Stanley KOLBE, Plaintiff, Appellant, v. BAC HOME LOANS SERVICING, LP, d/b/a Bank of America, N.A.; Balboa Insurance Company, Defendants, Appellees, 738 F.3d 432

Opinion En Banc LYNCH, Chief Judge, with whom HOWARD, Circuit Judge, and KAYATTA, Circuit Judge, join. The judgment of dismissal entered by the district court is affirmed by an equally divided en banc court. See Savard v. Rhode Island, 338 F.3d 23, 25 (1st Cir. 2003) (en banc). Opinions follow. The result of the evenly divided vote of the en banc court is to affirm the district court’s dismissal of the complaint for failure to state a claim. See Savard v. Rhode Island, 338 F.3d 23, 25 (1st Cir.2003) (en banc).' This opinion explains why we think that result is correct and required by law. I.

This is a contract dispute over the terms of a mortgage contract between the borrower, plaintiff-appellant Stanley Kolbe, and the servicer of his loan, defendant-appellee BAC Home Loans Servicing, LP (“BAC” or “the Bank”). Kolbe sued the Bank in a putative class action for damages alleged to have arisen out of the Bank’s requirement that he maintain flood insurance in an amount sufficient to cover the replacement value of his home.

Kolbe contends that the Bank, under Covenant 4 of his mortgage contract, cannot require more than the federally mandated minimum flood insurance, which is the lesser of the principal balance of the loan or $250,000 in special flood hazard areas, and $0 in all other areas. The mortgage is insured by the Federal Housing Administration (“FHA”), and Covenant 4 is a standard uniform covenant prescribed by the FHA pursuant to federal law. See 24 C.F.R. § 203.17 (2012); Requirements for Single Family Mortgage Instruments, 54 Fed.Reg. 27,596, 27,603-07 (June 29, 1989) (hereinafter “Mortgage Requirements”).

The Covenant was promulgated after notice and comment rulemaking. We conclude that Kolbe has failed to state a claim for breach of contract. Three interrelated strands of reasoning support our conclusion. The first is straightforward application of the typical principles of contract

interpretation. When interpreting a written contract, we look at text, context, and purpose to discover whether a proffered reading of the contract is reasonable. For contract language mandated by a federal regulation, this context includes the regulation and the federal policy underlying the regulatory scheme. As a purely textual matter, the Bank offers the most natural reading of the disputed-language. Yet even if an argument exists that Kolbe’s textual reading is plausible, context confirms that the Bank’s reading is correct and Kolbe’s reading is incorrect.

As we will describe, particularly under our third strand of reasoning, Kolbe’s reading would hinder federal housing policy and conflict with other guidance from the federal government regarding flood insurance. Interpreting the text in context, as we would do with any contract, we conclude that the Bank’s reading is correct. Second, we apply special principles for interpreting uniform contract language. Covenant 4 is a uniform clause used in millions of mortgages nationwide by many different lenders, so we give it one uniform meaning rather than multiple inconsistent meanings.

Extrinsic evidence of the parties’ unique intentions regarding a uniform clause is generally uninformative because unlike individually tailored contracts, uniform clauses do not derive from the negotiations of the specific parties to a contract. Instead, courts seek to determine the uniform meaning of the clause as a matter of law, a task appropriate for the motion to dismiss stage. Kolbe cannot avoid dismissal on the grounds that his specific understanding or the actions of the parties create an ambiguity.

Third, the fact that the Covenant was drafted and mandated by the United States requires that its meaning be that meant by the United States when it drafted the regulation. The role that the Covenant plays in an important regulatory scheme requires that result. The language of the Covenant was not drafted or negotiated by the parties and was not the result of give-and-take in the marketplace. Rather, it was created and mandated in order to further important federal policies.

While on the Covenant’s plain language and context, we think the meaning is clear, were there doubt, we would defer to the position articulated to us by the United States in its amicus brief; in this case, the United States’ position reinforces our conclusion reached in applying the first two principles. In its amicus brief to the en banc court, the United States has stated that Kolbe’s

interpretation is incorrect for a number of reasons, including that it “lacks any anchor in the statutory scheme.” Brief for the United States as Amicus Curiae Supporting Appellees at 2, Kolbe v. BAC Home Loans Servicing, LP, No. 11-2030 [hereinafter “United States Brief’]. Further, the United States says that Kolbe’s

interpretation “serves no practical end, and ... would seriously undermine federal housing policy.” Id. The United States’ position as set forth in the brief is entitled to deference; it is well-reasoned and is entirely consistent with its prior

interpretations of the clause expressed in various federal publications. This is an issue for judges to decide. The law does not allow a jury to decide that federal policy is otherwise, or that the contract language required by the United States does not have the eminently reasonable meaning urged by the United States, consistent with the policies that brought about the Covenant in the first instance. As we will discuss, Kolbe has also failed to state a claim for breach of the covenant of good faith and fair dealing. The district court correctly dismissed all of Kolbe’s claims. II.

Kolbe owns a home in Atlantic City, New Jersey in a special flood hazard area. On October 6, 2008, he borrowed $197,437 from Taylor, Bean & Whitaker Mortgage Corp. (“Taylor Bean”) in a mortgage loan secured by his home. The loan was guaranteed by the FHA, a part of the Department of Housing and Urban Development (“HUD”). The mortgage agreement contained a set of Uniform Covenants that are required by HUD regulations to be in every FHA-insured mortgage. One of the Uniform Covenants included in the mortgage is the following provision, which is at issue: 4. Fire, Flood and Other Hazard Insurance.

Borrower shall insure all improvements on the Property, whether now in existence or subsequently erected, against any hazards, casualties, and contingencies, including fire, for which Lender requires insurance. This insurance shall be maintained in the amounts and for the periods that Lender requires. Borrower shall also insure all improvements on the Property, whether now in existence or subsequently erected, against loss by floods to the extent required by the Secretary. The “Secretary” referred to in Covenant 4 is the Secretary of HUD.

This case presents the issue of whether the amount of flood insurance required by HUD is a floor or a ceiling. Kolbe’s home is in an area designated by the Federal Emergency Management Agency (“FEMA”) as having “special flood hazards,” and as such HUD required (and still requires) that flood insurance must be maintained in “an amount at least equal to either the outstanding balance of the mortgage ... or the maximum amount of the NFIP insurance available with respect to the property improvements, whichever is less.” 24 C.F.R. § 203.16a(c) (emphasis added).

The original mortgage holder, Taylor Bean, never required Kolbe to maintain greater flood insurance than the minimum federally required amount, and at all times, Kolbe maintained flood insurance in excess of the outstanding loan balance. Taylor Bean declared bankruptcy and ceased operations in August 2009. At some point, the Bank became the servicer of Kolbe’s loan.

In November 2009, the Bank sent Kolbe a letter notifying him that it was requiring him to purchase an additional $46,000 in flood insurance coverage; the Bank has asserted, and Kolbe has not disputed, that this additional insurance would bring Kolbe’s total flood insurance coverage to the replacement cost of the property. Kolbe alleges that the Bank had a nationwide policy of requiring flood insurance at a level that often exceeds the principal balance of the loan.

The letter notified Kolbe that if he did not purchase the required flood insurance within about six weeks, the Bank would purchase the insurance at his expense and charge him for the cost, a practice known as “lender-placed insurance”; the letter urged Kolbe to avoid lender-placed insurance by purchasing his own insurance. A second letter reiterated the requirement. Kolbe purchased the insurance on his own; thus the Bank never had to purchase lender-placed insurance on his behalf. III.

On February 23, 2011, Kolbe filed a class action, complaint in the district court against the Bank alleging it breached the mortgage contract and violated the implied covenant of good faith and fair dealing by requiring the additional flood insurance. The first count of Kolbe’s complaint alleged breach of Covenant 4 of the mortgage contract. Under Kolbe’s theory, the Covenant precluded the Bank from requiring Kolbe to maintain any flood insurance in excess of the amount required by the Secretary of HUD, which in Kolbe’s case was the principal balance of the loan. See 24 C.F.R. § 203.16a(c).

The second count alleged a breach of the implied covenant of good faith and fair dealing. This count alleged that “[b]y requiring Plaintiff ... to maintain and pay for flood insurance coverage in excess of the coverage required by [his] mortgage agreement ], Defendants acted in bad faith and breached the implied covenant of good faith and fair dealing....” Kolbe sought to represent a putative class of all other borrowers with similar mortgages owned or serviced by the Bank who were required to purchase flood insurance above the amount of the outstanding balance of their loans.

Kolbe also sought a jury trial as to all claims. On August 18, 2011, the district court granted the Bank’s motion to dismiss all claims. The court concluded that the first two sentences in Covenant 4, which allowed the Bank to require insurance for “any hazards ... in the amounts-and for the periods that Lender requires,” unambiguously gave the Bank the right to choose the amount of flood insurance it required. Kolbe v. BAC Home Loans Servicing, L.P., No. 11-10312-NMG, 2011 WL 3665394, at *3-5 (D.Mass. Aug. 18, 2011).

The district court also dismissed the count for breach of the covenant of good faith and fair dealing because it concluded that the Bank’s flood insurance requirement was based on FEMA policy guidelines and was not unreasonable. Id. at *5. Kolbe appealed, and a divided panel of the First Circuit vacated the dismissal. See Kolbe v. BAC Home Loans Servicing, LP, 695 F.3d 111, 113-14 (1st Cir.2012). The panel majority held that both Kolbe’s

interpretation and the Bank’s'interpretation of the contract could be found reasonable by a trier of fact, and therefore that the district court erred in dismissing the breach of contract claim. Id. at 122. The panel majority also held that the breach of good faith claim could go forward either on the theory that the Bank intentionally breached the contract, or that the Bank demanded greater insurance based on the improper motivation of potential profit from placement of lender-placed insurance with affiliated companies. Id. at 123-24.

Judge Boudin dissented, arguing that the contract and federal policy plainly allowed the Bank to require more flood insurance and there was no independent claim under the implied covenant. Id. at 126-29 (Boudin, J., dissenting). We granted rehearing en banc, and vacated the panel’s decision. Order, Kolbe v. BAC Home Loans Servicing, LP, No. 11-2030 (1st Cir. Nov. 1, 2012). IV. To interpret Kolbe’s mortgage agreement, we start with the legal rules applicable to the construction of this particular contract language. Contract

Interpretation in Light of Context and Purpose In all contracts, courts must construe contract language in light of the purposes the language was meant to achieve, and in the context of the relevant commercial or regulatory schemes within which the contract is situated. See Simonson v. Z Cranbury Assocs.

P’ship, 149 N.J. 536, 695 A.2d 222, 224 (1997) (“[A] contract should not be construed literally so as to defeat the probable intention of the parties; rather, _ particular words or clauses may be qualified by the context and given the meaning that comports with the probable intention.” (internal quotation marks and citation omitted)); OneBeacon Ins. Co. v. Georgia-Pacific Corp., 474 F.3d 6, 7 (1st Cir.2007) (“The issue being one of contract

interpretation, we look to language and other common indicia (e.g., context, inferred purpose).”); Restatement (Second) of Contracts § 202 cmt. b (“The meaning of words ... commonly depends on their context.... When the parties have adopted a writing as a final expression of their agreement,

interpretation is directed to the meaning of that writing in the light of the circumstances.”). In particular, contract language must be interpreted in the context of applicable statutes and regulations. See 5 Corbin on Contracts § 24.26, at 271 (rev. ed.1998) (“Words and other symbols must always be interpreted in the light of the surrounding circumstances, and the existing statutes and rules of law are always among these circumstances.”). The typical principles of contract

interpretation are supplemented by two additional sets of rules of contract construction particularly relevant to Covenant 4: those for construction of uniform clauses, and those for construction of contract language drafted by the United States and required by federal law to be in the contract. Although these principles are applications of the general rule that contracts are interpreted in light of context, the methodology varies somewhat from that used when interpreting a contract with unique language negotiated by the two parties.

Uniform Clauses When a contract uses uniform language that is contained in a large number of contracts, as is the case here, it is a well-established common law principle of contract

interpretation that such contracts are “interpreted wherever reasonable as treating alike all those similarly situated, without regard to their knowledge or understanding of the standard terms of the writing.” Restatement (Second) of Contracts § 211(2). A variety of state and federal courts have acknowledged this principle. Because uniform contracts are interpreted uniformly across cases whenever it is reasonable to do so, extrinsic evidence about what a particular party intended or expected when signing the contract is generally irrelevant. See, e.g., Sharon Steel Corp. v.

Chase Manhattan Bank, N.A., 691 F.2d 1039, 1048 (2d Cir.1982) (“Boilerplate provisions are thus not the consequence of the relationship of particular borrowers and lenders and do not depend upon particularized intentions of the parties to an indenture.

There are no adjudicative facts relating to the parties to the litigation for a jury to find and the meaning of boilerplate provisions is, therefore, a matter of law rather than fact.”); 2 Farnsworth, Farnsworth on Contracts, § 7.11, at 304-OS (3d ed.2004) (“This rule plainly subordinates the meaning that an individual party may have attached to the contract language to the goal of equality of treatment for parties that are similarly situated.”). The issue of interpreting form contract language frequently arises in the context of class action certification.

Several federal courts have certified classes for contract disputes over form contracts because the form contracts are interpreted uniformly across members of the class, and thus the outcome does not depend on extrinsic evidence that would be different for each putative class member. See, e.g., Vedachalam v. Tata Consultancy Servs., Ltd., 18 Wage & Hour Cas.2d (BNA) 1677, 2012 WL 1110004, at *9 (N.D.Cal., Apr. 2, 2012) (“[I]n construing the form contract between Defendants and class members, the Court need not delve into the actual knowledge of individual class members.”); Peoples v.

Sebring Capital Corp., 52 Fed. R. Serv.3d 197, 2002 WL 406979, at *8 (N.D.Ill. Mar. 15, 2002) (“The court also rejects the broader notion that it will generally have to examine the parties’ intent on a transaction-by-transaction basis.”). It is undisputed that Covenant 4 is a Uniform Covenant required by HUD for all FHA-insured mortgages, according to a regulation that went into effect after notice and comment. Requirements for Single Family Mortgage Instruments, 53 Fed. Reg. 25,434 (July 6, 1988); see also Mortgage Requirements, 54 Fed.Reg. at 27,596 (final notice issued after receiving comments).

In essence, HUD’s regulation requires that every FHA-insured mortgage contain a core of Uniform Covenants, while allowing the parties to an individual mortgage to add non-uniform covenants at the end of the contract. For 'example, Kolbe’s mortgage contains about four pages of Uniform Covenants and one page of nonuniform covenants. That Kolbe’s mortgage contract contains uniform HUD covenants is apparent on its face.

After information about the address and location of Kolbe’s home, the third paragraph states, “THIS SECURITY INSTRUMENT combines uniform covenants for national use and non-uniform covenants with limited variations by jurisdiction to constitute a uniform security instrument covering real property.” The mortgage then reads, “UNIFORM' COVENANTS. Borrower and Lender covenant and agree as follows.” Following this heading are sixteen numbered covenants, including the disputed Covenant 4 and Covenant 7, which also has significance to this case.

These Uniform Covenants form the heart of the mortgage contract, covering such topics as principal and interest payments, insurance and taxes, care of the property, grounds for acceleration of debt, and the liability of co-signers. After the Uniform Covenants, the mortgage reads, “NON-UNIFORM COVENANTS. Borrower and Lender further covenant and agree as follows.” The mortgage then includes five non-uniform covenants.

The bottom left corner of every page of the contract contains the label in capital, boldface type: “NEW JERSEY FHA MORTGAGE.” Upon reading the mortgage, it would have been clear to Kol-be or any reasonable person that the mortgage contained nationwide Uniform Covenants, including Covenant 4. It also would have been clear that this was an FHA mortgage, such that federal policy and regulatory pronouncements would be relevant to its

interpretation. Language Drafted By The Government When dealing with uniform contract language imposed by the United States, it is the meaning of the United States that controls. In interpreting such a government mandated term, a court’s assessment of context and purpose is informed by the traditional tools of legislative and regulatory construction. This is a matter of law to be determined by a court.

When the United States mandates that private parties .use uniform language for a certain type of contract, the United States is enacting a policy that all parties to that type of contract should be subject to identical obligations. Those obligations are the ones the United States intended them to be, as determined by a court, regardless of the personal

interpretation offered by a party. If such contracts were subjected to different meanings depending merely on whether a particular party’s

interpretation was plausible, it would not only undermine the efficiency benefits of standardization, but it would also undermine the federal policy that motivated the United States to impose uniform contractual obligations on parties in the first place. This case demonstrates the necessity of these principles.

The disputed contract language is a Uniform Covenant required by federal law for the nearly 7.8 million FHA-insured mortgages nationwide; we therefore seek to find, to the extent reasonable, one uniform meaning, rather than separate meanings that might vary from lender to lender, or even from borrower to borrower.

As one commentator puts it, “if the specified provision is expressly included in the contract in' the exact terms required, the provision must be interpreted and given effect in accordance with the intention of the legislature, regardless of what the contracting parties may have understood it to mean.” 5 Corbin on Contracts § 24.26, at 278. Numerous federal and state courts, including the Supreme Court, have affirmed these principles. In Illinois Steel Co. v.

Baltimore & Ohio Railroad Co., 320 U.S. 508, 64 S.Ct. 322, 88 L.Ed. 259 (1944), the Supreme Court adjudicated a contract dispute involving a uniform bill of lading that had been imposed by the Interstate Commerce Commission. The Supreme Court noted that “[s]ince the clauses of the uniform bill of lading govern the rights of the parties to an interstate shipment and are prescribed by Congress and the Commission in the exercise of the commerce power, they have the force of federal law and questions as to their meaning arise under the laws and Constitution of the United States.” Id. at 511, 64 S.Ct. 322.

The Supreme Court then approached the issue as a question of regulatory construction, and decided the purpose and effect of the clause itself. See id. at 513-16, 64 S.Ct. 322. Similarly, in Honeywell, Inc. v. United States, 661 F.2d 182 (Ct.Cl.1981), the Court of Claims (the predecessor to the Federal Circuit) construed a federal procurement regulation that had been incorporated into a government contract. The court held that under the rules for “regulation

interpretation,” the agency’s

interpretation received “controlling weight”; the court rejected the notion that it should “construe[ ] [the language] in order to give it the effect intended by both parties.” Id. at 186. See also Saavedra v. Donovan, 700 F.2d 496, 499 (9th Cir.1983) (noting that when a federal regulation mandated contract terms, the contractual party “had a legal duty to conform to' the actual wage determination, not just a contractual duty to conform to plausible

interpretations of contract provisions embodying the wage determination”); Lloyd v. Cincinnati Checker Cab Co., 67 Ohio App. 89, 36 N.E.2d 67, 69 (1941) (“[S]uch statutory provisions [required to be in the contract] are read into the. bond or contract ‘regardless of the intention of the parties.’ The liability thus created is obviously, therefore, not a contractual liability involving a meeting of the minds, but a purely statutory obligation.”). These principles have also been adopted in New Jersey. See above, note 6. In Paul Revere Life Insurance Co. v.

Haas, 137 N.J. 190, 644 A.2d 1098 (1994), the Supreme Court of New Jersey interpreted an insurance contract with a provision required by state statute. The court rejected an argument that it should consider the understanding of the insured in interpreting the required provision; rather, the court stated, “A specific provision integrated into the contract by force of a statute, as a matter of public policy, must be interpreted and given effect in accordance with the intention of the legislature, irrespective of how the contractors understood it.” Id. at 1106 (quoting Saffore v. Atl. Cas. Ins.

Co., 21 N.J. 300, 121 A.2d 543, 548 (1956) (quoting 3 Corbin on Contracts § 551, at 200-01 (I960))) (internal quotation marks omitted). Although Haas dealt with a state statute, there is no reason the same principle would not apply with full force to a provision required by a federal regulation. This court therefore must examine the text of the Covenant in light of the purposes for which' the United States imposed the language and the context of the relevant regulatory scheme.

This is in keeping with the basic common law principle that contract language should be interpreted in light of purposes and context, applied to the particular circumstance of uniform contract language imposed by the United States. Such an inquiry is appropriate for the motion to dismiss stage because interpreting regulatory text in light of government purposes is a matter of law that is emphatically the province of judges, not juries. See Northshore Min. Co. v. Sec’y of Labor, 709 F.3d 706, 708 (8th Cir.2013) (“This dispute involves the

interpretation of MSHA regulations, a matter of law that we review de novo.”); Marine Polymer Techs., Inc. v. HemCon, Inc., 672 F.3d 1350, 1358 (Fed.Cir.2012) (“Statutory

interpretation is a matter of law that we consider de novo.”); cf. Marbury v. Madison, 5 U.S. (1 Cranch) 137, 177, 2 L.Ed. 60 (1803) (“It is emphatically the province and duty of the judicial department to say what the law is.”); Diederich v. American News Co., 128 F.2d 144, 146 (10th Cir.1942) (“The power of the judge to pass upon questions of law is just as much an essential part of the process of trial by jury ... as is the power of the jury to pass upon questions of fact.”). . V. With these principles in mind, we turn to the Covenant at issue.

In performing our task of determining the uniform meaning of the Covenant as a matter .of law, we first examine the text in light of its context, then look to the United States’

interpretation. We repeat the language of Covenant 4 for convenience, dividing it into its three sentences:, 4. Fire, Flood and Other Hazard Insurance.

(1) Borrower shall insure all improvements on the Property, whether now in existence or subsequently erected, against any hazards, casualties, and contingencies, including fire, for which Lender requires insurance.

(2) This insurance shall be maintained in the amounts and for the periods that Lender requires.

(3) Borrower shall also insure all improvements on the Property, whether now in existence or subsequently erected, against loss by floods to the extent required by the Secretary [of HUD]. The Bank argues that in allowing the lender to require its chosen amount of insurance for “any hazards,” the first two sentences clearly give the Bank the authority to choose the required amount of flood insurance. Kolbe argues that the only provision addressing flood insurance in Covenant 4 is the third sentence, and thus the Bank could not require more flood insurance than the amount required by HUD, which (in Kolbe’s case) was the principal loan balance. We agree with the contract

interpretation offered by Judge Boudin in his panel dissent. We adopt and incorporate Judge Boudin’s reasoning and expand. See Kolbe, 695 F.3d at 127-29 (Boudin, J., dissenting). The Bank offers the only plausible reading of the uniform text, against the context. As we discuss later, this reading is confirmed by the intent of the United States. Simply put, the first two sentences allow the Bank to choose the amount of insurance for “any hazards,” and that includes flood insurance because floods are hazards.' Dictionary

definitions confirm the common understanding that floods are hazards, and even the panel majority acknowledged that “[fjloods unquestionably are a type of hazard, and they are thus literally within the scope of the first sentence.” Kolbe, 695 F.3d at 117 (majority opinion). Although the third sentence also addresses flood insurance, it adds an independent requirement: that the borrower maintain HUD’s minimum level of flood insurance in addition to the lender’s minimum.

Because both HUD’s and the lender’s flood insurance requirements are minimum requirements, they are perfectly consistent, and the borrower can meet both requirements by maintaining flood insurance in the amount of the higher requirement. Contrary to Kolbe’s arguments, there is no need to read the first two sentences to exclude floods in order to avoid making any provision superfluous, or to resolve a conflict between a specific provision and a general provision under principles of contract

interpretation. The Bank’s

interpretation is also more consistent with another covenant of the contract, Covenant 7, as we explain in another section- below. This Covenant empowers the lender to purchase insurance to “protect the value of the Property,” suggesting that the lender’s economic interests are not limited to the principal balance of the loan. Kolbe also argues that the title of Covenant 4 — “Fire, Flood and Other Hazard Insurance” — supports his reading. Kolbe argues that the title signifies that the paragraph deals separately with fire insurance and flood insurance.

Because the first sentence refers to “hazards ... including fire,” but does not mention floods, while the third sentence singles out flood insurance, Kolbe concludes that only the first sentence deals with fire insurance and only the third sentence deals with flood insurance. This argument is a non sequitur. The first sentence covers “any hazards, ... including fire” (emphasis added). In the context of a sentence covering “any hazards,” the listing of fire as an example clearly does not imply an exclusion of other hazards.

It would be unnatural and illogical to read “any hazards, ... including fire” to mean “all hazards except floods.” The government flood insurance requirement is mentioned separately in the final sentence to comply with National Flood Insurance Act (NFIA) and .HUD legal requirements regarding flood insurance. See 42 U.S.C. § 4104a(a)(3); 24 C.F.R. § 203.16a(a)(2). The title merely reflects that flood and fire are two kinds of hazards that are specifically mentioned in the Covenant. If anything, the phrasing of the title supports the Bank.

By using the phrase “Other Hazard Insurance” after listing fire and flood, the title says that both fire and flood are instances of hazards, which leads to the conclusion that flood insurance is included in the first sentence. We conclude that the Bank’s reading of the text, is the only plausible reading in the relevant context. For contract language mandated by a federal regulation that implicates the federal mortgage insurance and flood insurance programs, this context includes the broader regulatory schemes and the federal policy underlying those schemes.

In essence, when this covenant is understood in context against its purposes and federal housing policy, the only reasonable

interpretation of this language is that offered by the Bank. Am examination of the context removes any claim of ambiguity. Covenant 4 traces its origins to a HUD regulation that bears directly on the question at hand. The regulation is titled “Mortgagor and mortgagee requirement for maintaining flood insurance.” 24 C.F.R. § 203.16a. In pertinent part, that regulation states that both the mortgagee and the mortgagor must “obtain and ... maintain NFIP flood insurance coverage on the property improvements during such time as the mortgage is insured.” Id. § 203.16a(a)(2).

As to the amounts, the regulation states: “The flood insurance must be maintained such time as the mortgage is insured in an amount at least equal to either the outstanding balance of the mortgage, less estimated land costs, or the maximum amount of the NFIP insurance available with respect to the property improvements, whichever is less.” Id. § 203.16a(c) (Emphasis added.). And there is good reason why HUD required lenders and borrowers to “maintain” flood insurance in “at least” certain amounts, and not in “no more” than certain amounts, as Kolbe would have it.

As the United States said at oral argument: And there are good reasons for that. The first is that in a normal case [the] borrower defaults, the bank forecloses on the property — assigns it to HUD and then walks away. And HUD pays them the insurance proceeds of the mortgage insurance. Then HUD is responsible for selling the property and reimburses the mortgage insurance fund with the proceeds of the sale. But of course if the house has been destroyed by a flood — there is nothing for HUD to sell.

And so there is no way to reimburse the mortgage insurance fund and that is why HUD regulations have specifically provided since 1971, that flood damage has to be repaired by the lender before the property can be re-conveyed. In its brief, the United States also explains the unreasonable consequences that would result from Kolbe’s reading. In response, Kolbe argues that federal policy supports his

interpretation. He also argues that the position of the United States articulated in the brief is entitled to no deference because (

a) it is stated in an amicus brief, and (

b) in his view, it is inconsistent with the position the United States took earlier. Before addressing the policy arguments, we provide background on the relevant regulatory schemes to explain the arguments and our conclusion. Federal Flood Insurance and Housing Policy Two federal statutory and regulatory schemes factor into this case: the National Flood Insurance Act (“NFIA”) and the FHA’s mortgage insurance program.

In 1968, Congress passed the NFIA, 42 U.S.C. §§ 4001-4129, to make flood insurance available and to promote the use of flood insurance in areas of the country with flood risk, see id. § 4002(b) (declaration of congressional purpose). Congress found that floods caused substantial economic and personal hardships, but that it was not economical for private insurance companies to provide flood insurance. Id. § 4001(a),(b). To remedy the situation, Congress authorized a program in which the United States would partner with private insurance companies to provide flood insurance. Id. § 4001(b)-(d).

Under the National Flood Insurance Program (NFIP), the United States makes flood insurance available in states and communities that agree to participate in the program. 42 U.S.C. § 4012(c). In flood-prone areas (i.e., those deemed “areas having special flood hazards” by FEMA) where flood insurance is available, the NFIA requires federally regulated lenders hot to make mortgage loans unless the borrower obtains flood insurance at least up to the full principal balance of the loan (or in the maximum amount available, if that is less). Id. § 4012a(b)(l).

In addition, federal financial assistance for homes in special flood hazard areas is forbidden unless the home is covered by flood insurance at least equal to the lesser of the loan balance or the maximum amount available. Id. § 4012a(a). Although the insurance is provided by private insurers to the extent possible, id. § 4011(c), the United States supports the program by offering subsidy payments and reinsurance to the private insurers, id. § 4054, 4055. The FHA was created in 1935 as a result of the National Housing'Act of 1934, 12 U.S.C. § Í701 et seq.

The FHA promotes affordable home ownership by providing mortgage insurance to private lenders, cf. id. § 1709; Mission/U.S. Department of Housing and Urban Development (HUD), http://portal.hud.gov/ hudportal/HUD?src=/about/mission (last visited May 16, 2013) (mission statement of HUD to “create strong, sustainable, inclusive communities and quality affordable homes for all”). If a borrower defaults on an FHA-insured mortgage, the lender can convey the mortgage or title to the property to HUD and collect insurance benefits from the United States to compensate for any losses on the mortgage. See 12 U.S.C. § 1710.

However, if the property has suffered damage from “fire, flood, earthquake, hurricane, or tornado,” then the lender cannot collect insurance benefits from the United States unless it has repaired the damage or taken a deduction from the insurance benefits for the cost of repairing the damage. 24 C.F.R. § 203.379 (emphasis added). Effectively, this scheme allocates the risk of most defaults on FHA-insured mortgages to the United States, but it allocates the risk of certain hazard losses (including flood losses) to the lender.

Policy Arguments Given this background and context, it is not surprising that the United States is able to confirm that HUD has “never endorsed such a policy” of construing Covenant 4 as “a federal ceiling for flood insurance coverage rather than a floor.” The United States explains that Kolbe’s reading conflicts with the overall structure of FHA mortgage insurance. HUD’s mortgage insurance program places the risk of flood and other hazard losses on the lender, see 24 C.F.R. § 203.379, and so gives the lender the authority to determine the amount of flood insurance necessary to protect its, investment.

As the -United States describes, “[tjhat is the purpose of Paragraph 4: because the lender ultimately bears the risk of uninsured hazard losses, FHA’s standard mortgage contract allows the lender to specify the types and amounts of all hazard insurance — including flood insurance — that the borrower must carry.” United States Brief at 15. In addition, Kolbe’s

interpretation of Covenant 4 would lead to anomalous and untoward results. Under Kolbe’s reading of Covenant 4, the only sentence addressing flood insurance is the third sentence, which obligates the borrower to maintain insurance in the amount required by the Secretary of HUD. But HUD only requires flood insurance in special flood hazard areas. Thus, under Kolbe’s reading, a lender could not require a penny of flood insurance for homes in moderate flood risk areas.

Special flood hazard areas are defined as areas subject to at least a one percent chance of flooding in any given year, which equates to a twenty-six percent chance of flooding over the course of a thirty year mortgage. Homes in moderate flood risk zones, while falling short of the risk threshold for a special flood hazard area, may nonetheless face significant flood risk. In fact, over twenty percent of NFIP flood-insurance claims and about one third of federal disaster relief payments for flooding are related to properties outside of special flood hazard areas.

National Flood Insurance Program, Flood Facts, http://www.floodsmart.gov/ floodsmart/pages/flood-jfacts.jsp. There would be no reason to forbid the lender from requiring any flood insurance on such homes, yet allow the lender to require as much insurance as it wishes for other hazards that are extremely unlikely to occur, such as earthquakes or tornados in certain parts of the country. Such an irrational policy objective could not plausibly be attributed to HUD, and the United States’ brief confirms that HUD did not intend such a result.

The result urged by Kolbe would seriously impair federal housing policy as articulated by the United States. Kolbe’s

interpretation would prevent lenders in some cases from requiring adequate flood insurance, particularly for homeowners with mortgages above $250,000 (the maximum federal requirement) or homes outside of special flood hazard areas, where the United States does not require any flood insurance. United States Brief at 21-22. Kolbe’s

interpretation would not only frustrate HUD policy, but it “is impossible to reconcile with Congress’s objective in the [NFIA], which was-not to prohibit the use of flood insurance in federally insured housing but to encourage it.” Id. at 24. The United States finds it foreseeable that lenders would react to Kolbe’s

interpretation by “declinfingj to offer FHA-insured loans in areas facing even marginal flood risks, or charging] substantially greater interest rates for such loans,” thus hindering affordable home ownership. Id. at 24. Kolbe and supporting amici posit that federal housing policy could support their contract reading. These policy arguments revolve around the fact that a primary purpose of HUD and the FHA is to promote affordable home ownership.

Because flood insurance can be expensive, a provision limiting the lender’s ability to require flood insurance could reduce one component of the initial cost of home ownership for FHA borrowers. This argument that the policy of lowering housing costs supports Kolbe’s

interpretation is anchored in speculation rather than the record of the Covenant’s actual context and purpose. Moreover, its economic assumptions do not bear scrutiny. Restricting the amount of flood insurance only reduces the buyer’s monthly payment if the lender, so restricted, fails to factor the increased risk into the interest rate charged. Kolbe also ignores the fact that the purchase of flood insurance results in either an increase in home ownership costs (in the event of no flood) or a decrease in home ownership costs (in the event of a flood).

And Kolbe offers no evidence that the FHA somehow considered the risk-adjusted balance of the effects on costs to be detrimental to consumers. In short, the notion that the FHA wanted to make sure that consumers could under-insure for flood loss is complete and improbable speculation. And this

interpretation by the United States was provided before Kolbe entered into his mortgage with Taylor Bean, as discussed below. Kolbe further dismisses the United States’ brief as a “newly minted

interpretation [that] is flatly inconsistent” with past HUD practice. This is simply not so. Earlier HUD pronouncements support the United States’ present assertions and the Bank’s

interpretation, and are inconsistent with Kolbe’s

interpretation. Kolbe’s inconsistency argument is largely based on HUD handbooks and guidance documents that list “flood insurance” and “hazard insurance” as separate categories. Kolbe argues that these documents show that HUD has long treated hazard insurance and flood insurance separately, reflecting a broader industry practice of excluding flood coverage from hazard insurance policies. Because HUD and industry practice treat hazard insurance and flood insurance as separate categories, Kolbe asserts that the mention of hazards in the first sentence should be read to exclude floods.

The panel majority found this separation to be significant. It is, but the difference reinforces the Bank’s reading. Kolbe’s argument confuses the question at issue. The question is not whether the category of “hazard insurance” includes “flood insurance”; the question is whether floods are hazards, and thus whether a reference in Covenant 4 to “any hazards” includes floods. On this question, both HUD practice and the pattern of industry usage favor the Bank and the United States’

interpretation, not Kolbe’s. We explain why. In the middle of the twentieth century, insurance companies began issuing comprehensive hazard insurance policies that covered against a wide variety of risks. Crusto, The Katrina Fund: Repairing Breaches in Gulf Coast Insurance Levees, 43 Harv. J. on Legis. 329, 334 (2006). These comprehensive hazard insurance policies consist of “named peril” policies that only cover an enumerated list of hazards, and “all-risk” policies that cover all physical hazards except those specifically excluded.

Thomas & Randall, New Appleman on Insurance Law § 41.02[l][a], at 41-15 (library ed.2011). More recently, all-risk policies have eclipsed named peril policies as the most common form of homeowners insurance. Id. § 42.02[1], at 42-60. Yet virtually all standard hazard insurance policies, including all-risk policies, contain a specific “flood exclusion” provision that excludes flooding and water damage from coverage. Id. § 43.02[3][a], at 43-14.

The fact that HUD documents list “flood insurance” and “hazard insurance” as separate categories reflects the reality that homeowners who want flood insurance will need to purchase it separately from an all-risk hazard insurance policy. It does not support an inference that HUD is stating that floods are not hazards; rather, it is stating the opposite. The reason that such an express flood exclusion is necessary in a hazard insurance policy covering all risks is because flooding is considered a risk (or alternatively, a hazard), and thus would be covered by the hazard insurance policy absent such an exclusion.

HUD regulations and the NFIA confirm the industry understanding that floods are hazards. For example, HUD requires flood insurance on FHA-insured mortgages in “area[s] having special flood hazards.” 24 C.F.R. § 203.16a(b); see also 42 U.S.C. § 4012a(a) (mandating that federally regulated lenders require flood insurance on homes in “area[s] having special flood hazards”). Other HUD pronouncements, including a different part of the 1994 HUD Handbook cited by Kolbe, also support the United States’

interpretation, but contradict Kolbe’s

interpretation. As we have noted, under Kolbe’s

interpretation, a lender cannot require any more flood insurance than what HUD requires, which would mean zero flood insurance outside of special flood hazard areas. Yet HUD has been quite clear on multiple occasions that lenders can require flood insurance outside of special flood hazard areas.

For example, in a 1990 letter to mortgagees of FHA-insured loans, the FHA Commissioner wrote, “Menders are free to consider requiring flood insurance in participating communities on the basis of their own business judgement, even if the building that is the security for a loan is located outside of an SFHA [special flood hazard area].” Mortgage Letter 90-16, 1990 WL 10022448, at *2.

A handbook issued by HUD in 1994 states, “In areas designated B and C (with suffixes) [on FEMA maps], [flood] insurance is available but not required by HUD (although mortgagees may require it under the same terms and conditions as those that apply to other dwelling insurance).” HUD Handbook 4330.1, 2-ll(E)(2) (emphasis added). Quite significantly, FEMA recommended in its 2007 guidelines that lenders do precisely what the Bank did: require homeowners in special flood hazard areas to maintain replacement cost flood insurance.

See FEMA, National Flood Insurance Program: Mandatory Purchase of Flood Insurance Guidelines 27 (2007). We will not read HUD regulations as preventing lenders from following FEMA flood insurance guidelines with respect to FHA-insured mortgages. See Mortgage Letter 90-16, 1990 WL 10022448, at *1 (“[W]e want to bring HUD policy in conformance with that of FEMA.”). Kolbe raised another example of purported inconsistency at oral argument. Kolbe notes that the 1994 HUD Handbook states that a lender “may not insist on more [insurance] coverage than is necessary to protect its investment.” HUD Handbook 4330.1, 2-ll(B).

He argues that only insurance in the amount of the principal loan balance is necessary to protect the lender’s investment; thus, this handbook limited the lender’s discretion and thereby conflicts with the conclusion of the United States’ brief. Kolbe’s argument fails for several reasons, including that its factual premise is untrue. First, his argument conflicts with Covenant 7 of the mortgage contract.

Covenant 7 is the force-pay provision that’ not only allows the lender to protect itself when the borrower fails to comply with his obligations, including those under Covenant 4, but also allows the lender to charge the borrower for the resulting cost incurred by virtue of the borrower’s breach. It provides that “[i]f Borrower ... fails to perform any ... covenants and agreements contained in this Security Instrument, ... then Lender may do and pay whatever is necessary to protect value of the Property and Lender’s rights in the Property, including payment of taxes, hazard insurance....

Any amounts disbursed by Lender under this paragraph shall become an additional debt of Borrower.... ” The two covenants must therefore be read together in a manner that aligns duty, breach, and remedy. That alignment appears perfectly and plainly if Covenant 4 is read, as we read it, to allow the lender to require the borrower to procure flood insurance up to an amount necessary to protect the value of the property.

Conversely, under Kolbe’s view, the Covenant 4 duty is only to buy flood insurance in amounts that will often be far less than that necessary to protect the value of the property, but the remedy for a breach of that duty, under the plain language of Covenant 7, is that the borrower may be required to reimburse the lender for the cost of flood insurance for the full amount necessary to protect the value of the property.

Second, FEMA’s guidelines confirm the fact, described by the United States in its amicus brief, that the lender has an economic interest in the borrower maintaining replacement cost flood insurance. Finally, it is a matter of common sense that a lender has an interest not only in the principal balance of the loan but in maintaining a performing loan that will provide a stream of interest payments; if the borrower has enough insurance to rebuild his home in the event of a flood, it is more likely that the borrower will remain current on the loan and continue to make payments.

We again explain why three strands of reasoning support our conclusion. Using the ordinary tools of contract

interpretation, we view the text of Covenant 4 in the context of federal housing policy. This examination convinces us that the Bank’s

interpretation is correct. Because this covenant is a uniform clause, we determine its uniform meaning as a matter of law, and do not allow Kolbe to vary from that meaning on the basis of extrinsic evidence unique to his transaction. This leads into the third strand: the fact that this- language wás drafted and imposed by the United States in a regulation. On this record, we think the Covenant’s purpose is plain. We have no doubts about the meaning of Covenant 4 under any of the three tests, but if we did, we would resolve those doubts by deferring to the United States’ reasonable

interpretation. See Auer v. Robbins, 519 U.S. 452, 461, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997). Under the doctrine of “Auer deference,” we accept an agency’s

interpretation of its own regulation “unless ‘plainly erroneous or inconsistent with the regulation.’ ” Id. (quoting Robertson v. Methow Valley Citizens Council, 490 U.S. 332, 359, 109 S.Ct. 1835, 104 L.Ed.2d 351 (1989)). Although Covenant 4 appears in a contract between private parties, it derives from a duly enacted HUD regulation, in which HUD promulgated the language and mandated that private parties include the. language in mortgage contracts for FHA-insured mortgages. See Mortgage Requirements, 54 Fed.Reg. at 27,603-07. Auer deference applies here just as it does to any other agency

interpretation of a regulation. Indeed, multiple courts of appeals have accorded deference to agency

interpretations of contract terms that were promulgated and mandated by a federal regulation. See Saavedra v. Donovan, 700 F.2d 496, 499 (9th Cir.1983) (according “deference to an agency’s reasonable and conforming

interpretation of its own regulation”); Honeywell Inc. v. United States, 661 F.2d 182, 185 (Ct.Cl.1981) (“['I]n construing administrative regulations, the ultimate criterion is the administrative

interpretation, which becomes of controlling weight unless it is plainly erroneous or inconsistent with the regulation.... The fact that a regulation may be incorporated into a contract does not require a different rule for regulation

interpretation.”). Applying Auer deference, it is a simple matter to uphold the United States’

interpretation of Covenant 4, which accords with the Bank’s

interpretation. Far from being “plainly erroneous or inconsistent with the regulation,” the United States’

interpretation is consistent with the most natural reading of the regulation’s text. Further, it is supported by persuasive articulations of federal policy as discussed earlier and contained in the United States’ brief. Kolbe insists that Auer deference is inappropriate, citing to Christopher v. SmithKline Beecham Corp., — U.S.-, 132 S.Ct. 2156, 183 L.Ed.2d 153 (2012), a case in which the Supreme Court rejected and refused to extend Auer deference to a United States brief that was inconsistent with past agency practice and the governing statute.

The agency in Christopher submitted a brief declaring an industry practice illegal, but the Court noted that this brief was inconsistent with decades of declining to bring enforcement actions, which created a justified expectation that the practice did not violate the relevant regulations. See id. at 2167-68. This case is distinguishable from Christopher. Nothing in HUD’s past practice is inconsistent with the position articulated in its brief.

To the contrary, HUD has de-dared in the past that lenders can require flood insurance above HUD requirements outside of special flood hazard areas, which supports the position in its brief but is inconsistent with Kolbe’s position. Christopher provides no support for rejecting Auer deference in this case. We stress that Auer deference is not necessary to our conclusion. Even if Kol-be were correct that Christopher governs this case, he would still lose. In Christopher, while rejecting Auer deference, the Court granted the agency a lesser measure of deference derived from Skidmore v.

Swift & Co., 328 U.S. 134, 140, 65 S.Ct. 161, 89 L.Ed. 124 (1944): “deference proportional to the ‘thoroughness evident in [the agency’s] consideration, the validity of its reasoning, its consistency with earlier and later pronouncements, and all those factors which give it power to persuade.’ ” Christopher, 132 S.Ct. at 2169 (quoting United States v. Mead Corp., 533 U.S. 218, 228, 121 S.Ct. 2164, 150 L.Ed.2d 292 (2001)). Here, the. United States’ brief contained thorough consideration and valid reasoning, was consistent with other HUD pronouncements, and was persuasive of its own force.

The lesser Skidmore deference easily would have sufficed to sustain its

interpretation. Indeed, we would agree with the United States’

interpretation even if we gave it no deference at all. Kolbe has failed to state a claim for breach of contract. VI. The claim for breach of the covenant of good faith and fair dealing also fails. In every contract, there exists an implied covenant of good faith and fair dealing. Kalogeras v. 239 Broad Ave., L.L.C., 202 N.J. 349, 997 A.2d 943, 953 (2010). Under this covenant, “neither party shall do anything which will have the effect of destroying or injuring the right of the other party to receive the fruits of the contract.” Id. (quoting Palisades Props., Inc. v.

Brunetti, 44 N.J. 117, 207 A.2d 522, 531 (1965)) (internal quotation marks omitted). In addition, where a contract grants a party discretion, the party must exercise that discretion reasonably. Wilson v. Amerada Hess Corp., 168 N.J. 236, 773 A.2d 1121, 1130 (2001).

Kolbe’s complaint contains only a single allegation that the Bank breached the implied covenant: “By requiring Plaintiff and the Class to maintain and pay for flood insurance coverage in excess of the coverage required by their mortgage agreements, Defendants acted in bad faith and breached the implied covenant of good faith and fair dealing contained in the mortgage agreements.” This allegation is wholly dependent on the premise that the Bank breached the contract, and it therefore fails with the failure of the breach of contract claim.

By failing to allege it in his complaint, Kolbe has waived any other claim regarding the covenant of good faith. Even if we were to assume in Kolbe’s favor that he preserved this argument, raised for the first time on appeal, that “the only reason Defendants demanded additional flood insurance was an improper effort to self-deal ... collecting for itself or its affiliates insurance brokerage commissions and excessive premiums,” it fails. Kolbe’s self-dealing claim fails the standard of plausibility necessary to survive a motion to dismiss. Ashcroft v.

Iqbal, 556 U.S. 662, 678, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009). Kolbe’s allegations do not support a plausible inference that he personally has suffered any injury or that the Bank has abused him. The Bank sent Kolbe a letter in which it urged him to purchase his own insurance. This letter gave Kolbe about six weeks notice to purchase his own insurance.

It warned Kolbe of the potential negative consequences of lender-placed insurance, stating that the insurance “may be more expensive and will likely provide less coverage than was previously in effect or that you can obtain on your own,” as well as mentioning the potential commissions.

The letter implored him to purchase his own insurance: “We encourage you to act now and obtain flood insurance in the necessary amounts to avoid incurring the cost of our buying Lender-Placed Insurance.” The Bank followed up a month later with a second letter, again notifying Kolbe of the insurance requirement and stating that he could avoid lender-placed insurance by purchasing his own insurance. The Bank’s disclosure and warning hardly support a claim of abusive self-dealing. Of course, Kolbe did purchase his own insurance, presumably at a fair market rate.

The Bank did not force-place any insurance, and thus did not collect any commissions or premiums from Kolbe. Kolbe did not suffer any harm; the only “injury” he claims to have suffered is from the cost of obtaining his own insurance. But, as we have said, the requirement that he do so was legal, and so there was no injury. He may not raise a claim, apparently on behalf of others, that affiliates of the Bank collected and profited from commissions or premiums on lender-placed insurance.

Further, even as to that issue, Kolbe’s complaint makes no allegation that plausibly suggests that his lender required he obtain additional flood insurance beyond an amount necessary to protect the lender’s legitimate interests, or that it required him to purchase anything at all from the lender or anyone associated with the lender. To the contrary, the very letters to which Kolbe points in his complaint make clear that, in requiring the additional coverage, the lender urged Kol-be-twice-to obtain the insurance on his own from someone other than the lender.

In short, taking Kolbe’s allegations on then-face, they fail to make out any claim for a breach of the lender’s contractual commitments, express or implied. Kolbe and supporting amici have attempted to turn this case into a broader hearing on alleged abuses in the practice of lender-placed insurance. That is a separate problem, and one independent of the clause we have construed.

Accepting Kolbe’s allegations as trae, he ended up with more insurance than he would have chosen to purchase on his own, but he unquestionably received value for the additional cost: sufficient insurance to rebuild his home in the event of a flood. We take judicial notice that the Atlantic Coast suffered a major flood last fall from Hurricane Sandy; the damage was so significant that Congress appropriated $9.7 billion to replenish the NFIP’s insurance fund, and an additional $51 billion to aid storm victims. Final Passage by Congress to $51 Billion in Storm Aid, N.Y.

Times, Jan. 29, 2013, at A21; Congress Approves $9.7 Billion in Insurance Aid for Hurricane Victims, N.Y. Times, Jan. 5, 2013, at A14. Kolbe’s hometown of Atlantic City sustained significant damage. Empty of Gamblers and Full of Water, Atlantic City Reels, N.Y. Times, Oct. 30, 2012, at Al. This event served as a sad reminder of the value of replacement cost flood insurance for homeowners, particularly in flood-prone areas. Further, the Bank did not act unreasonably in requiring this insurance.

The Bank was following FEMA’s guidance, and as discussed above, the increased insurance protected the Bank’s reasonable and legitimate economic interests. Kolbe’s complaint fails to state a claim for relief, and the district court correctly granted the motion to dismiss. VII. This opinion does not attempt to respond to the opposing opinion written by Judge Lipez for himself and two of our colleagues. Rather, the opinion of Judge Kayatta does respond, and I join him. LIPEZ, Circuit Judge, with whom TORRUELLA, Circuit Judge, and THOMPSON, Circuit Judge, join.

Appellant Stanley Kolbe claims that he and his mortgage lender agreed in 2008 that his obligation to buy flood insurance was capped at the amount of his outstanding principal balance, consistent with their common understanding of a uniform covenant included in all mortgages insured by the Federal Housing Administration (“FHA”). Five years later, in an amicus brief submitted to the en banc court in this case, the federal government announced for the first time that the covenant at issue must be read to give lenders the discretion to increase the flood insurance requirement at any time.

Our colleagues conclude that the government’s

interpretation retroactively nullifies the bargain allegedly struck by Kolbe and his lender, even though that agreement rested on a reasonable construction of the provision and, importantly, is consistent with federal law. Our colleagues’ judgment constitutes extraordinary intervention into the contractual dealings of two private parties.

In effect, they conclude that a federal agency, through coxirt intervention, may rewrite an agreement even though the agency is not a party to the deal, and has no role in its enforcement, simply because a different agreement would better serve the government’s newly clarified priorities. There is no justification for such a wholesale abandonment of common law contract princi-pies by our colleagues.

Regrettably, the even division of views on the en banc court means that the decision of the district court dismissing Kolbe’s breach of contract and good faith and fair dealing claims against Bank of America will be reinstated. At bottom, this is a straightforward motion-to-dismiss case. Kolbe asserts that he and his original lender, Taylor, Bean & Whitaker Mortgage Corp. (“Taylor Bean”), agreed that Kolbe was required to maintain only the statutory minimum amount of flood insurance on his property throughout the duration of his mortgage.

Taking that factual assertion as true, as we must, Bank of America’s demand that Kolbe increase his flood coverage provides the basis for a plausible claim that the Bank committed a breach of contract. Kolbe likewise alleges facts that would permit a jury to find that the Bank made its demand to serve its own financial interests, in violation of the implied covenant of good faith and fair dealing. Whatever the ultimate resolution of his contentions, Kolbe has done enough to defeat the Bank’s motion to' dismiss and, hence, is entitled to move forward with his case, including discovery.

As we shall explain, denying him that opportunity is indefensible. I. We will not here reprise the textual analysis of Paragraph 4 of Kolbe’s mortgage that appears in the majority panel opinion previously vacated by the en banc court. See App’x. The discussion there makes plain that the provision is ambiguous, and its lack of clarity is underscored by the lack of consensus in the decisions of other courts. Compare, e.g., Morris v. Wells Fargo Bank N.A., No. 2:11-cv-00474, 2012 WL 3929805, at *7-8 (W.D.Pa.

Sept. 7, 2012) (denying motion to dismiss breach of contract claim involving same language) (stating that, “[a]t the very least, plaintiffs

interpretation is tenable”), and Wulf v. Bank of Am., 798 F.Supp.2d 586, 588 (E.D.Pa.2011) (same), with, e.g., McKenzie v. Wells Fargo Home Mortg., Inc., No. 3:11-cv-04965-JCS, 2012 WL 5372120, at **13-15 (N.D.Cal. Oct. 30, 2012) (adopting reasoning of Kolbe panel dissent).

Our colleagues, who acknowledge the “set of sharply conflicting district court opinions” on this issue across the country, nonetheless reject the notion of ambiguity and assert that appellee Bank of America offers “the only plausible reading in the relevant context.” The problem is- that this conclusion of non-ambiguity is procured by means of hindsight, with substantial weight given to the government’s newly announced view and the policies offered to justify it. We do not minimize the importance of the government’s

interpretation of Paragraph 4. We acknowledge that the additional context revealed in the government’s amicus brief sheds helpful light on the meaning that was intended when the covenant was crafted by the Department of Housing and Urban Development in the 1980s. Our disagreement is not with our colleagues’ lengthy exegesis on the nature of contract construction as applied to uniform clauses and language drafted by the United States. Rather, those principles are inapplicable to the specific issue before us.

Our colleagues insist that what the government says with clarity in 2013 overrides the meaning that Kolbe and his lender ascribed to Paragraph 4 five years earlier. Yet the government’s effort now to dispel the confusion generated by its poorly drafted language cannot erase the ambiguity that confronted Kolbe and his lender when they signed their mortgage agreement.

As the panel majority explained, the textual signals in Paragraph 4 point in both directions, and there is nothing implausible about a federal agency charged with promoting home ownership choosing to adopt Congress’s benchmark for flood insurance coverage as a way to balance affordability and risk avoidance. The government’s clarification does not magically eliminate the mixed message communicated by the language and structure of Paragraph 4.

It therefore does not change the fact that Kolbe and his lender could have agreed that the phrase requiring flood insurance “to the extent required by the Secretary” fixed Kolbe’s obligation at the statutory minimum for the duration of the loan. Hence, the pivotal issue — the one that sharply separates our view from that of our colleagues — is not whether the text of Paragraph 4 is ambiguous, but whether the government’s belated clarification should be decisive in this case. Our colleagues answer that question in the affirmative.

In other words, even if they conceded the covenant’s ambiguity, they would still refuse to allow Kolbe’s lawsuit to proceed. They maintain that the government’s newly offered construction of Paragraph 4 not only governs mortgage agreements entered into subsequent to the pronouncement, see Auer v. Robbins, 519 U.S. 452, 457-58, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997) (holding that agency’s

interpretation of its own regulations is owed deference), but also operates retroactively to supersede the shared understanding of private parties who previously entered into mortgages containing the flawed language. Accordingly, ambiguity aside, they conclude that the district court properly dismissed Kolbe’s complaint alleging that Bank of America, the successor-in-interest to Taylor Bean, improperly forced Kolbe to purchase additional flood insurance.

Our colleagues identify three separate strands of reasoning to support their conclusion, all of which essentially reduce to the same proposition: the government’s explanation of uniform contract terms that it promulgated trumps any other shared understanding of those terms by private contracting parties. Whatever the force of that principle in other circumstances, we strongly disagree that the government may reach back in time to override lawful agreements between two private parties who shared the same understanding of their mutual commitment. Cf. Christopher v.

SmithKline Beecham Corp., — U.S. -, 132 S.Ct. 2156, 2168, 183 L.Ed.2d 153 (2012) (“It is one thing to expect regulated parties to conform their conduct to an agency’s

interpretations once the agency announces them; it is quite another to require regulated parties to divine the agency’s

interpretations in advance____”). Federal law does not demand such a result, and our responsibility to respect private contracts should preclude such a substantial departure from legal norms. II. Even under the Bank’s and government’s view of Paragraph 4, lenders may exercise their discretion to do what Kolbe maintains that Taylor Bean did here: issue a HUD-insured home loan contingent on the borrower’s maintaining flood insurance throughout the mortgage period in an amount equal to the loan’s outstanding principal balance, i.e., at the statutory minimum for loans less than $250,000.

This is so because Paragraph 4 does not bar lenders from committing to a specific amount of required coverage for the duration of the mortgage. The pertinent sentence in the uniform covenant states only that the hazard insurance required by the lender “shall be maintained in the amounts and for the periods that Lender requires.” The lender could thus choose the statutory minimum as its required “amount[]” for the entire “period[ ]” of the loan.

The conflict in this case arises from the fact that, under the construction of Paragraph 4 advanced by Kolbe, specifying the amount and period was unnecessary because the uniform covenant itself capped the flood insurance requirement at the statutory minimum, while under the Bank’s construction, the covenant allows the lender to change the amount at any time. We presume that our colleagues would reach a different conclusion if Kolbe and Taylor Bean had signed a supplemental document setting the minimum amount of insurance as the amount required for the entire loan period.

In that case, the lender would have expressly exercised the discretion to which the Bank claims entr-tlement by choosing an amount of coverage that Congress, in the National'Flood Insurance Act, deemed adequate. Our colleagues nonetheless contend that federal policy bars us from enforcing an agreement containing those terms — based on the parties’ joint understanding of Paragraph 4, rather than on a separate written document — against Taylor Bean’s sueces-sor-in-interest, Bank of America. We elaborate below on why we believe our colleagues are wrong. A.

Integrity of Contracts If we applied ordinary contract principles to this case, the ambiguity in Paragraph 4 would foreclose dismissal of the complaint because Kolbe would be entitled to show that his understanding of the provision reflects the actual intention of the contracting parties. ' The New Jersey Supreme Court “permitís] a broad use of extrinsic evidence to achieve the ultimate goal of discovering the intent of the parties,” Conway v. 287 Corporate Ctr.

Assocs., 187 N.J. 259, 901 A.2d 341, 347 (2006), and we could not achieve that goal here without giving Kolbe the opportunity to develop the facts in support of his claim that Taylor Bean understood Paragraph 4 as he did. The New Jersey high court has described the court’s contract-interpretation role as follows: In the quest for the common intention of the parties to a contract the: court must consider the relations of the parties, the attendant circumstances and the objects they were trying to attain.

An agreement must be construed in the context of the circumstances under which it was entered into and it must be accorded a rational meaning in keeping with the express general purpose. Tessmar v. Grosner, 23 N.J. 193, 128 A.2d 467, 471 (1957); see also, e.g., Pacifico v. Pacifico, 190 N.J. 258, 920 A.2d 73, 77 (2007) (“[I]t is a basic rule of contractual

interpretation that a court must discern and implement the common intention of the parties.”). If, as Kolbe maintains, the evidence demonstrated that both he and Taylor Bean understood Paragraph 4 to cap his flood insurance obligation at the amount “required by the Secretary”: — i.e., at the statutory minimum — application of traditional contract law principles would end the matter. Where there is no dispute between the contracting parties about which of two reasonable

interpretations of their agreement is correct, the parties’ shared understanding surely would govern — barring some collateral reason to depart from ordinary principles. See Conway, 901 A.2d at 347 (“‘The polestar of construction is the intention of the parties to the contract....’” (quoting Atl. N. Airlines v. Schwimmer, 12 N.J. 293, 96 A.2d 652, 656 (1953))). The Bank, as Taylor Bean’s successor-in-interest, would stand in the original lender’s shoes, and would be bound by that shared understanding.

The Bank and our colleagues assert that this is an instance where ordinary contract principles do not apply because the language under scrutiny derives from a government source and must be interpreted uniformly in every instance. The Bank cites federal and New Jersey case law to support its contention that the government’s construction of government-generated contractual language, whether derived directly from a statute or drafted by an agency to carry out its regulatory mission, prevails even over the mutually agreed upon understanding of the parties.

The Bank, in other words, asserts that the government’s wishes about how the parties should have understood the ambiguous language override the parties’ actual understanding of the language. In so arguing, the Bank (and our colleagues) inexplicably treat this case as indistinguishable from the cases on which they rely to establish the principle of uniformity. From multiple perspectives, however, this case is unlike that precedent. Most significantly, each of the decisions highlighted by the Bank involved a dispute between the contracting parties about the meaning of language in their agreement. See Ill.

Steel Co. v. Balt. & Ohio R.R. Co., 320 U.S. 508, 508-509, 64 S.Ct. 322, 88 L.Ed. 259 (1944); Honeywell Inc. v. United States, 661 F.2d 182, 185-86 (Ct.Cl.1981); Paul Revere Life Ins. Co. v. Haas, 137 N.J. 190, 644 A.2d 1098, 1103 (1994). Hence, some principle of contract

interpretation was necessary to resolve the conflict. When such a dispute is between parties of unequal bargaining power, the stalemate is usually resolved by adopting the meaning most favorable to the non-drafting party, a method of

interpretation known as contra proferentem. See Pacifico, 920 A.2d at 78. That principle also operates in specific contexts. See, e.g., Kieffer v. Best Buy, 205 N.J. 213, 14 A.3d 737, 743 (2011) (stating that ambiguity in an indemnity provision is construed against the indemnitee); Marcinczyk v. N.J. Police Training Comm’n, 203 N.J. 586, 5 A.3d 785, 789 (2010) (stating that ambiguity in exculpatory contracts “must be resolved against the drafter of the agreement” (quoting Gershon v. Regency Dining Ctr., Inc., 368 N.J.Super. 237, 845 A.2d 720, 726 (2004))); Simonetti v. Selective Ins.

Co., 372 N.J.Super. 421, 859 A.2d 694, 698 (2004) (stating that ambiguity in an insurance contract “must be resolved against the insurer”). Similarly, when government-generated language is ambiguous in a dispute between contracting parties, courts are inclined to defer to the government’s

interpretation of its own language. See, e.g., U.S. Bank, N.A. v. Hough, 210 N.J. 187, 42 A.3d 870, 877 (2012) (noting that “we defer to an agency’s

interpretation of ... [a] regulation, within the sphere of [its] authority, unless the

interpretation is ‘plainly unreasonable’ ” (alterations in original) (internal quotation marks omitted)). But given the paramount importance of the parties’ intentions in resolving contract disputes, it is a considerably more dramatic departure from basic contract law to accept the government’s

interpretation of ambiguous language as decisive where both parties to a private contract manifested a contrary, consistent understanding of the language. The allegations in Kolbe’s complaint permit a finding that such an understanding existed. See, e.g., Marcinczyk, 5 A.3d at 788-89 (“[P]arties bargaining at arms-length may generally contract as they wish, subject only to traditional defenses such as fraud, duress, illegality or mistake.” (citations omitted)).

So long as the parties’ agreement does not violate important policy objectives, we cannot accept that the government’s interpretive authority may intrude so deeply into private contractual agreements. Cf. Shaw v. City of Jersey City, 174 N.J. 567, 811 A.2d 404, 411 (2002) (adopting construction of statute that is “consistent with both legislative design and the reasonable expectations of [the insured]” (emphasis added)). As we discuss in the next section, Kolbe’s construction of Paragraph 4 of his mortgage agreement does not conflict with the policies embodied in the National Flood Insurance Act.

The private nature of Kolbe’s mortgage agreement also distinguishes this case from some of those cited by the Bank and our. colleagues. In Honeywell, which arose in the unique context of military contracts, the United States was one of the parties and the language interpreted was purely regulatory, and. not incorporated into a contract. See 661 F.2d at 184 (noting dispute concerning Armed Services Procurement Regulation 15-205.34). The decision resulted from the appeal of a ruling by the Armed Services Board of Contract Appeals — a far cry from this private contract action. Similarly, in Saavedra v.

Donovan, 700 F.2d 496, 499-500 (9th Cir.1983), cited by our colleagues, the court rejected a government contractor’s claim in an enforcement proceeding that his failure to pay government-required fringe benefits was attributable to ambiguous language in the contracts. The decision in Lloyd v. Cincinnati Checker Cab Co., 67 Ohio App. 89, 36 N.E.2d 67 (1941), also cited by our colleagues, is similarly inapt.

The defendant there had sought to set off an insurance claim against a statutory assessment it owed because its insurer (a mutual assessment company) had been liquidated and taken over by the Ohio Superintendent of Insurance. The court held that the assessment was a non-contractual obligation of the defendant owed to the state as trustee for the insurer’s creditors and was thus “not a contractual liability involving a meeting of the minds.” Id. at 69 (“The debt of the defendant herein involved is a debt created by force of statute, not a debt created by any voluntary act of the parties.”).

Our colleagues also rely on Sharon Steel Corp. v. Chase Manhattan Bank, N.A., 691 F.2d 1039 (2d Cir.1982), for the proposition that “uniform contracts are interpreted uniformly across cases whenever it is reasonable to do so.”. This principle, they say, means that “extrinsic evidence about what a particular party intended or expected when signing the contract is generally irrelevant.” However, the context of Sharon Steel, which involved the debt securities market, was markedly different from this personal mortgage dispute.

The case concerned boilerplate language in “successor obligor clauses” in debenture indentures, and the Second Circuit empha-' sized that “uniformity in

interpretation is important to the efficiency of capital markets.” Id. at 1048. Kolbe’s and his lender’s allegedly common understanding of his personal loan has no equivalent implications. Indeed, the very cases that gave birth to the interpretive principles that the Bank and our colleagues invoke also involve contexts far removed from the mortgage relationship of Kolbe and Taylor Bean. In Auer, the plaintiffs were officers of the St.

Louis Police Department who brought suit against the city’s Board of Police Commissioners seeking overtime pay they claimed was owed under a provision of the Fair Labor Standards Act. See 519 U.S. at 455, 117 S.Ct. 905. The Board argued that the officers were not entitled to such pay based on a statutory exemption. See id. The Supreme Court deferred to the Secretary of Labor’s

interpretation of applicable regulations, which was provided in an ami-cus brief requested by the Court. Unlike here, there was no contract provision in dispute that was reasonably subject to a common understanding by the contracting parties. The circumstances in Bowles v. Seminole Rock & Sand Co., 325 U.S. 410, 65 S.Ct. 1215, 89 L.Ed. 1700 (1945), noted by our colleagues as the origin of the concept of Auer deference, are even more distant from the breach-of-contract case before us.

The dispute there was between a government official — the Administrator of the Office of Price Administration — and a manufacturer of crushed stone that was subject to a maximum price regulation. In determining the permissible price the manufacturer could charge pursuant to the regulation, the Court stated that it “must necessarily look to the administrative construction of the regulation if the meaning of the words used is in doubt.” Id. at 414, 65 S.Ct. 1215. It is also noteworthy that, in all three of the cases cited by the Bank, the co

[…]