal law” also open the door into
federal court. See Grable & Sons Metal Prod., Inc. v. Darue Eng’g & Mfg., 545 U.S. 308, 312
(2005). We ask whether (1) “a state-law claim necessarily raise[s] a stated federal issue,”
(2) that is “actually disputed and substantial,” (3) “which a federal forum may entertain without
disturbing any congressionally approved balance of federal and state judicial responsibilities.”
Id. at 314. This pathway is a “slim category,” Empire Healthchoice Assur., Inc. v. McVeigh,
547 U.S. 677, 701 (2006), that is to be read narrowly, Mikulski v. Centerior Energy Corp.,
501 F.3d 555, 568 (6th Cir. 2007) (en banc). In this case, MICH. COMP. LAWS §§ 445.1626 and
445.1628 reference the Garn-St. Germain Act, but the question they raise is not sufficiently
substantial to justify federal question jurisdiction.5

Grable provides direction here. In that case, the Supreme Court held that a state quiet
title action implicated a substantial question of federal law because it required the interpretation
of the federal tax code. Grable, 545 U.S. at 314–15. The entire dispute would be resolved by
interpretation of the federal provision. Id. at 315. The Court reasoned that the question was
substantial because the IRS “has a strong interest in the prompt and certain collection of
delinquent taxes” and a direct interest in vindication via a federal forum. Id. (citation omitted).
The Court also explained that federal judges have more experience in federal tax matters. Id.
Finally, the Court noted that allowing disputes over federal tax title provisions would only have a
“microscopic impact on the federal-state division of labor.” Id.

Applying three-prong test articulated in Grable makes it clear that no federal question
exists here. While the first prong favors jurisdiction because §§ 445.1626 and 445.1628
reference the Garn-St. Germain Act and nominally raise a federal issue,6 the second and third

5Notably, the dissent does not argue that the Garn-St. Germain Act contains an express or implied cause of
action, and therefore implicitly acknowledges that Congress did not intend to create one.
6The dissent suggests that “the case for subject matter jurisdiction here is even stronger” than in City of
Chicago v. Int'l Coll. of Surgeons, 522 U.S. 156 (1997) (ICS) because the claim would not exist without the federal
statute. Dissenting Op. at 19. In ICS, the Chicago Landmarks Commission made a preliminary determination that
seven buildings qualified for designation as a landmark district. ICS, 522 U.S. at 159–60. The plaintiffs challenged
the administrative decision by filing two complaints in state court via a state cause of action—the Illinois
Administrative Review Law—alleging Due Process, Equal Protection, and Takings Clause violations under the Fifth
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 9

prongs disfavor jurisdiction. This question is not substantial and creating a gateway into federal
court will specifically disrupt the deference Congress gave to state courts to regulate in their
traditional areas of expertise.

For the second prong of the Grable analysis, there are four “substantiality” factors to
consider: (1) “whether the case includes a federal agency, and particularly, whether the agency’s
compliance with the federal statute is in dispute”; (2) “whether the federal question is important
(i.e. not trivial)”; (3) “whether a decision on the federal question will resolve the case (i.e. the
federal question is not merely incidental to the outcome)”; and (4) “whether a decision on the
federal question will control many other cases (i.e. the issue is not anomalous or isolated).”
Mikulski, 501 F.3d at 570. All four cut against finding substantiality. The first factor “is both
objective and apparent, and in this case weighs against characterizing the federal interest as
substantial because there is no federal agency in this dispute.”7 Id. (citing Empire, 547 U.S. at
700). The second factor—which focuses on “the importance of the issue to the federal system as
a whole,” Gunn v. Minton, 568 U.S. 251, 260 (2013)—also weighs against substantiality. The
Garn-St. Germain Act sets a limit on state regulation in one very particular aspect of state
property law. See Columbia Gas Transmission, LLC v. Singh, 707 F.3d 583, 590 (6th Cir. 2013)
(holding that the second factor “point[ed] away from a substantial federal interest” when the
claim involved “a typical state-law property issue”). It is hard to fathom how determining the
contours of the nine circumstances in which states may prohibit a lender from enforcing a due-
on-sale clause is especially important to the federal government. As for the third factor, even if

and Fourteenth Amendments. Id. at 160. The defendants removed to federal court, and the Supreme Court
eventually considered whether federal courts have original jurisdiction to review claims that local administrative
actions violate federal law. Id. at 163. The Court summarily concluded that the federal claims at issue
“unquestionably” raised substantial issues of federal law. Id. at 164. That is not the case here, where no
constitutional issues are at play. See Grable, 545 U.S. at 320 n.7 (“[C]onstitutional questions may be the more
likely ones to reach the level of substantiality that can justify federal jurisdiction.” (citation omitted)). In addition,
ICS preceded Grable, and the Grable Court did not rely on ICS to craft its three-prong test that we apply today. See
id. at 313.
7The dissent mentions that the Federal Home Loan Mortgage Corporation and the Federal National
Mortgage Corporation have an interest in the stability of the housing market and the uniform interpretation of the
Garn-St. Germain Act. Dissenting Op. at 22. While that may be true, the nine exceptions at issue leave so little
room for dispute that it should make no difference whether a state or federal court interprets them. And in any
event, those agencies are “not affected by the resolution of the dispute between these two parties.” Mikulski,
501 F.3d at 570.
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 10

we were to consider the parties’ claims, they dispute whether a due-on-sale clause was even
used, much less whether it was invoked in violation of the Garn-St. Germain Act’s nine
exceptions. We would not have to interpret federal law to resolve the dispute, and, even if we
did, other issues would remain. Finally, the fourth factor cuts against substantiality. As the
dissent notes, “only two states—Michigan and New Mexico—supply a cause of action for a
violation of 12 U.S.C. § 1701j-3,” Dissenting Op. at 26, so it is doubtful that a decision on the
federal question will control many other cases. Considering these factors, the federal question
embedded in Plaintiffs’ state law claim is not substantial.

As for the third Grable prong, entertaining the state law claim would certainly disturb the
balance of state and federal responsibilities. Grable, 545 U.S. at 314. We start by noting, again,
that Congress chose not to provide a remedy. Mikulski, 501 F.3d at 573 (explaining while “the
absence of a cause-of-action provision is not determinative, [it] certainly provides a starting point
for this part of the analysis”). And with good reason—state contract and property law are
uniquely within the province of the states. Shifting resolution of such traditional state law claims
from state courts to federal courts when the Garn-St. Germain Act does nothing more than
articulate a straightforward prohibition to the states, with nine plain exceptions that states are
perfectly capable of interpreting, is an unnecessary usurpation of state law domain.8 We see no
reason to manage state-law mortgage cases that would otherwise be barred from federal court
because of such a simple determination of law. This case is a far cry from Grable, where the
government was concerned about its ability to collect delinquent taxes, 545 U.S. at 315, or Smith
v. Kansas City Title & Trust Co., 255 U.S. 180 (1921) where the government’s constitutional
ability to issue bonds was called into question. And while we will not speculate on the amount
of cases this would bring into federal court, “even if the actual number of cases proved not to be
overwhelming, or even uncomfortably burdensome, it appears unlikely that Congress—through
its silence—intended to open the federal court door quite so wide.” Mikulski, 501 F.3d at 574.

8For example, state courts are perfectly capable of determining whether a lender exercised a due-on-sale
clause upon “a transfer to a relative resulting from the death of a borrower” or “a transfer where the spouse or
children of the borrower become an owner of the property.” 12 U.S.C. § 1701j-3(d)(5)–(6).
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 11

The dissent argues “That a federal court would throw the Act’s interpretation into the
hands of the states, and only the states, seems to undermine Congress’s expressed intent in the
text of the Act.” Dissenting Op. at 27. To the contrary, the only expressed intent in the statute is
that states cannot ban due-on-sale clauses unless they qualify as one of the nine types listed in
§ 1701j-3(d). Congress may provide a remedy and chose not to do so with this statute.
Congress, as it intended, left that task in the capable hands of the states because it did not want to
disrupt the federal-state balance in an area where states have a much stronger interest and
experience.

“As every schoolchild learns,” the states “retain substantial sovereign authority” of their
own property laws. Gregory v. Ashcroft, 501 U.S. 452, 457 (1991) (citing THE FEDERALIST NO.
45, at 292–93 (James Madison) (C. Rossiter ed., 1961) (“The powers reserved to the several
States will extend to all the objects which, in the ordinary course of affairs, concern the lives,
liberties, and properties of the people . . . .”)). As courts of limited jurisdiction, we can only
entertain questions of state law when Congress has authorized us to do so (i.e. diversity
jurisdiction) or when a state law raises a substantial question of federal law. This is not that case.

IV. CONCLUSION

In sum, because the federal statute does not create a cause of action, and the federal issue
nested inside Plaintiffs’ state law cause of action is not substantial, the district court lacked
subject matter jurisdiction. Therefore, we VACATE the district court’s judgment with
instructions to REMAND the case to state court.
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 12

_________________

DISSENT
_________________

KAREN NELSON MOORE, Circuit Judge, dissenting. This case began as a simple
dispute about whether a mortgage lender enforced a due-on-sale clause in a circumstance where
federal law prohibited enforcement. But the case has now veered into the significant question of
the meaning of the words “arising under” in 28 U.S.C. § 1331. The majority concludes that there
is no substantial federal question in this case, and consequently, that the federal courts lack
subject matter jurisdiction. In fact, the claim at issue turns exclusively on federal law and is well
within federal question jurisdiction under § 1331. For the reasons that follow, I respectfully
dissent.

I. BACKGROUND

Audrey Bantom is the personal representative for the Estate of Robert Cornell, Jr., and
Anthony Cornell (collectively “Plaintiffs”) resided in the Detroit, Michigan home that is at issue
in this case. R. 4 (Am. Compl. at 1) (Page ID #23). Bayview Loan Servicing and Thien Hoang
Tran are the Defendants. Before his death, Robert Cornell, Jr. entered into a mortgage contract
with Bayview, which is a mortgage lending company. Id. at 2 (Page ID #24). Most relevant to
the issue of our subject matter jurisdiction, the Plaintiffs allege that Bayview illegally foreclosed
on the home upon the death of their father, Robert, in violation of the federal Garn-St. Germain
Act, 12 U.S.C. § 1701j-3(d), and Michigan law, Mich. Comp. Laws § 445.1626. Id. at 4–5 (Page
ID #26–27). Michigan law provides a cause of action for this claim against Bayview. Mich.
Comp. Laws § 445.1628.

Perhaps not the most well-known statute, the Garn-St. Germain Act (“the Act”) is
nonetheless a fascinating piece of legislation in that Congress chose to interpose in areas of
law—property and contract—traditionally regulated by the states. Under the Act, a due-on-sale
clause is presumptively valid, 12 U.S.C. § 1701j-3(b)(1)–(2), unless one of nine exceptions
apply, id. at § 1701j-3(d). A due-on-sale clause is defined as, “a contract provision which
authorizes a lender, at its option, to declare due and payable sums secured by the lender’s
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 13

security instrument if all or any part of the property, or an interest therein, securing the real
property loan is sold or transferred without the lender’s prior written consent.” Id. at § 1701j-
3(a)(1). The Act prohibits enforcement of such clauses upon, among other circumstances, “a
transfer to a relative resulting from the death of a borrower” or “a transfer where the spouse or
children of the borrower become an owner of the property.” Id. at § 1701j-3(d)(5)–(6).
Michigan law then provides a cause of action for money damages, injunctions, and/or
declaratory judgments against lenders who violate these provisions of the Garn-St. Germain Act.
Mich. Comp. Laws §§ 445.1626, 445.1628. Michigan, or any state, cannot add to or subtract
from the federal law because the Garn-St. Germain Act preempts any state laws, including
judicial decisions, contrary to its provisions. 12 U.S.C. § 1701j-3(b)(1).

The context in which Congress enacted this law sheds light on the Act’s importance.
When Congress passed the statute, state legislatures and state courts restricted the enforcement of
due-on-sale clauses. Id.; S. Rep. No. 97–536 at 20–21 (1982) (“Senate Rep.”). Congress sought
to help new home buyers and lenders, as well as to clear-up confusion on the enforceability of
such clauses. Senate Rep. at 20–21. State due-on-sale clause restrictions “adversely affect[ed]
secondary mortgage markets, which rely on uniform, homogenous mortgage documents to
efficiently operate,” and these “restrictions . . . caused the Federal Home Loan Mortgage
Corporation and the Federal National Mortgage Corporation to alter their investment practices in
several states.” Id. at 21. Thus, these state restrictions were adversely affecting entities
sponsored by the federal government, and consequently, Congress intervened to protect the
federal interests at play. At the same time, Congress balanced helping new homebuyers and
lenders, 12 U.S.C. § 1701j-3(b)(1)–(2), with safeguards for existing homeowners, id. at § 1701j-
3(d). Clearly, in Congress’s judgment uniformity was required because the states were not
adequately doing the job of balancing these various interests. See id. at § 1701j-3(b)(1); Senate
Rep. at 20–21.1

1The Office of the Comptroller of the Currency, an independent bureau within the Department of Treasury,
shared Congress’s concerns. As it explained in an interpretive letter after the Act was passed:
Pursuant to the Comptroller’s power to regulate real estate lending activities of national
banks, on September 22, 1981, this Office issued a proposed rule . . . to ensure that the
congressional intent of encouraging national bank participation in the real estate finance market
would not be thwarted by state laws. The proposed regulation would have validated the inclusion
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 14

II. CLARIFYING THE FOCUS OF SUBJECT MATTER JURISDICTION

A. Federal Question or “Arising Under” Jurisdiction

A defendant can remove a case from state court if the plaintiff could have originally
brought the case in federal court. 28 U.S.C. § 1441(a). The federal district courts have “original
jurisdiction of all civil actions arising under the Constitution, laws, or treaties of the United
States.” 28 U.S.C. § 1331. The Plaintiffs originally filed their case in state court. Then the
Defendants removed the case to federal court on the basis that “Plaintiffs’ Complaint is founded
upon claims or rights arising under the laws of the United States.” R. 1 (Notice of Removal at 2–
3) (Page ID #2–3).2 Whether removal was proper, therefore, solely depends on the existence of a
federal question such that the case “arises under” federal law.

First, we must carefully dissect the anatomy of this case to determine what is actually
relevant for our jurisdictional inquiry. The Plaintiffs bring a claim under 12 U.S.C. § 1701j-3,
and Mich. Comp. Laws § 445.1626. Michigan law prohibits a lender from enforcing a “due-on-
sale clause in a residential real property loan in any circumstances” that are prohibited by the
Garn-St. Germain Act. Mich. Comp. Laws § 445.1626; 12 U.S.C. § 1701j-3(d). The state law,
therefore, turns exclusively on the Garn-St. Germain Act, which creates the federal right that the
Plaintiffs argue Bayview violated. As the Plaintiffs note, Mich. Comp. Laws § 445.1628, in turn,
supplies a cause of action. Appellant’s Br. at 8–9; Mich. Comp. Laws § 445.1628(3)
(authorizing suits for declaratory judgment and/or an injunction); id. at § 445.1628(4)
(authorizing actions for money damages). Accordingly, the state-created cause of action is

of due-on-sale clauses in real estate loans held by national banks in their portfolios and would
have made such clauses fully enforceable. In addition, the regulation would have revalidated due-
on-sale clauses in loans made prior to the rendering of a state court decision or the passage of a
state statute impairing the enforceability of such clauses.
Some time after publication of the proposed rule, Congress began consideration of
federal preemption of state laws limiting the enforceability of due-on-sale clauses. This
deliberation ultimately resulted in section 341 of the Garn-St Germain Depository Institutions Act
of 1982 (“Act”) (codified at 12 U.S.C. § 1701j-3) that established a federal preemption of state
restrictions on the enforcement of due-on-sale clauses.
Office of the Comptroller of the Currency, OCC Interpretive Letter (Jan. 13, 1986), 1986 WL 144000 at *1.
2The Plaintiffs allege that the amount in controversy exceeds only $25,000, R. 4 (Am. Compl. at 2) (Page
ID #24), and therefore, the Plaintiffs fail to overcome the $75,000 threshold for diversity jurisdiction under
28 U.S.C. § 1332(a).
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 15

simply a remedial vehicle for a plaintiff to vindicate a right, the contours of which are defined
entirely by federal law.

Turning to 28 U.S.C. § 1331, for a federal court properly to exercise federal question
jurisdiction under that statute, a federal question must appear on the face of a plaintiff’s well-
pleaded complaint. Louisville & Nashville R.R. Co. v. Mottley, 211 U.S. 149, 152 (1908). Here,
the Plaintiffs satisfied the well-pleaded complaint rule. Count I of the complaint recites the
§ 1701j-3(d) exemptions on which the Plaintiffs rely. R. 4 (Am. Compl. at 4–5) (Page ID #26–
27). Putting aside (for now) issues of whether this federal question is sufficiently “substantial,”
it is worth emphasizing that only this claim matters for the inquiry of whether we have federal
question jurisdiction. See City of Chicago v. Int’l Coll. of Surgeons, 522 U.S. 156, 166–68
(1997) (ICS). Although the Plaintiffs assert other claims that rest entirely on state law, a federal
court has discretion to hear those claims or not. 28 U.S.C. § 1367(c). In other words, this case
has two distinct jurisdictional hooks: Count I (the Garn-St.Germain Act and Mich. Comp. Laws
§§ 445.1626, 445.1628) is on the federal question hook; Counts II–V are on the discretionary
supplemental jurisdiction hook. To analyze properly whether we have subject matter
jurisdiction, we must imagine this case as if the Plaintiffs pleaded only the purported federal
question claim. See ICS, 522 U.S. at 166 (“The District Court’s original jurisdiction derives
from [a plaintiff’s] federal claims, not its state law claims.”). If a federal court determines that
jurisdiction lies over the claim with the federal question, it must hear the claim; and then,
separately, a court has discretion over whether to welcome the other claims that rest on state law
into federal court under § 1367. The Supreme Court has rejected as “mistaken” and “erroneous”
the “view that the district courts lack jurisdiction even over the federal claims” because “other,
nonfederal claims somehow take [a] complaint[] in [its] entirety (including the federal claims)
out of the federal courts’ jurisdiction.” Id. at 168. Thus, as it relates to subject matter
jurisdiction, the entire focus is on the Plaintiffs’ first claim, which incorporates federal law.

The majority misunderstands the importance of this separation between the purported
federal question claim and the other purely state-law claims. The majority reasons that, even if
this case required the interpretation of federal law, “other issues would remain”; and the majority
also believes that hearing the federal question claim here would require federal courts “to
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 16

manage state-law mortgage cases.” Majority Op. at 10. The majority’s reasoning misses the
mark. As explained, federal courts can hear the federal question claim and then decline to hear
the other state-law claims. Section 1367 gives federal courts the discretion to do so.

B. The Federal Cause of Action Analysis is Separate and Distinct from Subject Matter
Jurisdiction

The question of whether a federal court has subject matter jurisdiction is separate from
whether there is a federal cause of action. First, if the two analyses were one and the same, then
the longstanding rule that the federal courts can exercise federal question jurisdiction over state-
law causes of action would be wrong—and it is not. See Grable & Sons Metal Prods., Inc. v.
Darue Eng’g & Mfg., 545 U.S. 308, 312–13 (2005); ICS, 522 U.S. at 164; Smith v. Kansas City
Title & Trust Co., 255 U.S. 180, 199 (1921); Hopkins v. Walker, 244 U.S. 486, 490–91 (1917);
cf. De Sylva v. Ballentine, 351 U.S. 570, 571–82 (1956) (deciding on the merits a state-law claim
that involved the federal Copyright Act, without discussing jurisdiction), cited in Richard H.
Fallon, Jr., et al., Hart & Wechsler’s The Federal Courts and the Federal System 836 n.6 (7th ed.
2015). Second, Shoshone Mining Co. v. Rutter would be wrongly decided for the opposite
reason. There, the Supreme Court held that the federal courts lacked jurisdiction because the
claim turned on “a determination of the meaning and effect of certain local rules and customs”
related to mining rights, even though the plaintiff pleaded a federal cause of action. Shoshone,
177 U.S. 505, 509–10 (1900) (“A statute authorizing an action to establish a right is very
different from one which creates a right to be established. An action brought under the one may
involve no controversy as to the scope and effect of the statute, while in the other case it
necessarily involves such a controversy, for the thing to be decided is the extent of the right
given by the statute.”); see also Shulthis v. McDougal, 225 U.S. 561, 569–70 (1912) (a plaintiff’s
claim to title derived from federal law, but the federal statutes were not subject to “any
controversy respecting their validity, construction or effect”). The presence or absence of a
federal cause of action is not dispositive, in either direction, for the jurisdictional question.

For a classic example of this separation of remedies and jurisdiction, consider the
declaratory judgment statute. 28 U.S.C. § 2201. Congress added declaratory relief to the menu
of remedies available to plaintiffs in federal court, but Congress did not expand federal
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 17

jurisdiction in doing so. Skelly Oil Co. v. Phillips Petroleum Co., 339 U.S. 667, 671 (1950). On
the other hand, even if Congress explicitly forecloses a federal remedy, it does not follow that
Congress intends to foreclose federal jurisdiction. Erwin Chemerinsky, Federal Jurisdiction 307
(6th ed. 2012). Of course, if Congress wished to limit the federal courts’ jurisdiction, it could
enact such limits and vest jurisdiction in the state courts. See, e.g., 15 U.S.C. § 2310(d) (setting
certain thresholds for a claim under the Magnuson-Moss Warranty Act, which if not met, deprive
a federal court of jurisdiction to hear such a claim). Such a limit does not exist in the Garn-St.
Germain Act.

Importantly, the remedial question is easily answered in this case: Michigan law
explicitly provides a cause of action. Mich. Comp. Laws § 445.1628. We need to answer only
whether the federal courts have jurisdiction over a state-created cause of action under § 1331.
With that, we can explore the statute.

III. SUBJECT MATTER JURISDICTION IN THIS CASE

Under Article III of the United States Constitution, a federal question exists whenever
federal law is a potential ingredient in the litigation. Osborn v. Bank of the U.S., 22 U.S. (9
Wheat.) 738, 823 (1824). Section 1331, however, requires more than a mere federal ingredient.
And in this case, federal law is the whole recipe.

For over a century, the Supreme Court has recognized that federal question jurisdiction
will lie over state-law claims that have a federal question nested inside the state claim. See, e.g.,
Grable, 545 U.S. at 312–13; Smith, 255 U.S. at 199; Hopkins, 244 U.S. at 490–91. The test is
whether a state-law claim: (1) raises a federal issue, (2) that is “actually disputed and
substantial,” and where (3) hearing the state-law claim in federal court would not disturb the
federal-state balance of judicial responsibilities. Grable, 545 U.S. at 314. In sum, although a
plaintiff cannot use a “federal issue” as an automatic “password” to open federal courts, a federal
statute plus a federal cause of action is also not the only combination that unlocks jurisdiction
under § 1331. Id. A federal issue nested inside a state-law cause of action can also suffice.

The majority engages with this test and Supreme Court precedent in a cursory and
conclusory fashion, and it fails to understand the consequences of the Garn-St. Germain Act.
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 18

The majority states that “the Garn-St. Germain Act does nothing more than articulate a
straightforward prohibition to the states.” Majority Op. at 10. Not so. The Act, in addition to
preempting state laws contrary to its provisions, § 1701j-3(b)(1), prescribes the situations in
which due-on-sale clauses must be enforced, § 1701j-3(b)(2), details specific situations in which
such clauses cannot be enforced, § 1701j-3(d), and authorizes the Federal Home Loan Bank
Board to issue rules and regulations3 pursuant to the Act, § 1701j-3(e). Furthermore, the
majority suggests that Congress’s choice to preempt state laws in this area is insignificant. To
the contrary, Congress thought a sufficiently significant problem existed, as well as a sufficient
threat to federal interests, which prompted intervention into the traditionally state-regulated areas
of property and contract. Senate Rep. at 20–21. When Congress passed the Garn-St. Germain
Act and preempted any state law contrary to the Act’s provisions, Congress also created federal
rights. Most relevant here, individuals have “a federal right . . . to be free from a lender’s
purported enforcement of a due-on-sale clause” in the circumstances outlined in § 1701j-3(d).
See Dupuis v. Yorkville Fed. Sav. & Loan Ass’n, 589 F. Supp. 820, 822 (S.D.N.Y. 1984).4

A more thorough examination of whether the federal question in this case is “substantial”
is needed. But as a preliminary matter, no one contests that the Plaintiffs raise a federal issue
and that it is actually disputed. As it relates to Count I of the complaint, the applicability of the
Act “is actually in dispute,” and “it appears to be the only legal or factual issue contested” for
that claim. See Grable, 545 U.S. at 315. Therefore, I will focus on whether this federal question
is sufficiently “substantial” (it is), and whether welcoming this claim into federal court would
disturb the federal-state balance of judicial responsibilities (it would not).

3See, e.g., 12 C.F.R. § 591.1–591.6 (regulations on the preemption of state due-on-sale laws); 12 C.F.R.
§ 591.5(a) (“due-on-sale practices of Federal savings associations and other lenders shall be governed exclusively by
the Office’s regulations, in preemption of and without regard to any limitations imposed by state law on either their
inclusion or exercise including, without limitation, state law prohibitions against restraints on alienation,
prohibitions against penalties and forfeitures, equitable restrictions and state law dealing with equitable transfers.”).
4Notably, even the district court in Dupuis recognized the existence of a federal right, despite the district
court’s conclusion that the Act itself did not create a federal cause of action. 589 F. Supp. at 822–23. I do not
express any view on whether the Act contains a cause of action here. But, for the sake of argument, even if the Act
did not contain a cause of action, that is simply irrelevant because Michigan law provides one.
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 19

A. The Substantial Federal Question

1. Whether Federal Law is Directly and Exclusively Relevant

To start, this case should be situated alongside relatively recent Supreme Court precedent
that found federal question jurisdiction to lie over a state-law claim. In a sense, the case for
subject matter jurisdiction here is even stronger than in Grable and ICS. Grable involved a quiet
title action, the outcome of which turned on whether proper notice was given under a federal
statute. 545 U.S. at 311. ICS involved federal constitutional claims brought via a cause of action
created by state law. 522 U.S. at 164; see also 735 Ill. Comp. Stat. 5/3-103. In both cases, as
here, the validity of the claims depended exclusively on federal law. But not all quiet title
actions are based upon a federal question, and the same is true for claims under the Illinois
statute at issue in ICS. This Michigan cause of action involved here, however, would literally
not exist without federal law. The relevant substance of the Michigan law reads, “[a] lender shall
not enforce a due-on-sale clause in a residential real property loan in any circumstances under
which enforcement is prohibited under section 341(d) of the Garn-St. Germain depository
institutions act of 1982 . . . as currently in force.” Mich. Comp. Laws § 445.1626. For every
single action brought under this state-law claim, a federal law will be the singular consideration.

Supreme Court precedent does not draw bright lines when it comes to determining
whether a federal question is “substantial” or “essential” to a plaintiff’s claim. As Justice
Cardozo explained, whether a case “arises under” federal law calls for a “common-sense
accommodation of judgment to [the] kaleidoscopic situations” that present a federal question.
See Gully v. First Nat’l Bank, 299 U.S. 109, 117 (1936); see also Grable, 545 U.S. at 313. Of
course, a federal question could always be “lurking in the background” such that “countless
claims of right can be discovered to have their source or their operative limits in the provisions
of a federal statute or in the Constitution itself.” Gully, 299 U.S. at 117–18. Thus, “[t]o set
bounds to the pursuit [for a substantial federal question], the courts have formulated the
distinction between controversies that are basic and those that are collateral, between disputes
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 20

that are necessary and those that are merely possible.” Id. at 118.5 One could imagine a
spectrum of cases: on one end are cases where the federal issue is “so far removed from plain
necessity,” id. at 117, that it is not substantial or essential; on the other end are cases where the
federal issue is directly and actually in dispute and a claim turns exclusively on the federal issue,
see, e.g., Grable, 545 U.S. 308; ICS, 522 U.S. 156; Smith, 255 U.S. 180, such that the federal
issue is substantial and essential. Under this view, this prong of the Grable test turns on how
relevant state law is to the underlying substance of a plaintiff’s claim as compared to federal law,
and whether the federal question is at the forefront of the case rather than simply “lurking in the
background.”

The Garn-St. Germain Act further bolsters the conclusion that substantive state law is
irrelevant because the Act preempts state laws contrary to its provisions. 12 U.S.C. § 1701j-
3(b)(1). For example, states cannot create new rights by expanding upon the exemptions listed
in the Act. Id. at § 1701j-3(d). That is not to say preemption alone confers jurisdiction, see
Empire HealthChoice Assurance, Inc. v. McVeigh, 547 U.S. 677, 697 (2006), but it leaves no
doubt that only federal law is relevant, cf. Hart & Wechsler at 835 (noting that the Supreme
Court in Empire “described with some skepticism the argument that federal law would govern”
the issue of an attorney’s share of a state tort settlement); Empire, 547 U.S. at 701 (stating that it
was “hardly apparent why a proper ‘federal-state balance’ would place such a nonstatutory issue
under the complete governance of federal law”). Indeed, one cannot seriously dispute that
federal law is at the forefront of the Plaintiffs’ claim here. This case does not involve a
settlement or attorney’s fees issue that is ancillary to the actual merits of the claim, Empire,
547 U.S. 677, or a “backward-looking” legal malpractice claim, the outcome of which will not
change the previously answered federal question, Gunn v. Minton, 568 U.S. 251, 261 (2013), or a
situation in which the violation of federal law does not entirely determine the success of a state
tort claim, see, e.g., Merrell Dow Pharm. Inc. v. Thompson, 478 U.S. 804 (1986) (violation of a

5Cf. McGoon v. N. Pac. Ry. Co., 204 F. 998, 1001 (D. N.D. 1913) (explaining that cases do not “arise
under” federal law where “the right asserted in the complaint was not a right created by federal law,” or “such law
was only indirectly and remotely” relevant in the litigation), cited in Puerto Rico v. Russell & Co., 288 U.S. 476,
483–84 (1933) (“[Federal jurisdiction] may not be invoked . . . merely because the plaintiff’s right to sue is derived
from federal law . . . . The federal nature of the right to be established is decisive—not the source of the authority to
establish it.”).
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 21

federal statute created a rebuttable presumption of negligence); Mikulski v. Centerior Energy
Corp., 501 F.3d 555, 571 (6th Cir. 2007) (en banc) (violation of a federal statute satisfied some,
but not all, of the elements of the state tort claim). Instead, the Plaintiffs in this case allege that
Bayview violated a federal law that “is directly drawn into question,” Smith, 255 U.S. at 201, and
their claim “turn[s] exclusively on federal law,” ICS, 522 U.S. at 164. In short, the entirety of
the Plaintiffs’ claim rises or falls on the answer to the federal question.

The majority misreads—and arguably goes against—Supreme Court precedent in its
attempt to distinguish the present case. The majority states that, in this case, “no constitutional
issues are at play,” as in ICS, and that the Supreme Court “did not rely on ICS to craft” the
Grable test. Majority Op. at 9 n.6. The first part of the majority’s statement is irrelevant and the
second part is false. Grable expressly rejects that, as to the substantiality question, we should
differentiate constitutional claims from statutory claims. 545 U.S. at 319–20 n.7 (“There is . . .
no reason in [§ 1331’s] text or otherwise to draw such a rough line.”). What’s more, Grable did
not turn a blind eye to ICS. To the contrary, the Court saw ICS as consistent with the
substantiality test articulated in Grable. See id. at 313 (citing ICS as an example of a case that
implicated a substantial federal issue). Simply put, the majority’s reasoning on these points
cannot be squared with ICS or Grable.

2. The Importance of the Federal Question to the Federal System as a Whole

Moreover, “[t]he substantiality inquiry under Grable looks . . . to the importance of the
issue to the federal system as a whole.” Gunn, 568 U.S. at 260. When we view the substantiality
issue through this lens, we focus our sight on a broader picture and not on the parties in this
particular dispute. Id. (stating that in Grable, the Court “primarily focused not on the interests of
the litigants themselves, but rather on the broader significance of the [federal] question”). The
majority here, however, somehow sees only state interests—an odd observation for a statute in
which Congress displaced a variety of state laws.
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 22

We have previously outlined four factors to consider when assessing the substantiality of
a federal interest:

(1) whether the case includes a federal agency, and particularly, whether that
agency’s compliance with the federal statute is in dispute; (2) whether the federal
question is important (i.e., not trivial); (3) whether a decision on the federal
question will resolve the case (i.e., the federal question is not merely incidental to
the outcome); and (4) whether a decision as to the federal question will control
numerous other cases (i.e., the issue is not anomalous or isolated).

Mikulski, 501 F.3d at 570. Importantly, none of these factors are dispositive, and not all are
equally applicable in any one case. Id. Further, this list of factors is not exhaustive. Id. (“While
certain of these factors may be more applicable than others in any given set of circumstances, no
single factor is dispositive and these factors must be considered collectively, along with any
other factors that may be applicable in a given case.”). I will address each in turn.

Though true that this case does not involve a federal agency and a federal agency’s
compliance with the Act is not in dispute, I question the weight of this factor in light of Gunn,
which the Supreme Court decided after our decision in Mikulski. See Gunn, 568 U.S. at 260. If
we follow Gunn’s admonition not to focus solely on the litigants at hand, see id., we should
instead consider the federal government’s interest in the question more generally. Before
Congress passed the Garn-St. Germain Act, government-sponsored entities (i.e., the Federal
Home Loan Mortgage Corporation and the Federal National Mortgage Corporation) were
altering their investment strategies in several states because of uncertainty and a lack of
uniformity related to due-on-sale restrictions. Senate Rep. at 21. Both entities are currently
under the conservatorship of the Federal Housing Finance Authority.6 While the FHFA and
these government-sponsored entities may not have a stake in this particular dispute, the federal
government has an interest in the stability of the housing market and uniform enforcement of
due-on-sale clauses, and consequently, the scope of the Garn-St. Germain Act’s exceptions in
§ 1701j-3(d). Notably, in Grable, the federal government also lacked a direct interest in the
property at issue in the case. 545 U.S. at 317 n.4. But the Court still concluded that the federal

6Federal Home Loan Mortgage Corporation, https://www.usa.gov/federal-agencies/federal-home-loan-
mortgage-corporation-freddie-mac (last visited Oct. 5, 2018); Federal National Mortgage Association,
https://www.usa.gov/federal-agencies/federal-national-mortgage-association-fannie-mae (last visited Oct. 5, 2018).
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 23

question could “claim[] the advantages thought to be inherent in a federal forum.” Id. at 313. If
a lack of uniformity among the states motivated Congress to pass this Act, and if we believe that
the federal courts provide the advantage of uniform interpretation, then the meaning of the Act
“is an important issue of federal law that sensibly belongs in a federal court.” Id. at 315.

We have described the second Mikulski factor—whether the federal question is important
and not trivial—as “far more subjective.” 501 F.3d at 570. But in the present matter, we can
look to a more objective factor: the statute itself. At risk of belaboring the point, clearly
Congress thought this issue was important enough to preempt state laws, including judicial
decisions, in areas traditionally reserved to the states. 12 U.S.C. § 1701j-3(b)(1). The upshot is
that this case does not involve “a typical state-law property issue.” Columbia Gas Transmission,
LLC v. Singh, 707 F.3d 583, 590 (6th Cir. 2013). The issue is the applicability of exemptions in
a federal statute. In Singh, we further noted that “there is no allegation that [state courts] are not
enforcing the property rights of natural gas companies or are otherwise impeding the [federal
interest in] safe transmission of natural gas in a way that would call for a federal forum.” Id.
Congress found the opposite to be true in this case. See Senate Rep. at 20–21. Not only that, but
Congress also gave the power to promulgate rules and to publish interpretations of the Act to a
federal agency. 12 U.S.C. § 1701j-3(e). Considering the context in which Congress enacted the
Garn-St. Germain Act, as well as the Act’s plain text, the majority cannot say with a straight face
that hearing this claim in federal court would “disrupt the deference Congress gave to state
courts to regulate in their traditional areas of expertise.” Majority Op. at 9. The inescapable fact
is that Congress took away such deference from the states and replaced state law with a federal
statute.

As stated, however, preemption alone should not be dispositive. In Mikulski, we also
noted that “the IRS has never issued a rule” or “ever litigated an action” involving the provision
of law at issue. 501 F.3d at 571. Thus, Mikulski’s analysis of this factor seemed to blend in with
the first factor because we continued to reference the IRS’s role (or lack thereof) in that case. Id.
at 570–71. First, unlike Mikulski, the Federal Home Loan Bank Board has promulgated
regulations in this area. See supra footnote 3. Second, Mikulski’s suggestion that more frequent
litigation counseled in favor of finding jurisdiction cannot be squared with Grable’s and Merrell
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 24

Dow’s concern about welcoming in a “horde” of state cases into federal court. See Grable,
545 U.S. at 318 (describing Merrell Dow). Grable itself also gives no indication that the IRS
had frequently litigated the question at issue in that case. Taken together, the first two factors
militate in favor of finding that the federal question here is substantial.

The third and fourth factors—will answering the federal question dispose of “the case”
and will it control numerous other “cases”—require some clarification on the meaning of
“case(s).” Mikulski, 501 F.3d at 570.

The “case” in the third factor is really the “claim” on which the federal court’s subject
matter jurisdiction rests. To reiterate, the sole claim that matters for whether we have subject
matter jurisdiction over this case is Count I of the Plaintiffs’ complaint, which alleges a violation
of the Garn-St. Germain Act. See supra II.A. The Supreme Court has flatly rejected the notion
that we should look at any non-federal claims, which would fall under the discretionary
supplemental jurisdiction in 28 U.S.C. § 1367, when we consider whether we have subject matter
jurisdiction over the purported federal question claim. ICS, 522 U.S. at 167–68. If we
concluded that the Plaintiffs’ federal question claim fails on the merits, that could, in fact,
dispose of the entire case because the federal court need not hear the other state claims.
28 U.S.C. § 1367(c). More importantly, the answer to the federal question will completely
resolve the relevant state claim because that claim turns exclusively on federal law.
Accordingly, this third factor weighs in favor of substantiality.

As to the fourth factor, we have to reconcile the “numerous other cases” language with
the concern, once again, that we could throw open the courthouse doors to too many state claims.
Admittedly, Gunn seems to suggest that “cases” refers to the amount of litigation. See 568 U.S.
at 262 (“If the question arises frequently, it will soon be resolved within the federal system,
laying to rest any contrary state court precedent; if it does not arise frequently, it is unlikely to
implicate substantial federal interests.”). On this point, Gunn (and Empire) seem to be in tension
with Grable. In discussing the federal-state balance issue, Grable thought the fact that “it will be
the rare state title case that raises a contested matter of federal law” counseled in favor of
welcoming the state-law claim into federal court. 545 U.S. at 315. Assuming for the sake of
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 25

argument that we should focus on the amount of litigation, this factor leans against substantiality.
The Garn-St. Germain Act does not appear to be frequently litigated.

But this factor is not dispositive and there are good reasons to give it less weight in the
instant case. For one, hostile state-court precedent caused federal entities to alter their
investment strategies in several states, which then motivated Congress to pass the Garn-St.
Germain Act in the first place. Senate Rep. at 21. Additionally, even if this issue is not
frequently litigated, due-on-sale clauses are common in mortgage contracts, and government-
sponsored entities, lenders, and home buyers and sellers may find it beneficial to know how far
the Act’s due-on-sale enforcement exemptions reach. Finally, even if this factor weighs against
substantiality, the other considerations already mentioned overwhelm any concern about the low
amount of litigation.

In sum, on balance the Mikulski factors weigh in favor of substantiality. Moreover, those
factors should be considered alongside the fact that federal law is directly and exclusively
relevant to the Plaintiffs’ state-law cause of action. In the “selective process which picks
substantial [federal questions] out of the web and lays the other ones aside,” Gully, 299 U.S. at
118, we can pick this federal question out of the web so that it can claim “the advantages thought
to be inherent in a federal forum,” Grable, 545 U.S. at 313.

B. The Federal-State Balance

This is not the sort of “garden variety” state case that has raised concerns in prior cases.
For example, this case is nothing like Merrell Dow. In that case, the exercise of federal question
jurisdiction over state-law tort claims “would have attracted a horde” of cases “raising other state
claims with embedded federal issues.” See Grable, 545 U.S. at 318; Merrell Dow, 478 U.S. 804.
In essence, the Supreme Court in Grable clarified that Merrell Dow was concerned about the
“tremendous number” of negligence cases that would be swept into federal court, Grable,
545 U.S. at 318, if federal question jurisdiction existed “solely because the violation of [a]
federal statute is said to be a ‘rebuttable presumption’ or a ‘proximate cause’ under state law,”
Merrell Dow, 478 U.S. at 812. In contrast, “it is the rare state quiet title action that involves
contested issues of federal law,” and consequently, jurisdiction over that action “would not
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 26

materially affect, or threaten to affect, the normal currents of litigation.” Grable, 545 U.S. at
319. We have previously read these cases together in this manner. See Hampton v. R.J. Corman
R.R. Switching Co., 683 F.3d 708, 712–13 (6th Cir. 2012) (noting the Supreme Court’s
distinction between “the rare state quiet title action” in Grable and the “garden variety state tort
claim” in Merrell Dow, and concluding that, because the case involved a garden variety
negligence claim, the case was more like Merrell Dow). As it pertains to this federal law, only
two states—Michigan and New Mexico—supply a cause of action for a violation of 12 U.S.C.
§ 1701j-3. See Mich. Comp. Laws §§ 445.1626, 445.1628; N.M. Stat. §§ 48-7-20, 48-7-24.7
Welcoming these claims into federal court, therefore, would not “disturb[] any congressionally
approved balance of federal and state judicial responsibilities,” or clog the “currents of litigation”
in federal court. See Grable, 545 U.S. at 314, 319.

Indeed, Merrell Dow required clarification. Many commentators noted the confusion and
uncertainty the case caused, eventually leading the Supreme Court to correct course in Grable.
See James E. Pfander, Principles of Federal Jurisdiction 101 (West 2006) (“Judged solely by the
degree to which it brought clarity to jurisdictional law, one might consider the Merrell Dow
decision a failure.”); see also Hart & Wechsler at 824–25; Chemerinsky at 306–07. The Court
also admitted as much in its most recent decision in this area. See Gunn, 568 U.S. at 258 (stating
that Grable was “an effort to bring some order to this unruly doctrine”). The confusion, of
course, centered on the relevance of the absence of a federal cause of action. Hart & Wechsler at
824–25; Chemerinsky at 306–07.

Meanwhile, the majority resurrects the very confusion Grable sought to correct. One
reading the majority opinion might understandably think that the cause of action analysis is
determinative to the jurisdictional issue. See, e.g., Majority Op. at 10 (relying on the absence of
a cause of action to say that hearing this claim in federal court would “disturb the balance of state
and federal responsibilities.”). If the majority’s analysis on this prong is taken too seriously and
applied to its fullest extent, that would effectively undercut over a century of Supreme Court

7California and Indiana also incorporate parts of the Garn-St. Germain Act, but they do not appear to
supply a cause of action.
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 27

precedent that recognizes federal question jurisdiction even in the absence of a federal cause of
action. But at the very least, the majority goes against Grable’s reading of Merrell Dow.8

Finally, even if all fifty states adopted a law like Michigan’s, the exercise of jurisdiction
over such laws would not amount to a federal court disturbing the federal-state balance. Again,
Congress wanted the Garn-St. Germain Act to displace any “laws (including the judicial
decisions) of any State to the contrary.” 12 US.C. § 1701j-3(b)(1). If there is a disturbance in
the federal-state balance, it is not the disturbance the majority is looking for. Majority Op. at 10–
11. That a federal court would throw the Act’s interpretation into the hands of the states, and
only the states, seems to undermine Congress’s expressed intent in the text of the Act. To be
sure, state courts are presumptively competent to adjudicate issues of federal law, Tafflin v.
Levitt, 493 U.S. 455, 458–59 (1990), including this one. But that fact settles nothing about
whether the federal courts have jurisdiction. The majority, however, seems to believe that the
Tafflin v. Levitt presumption weighs against finding federal question jurisdiction. Majority Op.
at 10 n.8, 11. If so, the majority is wrong. Ultimately, this final prong of the Grable standard
focuses on congressional intent. In this case, the federal courts exercising jurisdiction is, if
anything, consistent with Congress’s intent to ensure that state courts do not restrict the
enforcement of due-on-sale clauses. See 12 U.S.C. § 1701j-3(b)(1); Senate Rep. at 20–21; see
also OCC Interpretive Letter, supra footnote 1, at *1.9

8Furthermore, the majority’s analysis is inconsistent with the analysis of our own precedent, which reads
Merrell Dow as the Supreme Court did in Grable. See Hampton, 683 F.3d at 711–13 (analyzing the cause of action
issue and the substantial federal question issue separately).
9Whether or not “every schoolchild” actually learns that “states retain substantial sovereign authority of
their own property laws,” Majority Op. at 11 (internal quotations omitted), I am confident that every law student
learns that Congress can regulate interstate commerce and preempt state laws. Congress did just that with the Garn-
St. Germain Act. And as a result, the Plaintiffs’ first claim here turns entirely on federal law and has nothing to do
with state law or state sovereignty. If the majority believes there is “an unnecessary usurpation of state law
domain,” id. at 10, the majority’s real problem is with Congress and the fact that it enacted a statute that displaces
certain state property and contract laws.
 No. 18-1245 Estate of Robert Cornell, Jr., et al. v. Bayview Loan Servicing, et al. Page 28

IV. CONCLUSION

The difficulty of the issue of subject matter jurisdiction is matched by its importance.
Subject matter jurisdiction is so fundamental that a court can (indeed a court must) rule on the
matter sua sponte, without any motions by a party. Fed. R. Civ. P. 12(h)(3). We are the first
circuit court of appeals to address this issue as it relates to the Garn-St. Germain Act. We have
done so without the benefit of any briefing from the parties on the jurisdictional issue of whether
the federal question is substantial.

This case presents an exceptionally unique cause of action and federal question. But that
makes this case a prime candidate to “squeeze[] into the slim category” of state-law claims that
“arise under” federal law. See Empire, 547 U.S. at 701; 28 U.S.C. § 1331. Indeed, the category
is slim because the claims that fit into it are unique.

For the reasons expressed above, I respectfully dissent.