861 A.2d 590 (D.C. 2004)
Pursuant to D.C.Code § 11-723 (1995), the United States Court of Appeals for the District of Columbia Circuit (“Circuit”) has certified the following question to this court:
Under District of Columbia law, is a client bound by a settlement agreement negotiated by her attorney when the client has not given the attorney actual authority to settle the case on those terms but has authorized the attorney to attend a settlement conference before a magistrate judge and to negotiate on her behalf and when the attorney leads the opposing party to believe that the client has agreed to those terms.
Makins v. District of Columbia, 349 U.S.App. D.C. 303, 312, 277 F.3d 544, 553 (2002).
For reasons set forth below, we answer the question in the negative. In so doing, we confine our analysis to the undisputed facts and those recited in the certified question.1
In November 1998, Brenda Makins, represented by John Harrison, Esquire, brought an action against the District of Columbia in the United States District Court for the District of Columbia claiming sex discrimination and retaliatory firing, in violation of Title VII (42 U.S.C. §§ 2000e et seq.). Makins had been employed in the District’s Department of Corrections from 1995 until her discharge in 1997. Her complaint sought reinstatement, compensatory damages, and attorney’s fees.
In the summer of 2000, at a pre-trial conference, the district judge referred Ma-kins’ case to a magistrate judge “for settlement purposes only” and ordered the District to “have present at all settlement meetings ... an individual with full settlement authority.” A similar admonition was absent as to Ms. Makins. A few days later, the magistrate ordered the “lead attorney(s) for the parties” to appear before him for a settlement conference; the order required that the “parties shall either attend the settlement conference or be available by telephone for the duration of the settlement conference.”
When the conference took place, Makins was not present. After two and a half hours of negotiations, Harrison and the attorneys for the District reached an agreement. Makins would receive $99,000 and have her personnel records amended from “discharged” to “resigned” (to preserve her retirement benefits if she were able to obtain other creditable employment). In return, Makins would dismiss her claims against the District. Mr. Harrison left the hearing room with cell phone in hand, apparently to call Ms. Makins. When he returned, the attorneys “shook hands” on the deal and later reduced it to writing. A few days later, when Harrison presented Makins with a copy for her signature, she refused to sign it. The District then filed a Motion to Enforce Settlement. Makins retained another attorney, and the court held an evidentiary hearing in which Harrison, Makins, and the lead attorney for the District testified.
The testimony of Makins and Harrison was at odds respecting whether Harrison had been given authority to settle absent a provision for her reinstatement to her job. The District Court, observing this “sharp conflict” in testimony, declined to resolve it. Instead, the court assumed arguendo that Harrison did not have actual authority to settle the case short of reinstatement. The court granted the District’s motion to enforce the settlement on the alternative ground that Harrison had apparent authority to bind Makins to the agreement. The court saw “no justification for the District of Columbia not to reasonably believe that Mr. Harrison had the full confidence and authority of his client.”
There is arguably some inconsistency as to the extent of authority required of an attorney in settlement negotiations.2 Indeed, a review of relevant case law and principles enunciated by the American Bar Association and the American Law Institute demonstrate some differences not only over the extent of authority, but also the appropriate definitions of authority. To the extent that there tends to be this inconsistency among the cases, it reflects, in part, a difference in the application or integration of agency law with legal ethics principles, the attorney-client relationship and policy considerations.
This dissonance may in part be seen as a result of the intersection of ethical guidelines and rules governing the client-lawyer relationship and the relationship of a principal to her agent in the context of settlement agreements. On the one hand, the District of Columbia Code of Professional Responsibility Ethical Consideration 7-7 provides that it is the exclusive authority of “the client to decide whether [s]he will accept a settlement offer Similarly, District of Columbia Rule of Professional Conduct 1.2(a) (2001) provides that a “lawyer shall abide by a client’s decision whether to accept an offer of settlement of a matter.”3 On the other hand, “it is well established that settlement agreements are entitled to enforcement under general principles of contract law”, Goozh v. Capitol Souvenir Co., 462 A.2d 1140, 1142 (D.C.1983) (quoting Brown v. Brown, 343 A.2d 59, 61 (D.C.1975)). Agency principles are applied to determine whether the attorney or agent had authority to bind his principal to the settlement contract. See, e.g., Bronson, supra note 3, 404 A.2d at 960; Feltman v. Sarbov, 366 A.2d 137, 139 (D.C.1976). Of course, an attorney can settle his client’s case if he or she has actual authority to do so. See United States v. Beebe, 180 U.S. 343, 352, 21 S.Ct. 371, 45 L.Ed. 563 (1901); Bronson, supra note 3, 404 A.2d at 963; Edwards v. Born, Inc., 792 F.2d 387, 389 (3d Cir.1986). Agency principles also recognize the authority of the agent to bind the client based on the doctrine of apparent authority.
The Restatement (Second) of Agency § 8 (1958) defines apparent authority as “the power to affect the legal relations of another person by transactions with third persons, professedly as an agent for the other, arising from and in accordance with the other’s manifestations to such third persons.” Thus, unlike actual authority, apparent authority does not depend upon any manifestation from the principal to her agent, but rather from the principal to the third party. Restatement (Second), supra, § 27, cmt. a. This court has stated that apparent authority arises when a principal places an agent “in a position which causes a third person to reasonably believe the principal had consented to the exercise of authority the agent purports to hold. This falls short of an overt, affirmative representation by a principal .... ” Feltman, supra, 366 A.2d at 139 (citing Drazin v. Jack Pry, Inc., 154 A.2d 553, 554 (D.C.1959)). In such circumstances, an agent’s representations need not expressly be authorized by his principal. The apparent authority of an agent arises when the principal places the agent in such a position as to mislead third persons into believing that the agent is clothed with the authority which in fact he does not possess. Id. at 140. Apparent authority depends upon “the third-party’s perception of the agent’s authority.” Sigal Construction Corp. v. Stanbury, 586 A.2d 1204, 1219 (D.C.1991) (citing Restatement (Second), supra, § 27) (other citation omitted). The third party’s perception may be based upon “written or spoken words or any other conduct of the principal which, reasonably interpreted, causes the third person to believe that the principal consents to have the act done on [her] behalf by the person purporting to act for [her].” RESTATEMENT (SECOND), supra, § 27.
We reiterate that apparent authority is an established doctrine in this court’s jurisprudence, and that settlement agreements are enforceable under general contract principles. But because apparent authority depends upon the principal’s manifestations to the third party, the issue before us is what conduct by a client in the settlement context is sufficient reasonably to cause a third person to believe that the attorney representing the client has full, final settlement authority, rather than something short of that. Whether an agent had apparent authority is a question of fact and the party asserting the existence of apparent authority must prove it. Feltman, supra, 366 A.2d at 140 (citing Rustler’s Steak House v. Environmental Assoc., Inc., 327 A.2d 536 (D.C.1974); McDonald v. Stone, 86 A.2d 624 (D.C.1952)). In determining whether the agent had apparent authority to bind the principal, “[c]onsideration should be given, inter alia, to the actual authority of the agent, the usual or normal conduct of the agent in the performance of his or her duties, previous dealings between the agent and the party asserting apparent authority, any declarations or representations allegedly made by the agent, and lastly, the customary practice of other agents similarly situated.” Management Partnership, Inc. v. Crumlin, 423 A.2d 939, 941 (D.C.1980). We take as a given that a third party in the shoes of the District of Columbia would reasonably assume that Makins had authorized attorney Harrison (1) to attend the settlement conference, and (2) to negotiate on her behalf; neither Makins nor amicus contends otherwise. We hold, however, that absent further manifestations by Makins— not Harrison— which are not contained in the certified question, there was insufficient conduct by the client to support a reasonable belief by the District that Harrison had full and final authority to agree to the settlement terms.
As pointed out, in the District of Columbia the decision to settle belongs to the client, see D.C. Rule of Prof. Conduct 1.2(a) (2004), a fact confirmed by our decisions. In Bronson, supra note 3, 404 A.2d at 960, for example, an attorney brought a declaratory judgment action to enforce a purported settlement agreement against the client. This court refused to bind the client to the agreement, because the decision to accept the agreement was the client’s and not the attorney’s. “[Regardless of the good faith of the attorney,” we stated, “absent specific authority, an attorney cannot accept a settlement offer on behalf of a client.” Id. at 963. In In re Hager, 812 A.2d 904 (D.C.2002), similarly, we sanctioned an attorney for violating a number of ethical rules relating to a purported settlement, including the principle that “ ‘[a] client’s right to accept or reject a settlement offer is absolute.’ ” Id. at 919 (quoting D.C. Bar Opinion 289). The Restatement (Third) of the Law Governing Lawyers echoes these ethical principles, observing that “[a]s between client and lawyer ... the following and comparable decisions are reserved to the client except when the client has validly authorized the lawyer to make the particular decision: whether and on what terms to settle a claim .... ” Restatement (Third) of the Law Governing Lawyers § 22 (2000). It is a principle of long standing that the decision to settle a claim belongs to the client and not the attorney. Beebe, supra, 180 U.S. at 351-52, 21 S.Ct. 371 (citations omitted). Retention of the attorney by itself is insufficient to bestow actual or apparent authority. Id. (“[T]he utter want of power of an attorney, by virtue of his general retainer only, to compromise his client’s claim, cannot, we think, be successfully disputed.”); Bronson, supra note 3, 404 A.2d at 963 (“[A]bsent specific authority, an attorney cannot accept a settlement offer on behalf of a client.”); Fennell v. TLB Kent Co., 865 F.2d 498, 502 (2d Cir.1989) (“A client does not create apparent authority for his attorney to settle a case merely by retaining an attorney.”); Edwards, supra, 792 F.2d at 390 (It is well settled that an attorney does not have the authority to compromise the client’s claim simply by virtue of his retention for the litigation.).
The Restatement (Third) further confirms the generally accepted distinction between the power to conduct negotiations and the power to end the dispute. See id. Conducting settlement negotiations is properly in the attorney’s domain: “[i]n the absence of a contrary agreement or instruction, a lawyer normally has authority to initiate or engage in settlement discussions, although not to conclude them ....” Id. § 22 cmt. C. Concluding those settlement negotiations, however, is strictly the client’s prerogative: “[t]he decision to settle is reserved to the client ... because a settlement definitively disposes of client rights.” Id. § 22 cmt. D; see also Auvil v. Grafton Homes, Inc., 92 F.3d 226, 227-28 (4th Cir.1996) (“While the record facts support the conclusion that Auvil manifested his attorney’s authority to negotiate a settlement, we do not agree that they support the conclusion that Auvil manifested his attorney’s authority to execute a settlement agreement.”) (emphasis in original).
These ethical principles are key to the issue before us, because they not only govern the attorney-client relationship, they inform the reasonable beliefs of any opposing party involved in litigation in the District of Columbia, as well as the reasonable beliefs of the opposing party’s counsel, whose practice is itself subject to those ethical constraints. It is the knowledge of these ethical precepts that makes it unreasonable for the opposing party and its counsel to believe that, absent some further client manifestation, the client has delegated final settlement authority as a necessary condition of giving the attorney authority to conduct negotiations. And it is for this reason that opposing parties— especially when represented by counsel, as here — must bear the risk of unreasonable expectations about an attorney’s ability to settle a case on the client’s behalf. “When a lawyer purports to enter a settlement binding on the client but lacks authority to do so, the burden of inconvenience resulting if the client repudiates the [unauthorized] settlement is properly left with the opposing party, who should know that settlements are normally subject to approval by the client and who has [received] no manifested contrary indication from the client.” Restatement (Third), supra, § 27 cmt. D (emphasis added). See Management Partnership, supra, 423 A.2d at 941 (Pertinent to the determination of apparent authority is “the customary practice of other agents similarly situated.”).
Applying these principles, we conclude that the two client manifestations contained in the certified question — sending the attorney to the court-ordered settlement conference and permitting the attorney to negotiate on the client’s behalf— were insufficient to permit a reasonable belief by the District that Harrison had been delegated authority to conclude the settlement. Some additional manifestation by Makins was necessary to establish that she had given her attorney final settlement authority, a power that goes beyond the authority an attorney is generally understood to have. The District, in its briefs, points only to actions and representation of record by Harrison, not Makins, as support for the reasonableness of its belief. Thus, it asserts that “Mr. Harrison represented that Ms. Makins was available by telephone and that he would consult with her when appropriate”; that “Mr. Harrison spoke on his cell phone with plaintiff at least three times during the conference”; and that “[a]t one point, Mr. Harrison left the room to phone plaintiff about the defendant’s latest settlement proposal, and returned, phone in hand, to accept the proposal with one new condition regarding amendment of personnel forms.” All of this information (including information purportedly about the client, Makins) was known to the District of Columbia only through representations made by Harrison, the attorney.4 As the Circuit Court stated in certifying the question to us: “Neither the District nor the magistrate ever heard from Makins, in person or by telephone. What the District derives from the telephone calls between Makins and Harrison amounts to nothing more than Harrison’s representations of — and the District’s educated guesses about — what was said in private between them, a disputed factual question the district court did not resolve.” Makins, 349 U.S.App. D.C. at 310, 277 F.3d at 551. Harrison’s conduct and representations about his own authority, in short, are not dispositive to whether Makins herself furnished the basis for a reasonable belief that he was authorized to conclude the settlement. See Fennell, supra, 865 F.2d at 502 (“Second Circuit case law ... explicitly rejects the notion that an agent can create apparent authority by his own actions or representations.”); Evans v. Skinner, 742 F.Supp. 30, 33 (D.D.C.1990) (“The defendants have only recited communications from the plaintiffs former attorney stating his belief that he had authority, but it is clear that an ‘agent cannot by his own acts imbue himself with apparent authority.’ ”) (quoting Fennell, supra, 865 F.2d at 503) (emphasis in Evans).
At the en banc argument, counsel for the District characterized the record as showing that Makins “sent” Harrison to the settlement conference, thus manifesting to the court and the District his apparent authority to settle her claim. But Makins had little choice, short of discharging Harrison, except to allow him to continue to represent her in the negotiations at the ordered conference. To execute a settlement agreement then and there is quite another matter.
Since Ms. Makins, as principal, did not make any manifestation of authority to the District’s attorneys, other than retaining Harrison, under the facts as certified in the question, a finding of apparent authority is precluded under the law of this jurisdiction. The District also presents several policy arguments supporting enforcement of settlement agreements on apparent authority grounds, none of which we find compelling. To be sure, settlement of disputes, both in trial courts and on appeal, is to be encouraged as sound public policy. Paul v. Bier, 758 A.2d 40, 56 (D.C.2000) (Schwelb, J., dissenting) (citing Autera v. Robinson, 136 U.S.App.D.C. 216, 218, 419 F.2d 1197, 1199 (1969)). However, we are not persuaded that the settlement process will be impeded simply by requiring some manifestation of the client’s authorization to support a claim of apparent authority in these cases where the client challenges the authority of his attorney to settle the claim. In addition, “[a]pparent authority is an equitable doctrine that places the loss on one whose manifestations to another have misled the latter.” Edwards, supra, 792 F.2d at 391; see also Restatement (Third), supra, § 27 cmt. D (As between the client and the third party, the third party should bear the risk of an unauthorized settlement because the third party “should know that settlements are normally subject to approval by the client.”). Our holding is consistent with this principle. Since Ma-kins manifested nothing by words or conduct on which reliance could be placed (she merely continued to retain Harrison), our answer to the certified question is not erosive to that policy.
We answer the certified question in the negative.
I am fully in agreement with the essential holding of the en banc court’s opinion that the enforceability of settlement agreements is governed by principles of contract and agency law, including the doctrine that a principal (the client) will be bound by the acts of an agent (the attorney) where the facts support a finding of apparent authority. See Makins v. District of Columbia, 838 A.2d 300, 306 (D.C.2003) (Ruiz, J., dissenting), vacated and reh’g en banc granted, 2004 D.C.App. Lexis 198. That doctrine, as we have long held, is grounded on a third party’s reliance on manifestations of the principal which lead the third party reasonably to expect that the acts of the agent are to be taken as authorized by the principal. Although we have previously held that the existence of apparent authority is a question of fact, see Feltman v. Sarbov, 366 A.2d 137, 140 (D.C.1976), I can join the majority’s narrow holding that the facts presented in the certified question—as interpreted by the majority to mean that Ms. Makins did nothing more than retain counsel—do not, as a matter of law, satisfy a necessary prerequisite for apparent authority: “Since Ms. Makins, as principal, did not make any manifestation of authority to the District’s attorneys, other than retaining Harrison, under the facts as certified in the question, a finding of apparent authority is precluded under the law of this jurisdiction.” (Emphasis added.)5
The certified question, however, presented the factual situation as one where the client, beyond merely retaining counsel, “has authorized the attorney to attend the settlement conference before a magistrate judge and to negotiate on her behalf and ... the attorney leads the opposing party to believe that the client has agreed to th[e] terms.” The majority takes the position that no inference of apparent authority should be permitted from the client’s authorization because she “had little choice, short of discharging Harrison, except to allow him to continue to represent her in the negotiations at the ordered conference.” But viewed through the eyes of a third party, Ms. Makins did have a choice, for the magistrate’s order provided that the District be represented by a person with settlement authority and that Ms. Makins “either attend the settlement conference or be available by telephone for the duration of the settlement conference.” The import of the magistrate’s order was evident, that the purpose of the conference was to have a negotiation with those who could come to a final settlement. Thus, Ms. Makins had the choice of attending personally, but appeared to have chosen, instead, to be available by telephone to her lawyer, whom, according to the certified question, she authorized to attend the conference to negotiate on her behalf.6
There are additional facts in the record of this case, moreover, that make the question of apparent authority intensely factual and a much closer issue. In particular, the attorney had been representing the client for several years throughout the administrative and judicial proceedings of her claim against the District, and (as testified by the client) she had been asked to be on call the day of the settlement conference and in fact had several telephone conversations with her counsel during the course of the settlement conference in which specific terms were discussed. Taking the facts in the certified question on their face and these further facts of record into consideration, supports the district court’s finding that there was apparent authority because there was a reasonable inference that opposing counsel could draw from the longstanding representation, specific authorization to negotiate settlement and the client’s conversations with her lawyer during the give-and-take of negotiations, that the client was informed and in agreement as terms of settlement were being actively discussed.
In Feltman, the court held there was apparent authority binding the client when the principal held out his retained counsel as the person to negotiate a lease. 366 A.2d at 140. Similarly in this case, the certified question posits that Ms. Makins held Mr. Harrison out as the person with whom the District should negotiate a settlement of her claims by “authoriz[ing] the attorney to attend the settlement conference before a magistrate judge and to negotiate on her behalf.” Although not stressed in our opinion in Feltman, the attorney in that case had conducted a previous negotiation for the lease of the same premises on behalf of the client, see id. at 138, 140, from which course of conduct the third party could reasonably draw the conclusion that the attorney spoke for his client on the occasion of the lease renewal as well. Likewise, in this case, there was longstanding representation by Mr. Harrison of Ms. Makins, beginning with the adverse action in the agency, continuing through a complaint before the EEOC, and culminating in the initiation and prosecution of a lawsuit in federal court. And, beyond this prior course of conduct similar to that present in Feltman, in this case there were several conversations between Mr. Harrison and Ms. Makins during the negotiations—conversations a fact finder could think the client knew or should have known would come to the attention of the court and the District’s counsel.7 Taken together, these actions by the client, though “short of an overt, affirmative representation” present a course of conduct that, the fact finder could find, placed the agent “in a position which causes a third person to reasonably believe the principal had consented to the exercise of authority the agent purports to hold.” Id. at 139 (quoting Ins. Mgmt. of Washington, Inc. v. Eno & Howard Plumbing Corp., 348 A.2d 310, 312 (D.C.1975)). Feltman, in other words, cannot meaningfully be distinguished if all the facts of record are taken into account and, arguably, presented fewer facts than this case to support apparent authority. The majority opinion relies on Feltman and does not purport to overrule it. Based on the authority of Feltman, we can hold that there is no apparent authority only if the facts are limited to the majority’s interpretation of the certified question that all the client did was retain an attorney. To the extent that the majority opinion holds otherwise, I disagree.8
I also disagree with the majority opinion’s additional and, in my view, unnecessary discussion that the third party (in this case, the District of Columbia attorneys) unreasonably relied on counsel’s representation that his client had agreed to the settlement. The majority opinion places key emphasis on the ethical principle that it is the client who decides when, and on what terms, to settle. See D.C. Rule of Prof. Conduct 1.2(a). This rule guides not only the attorney vis a vis the client, but also the reasonable expectations of other lawyers. With respect to the latter, the majority argues that because lawyers understand that the ultimate settlement decision is for the client, opposing counsel could not reasonably expect that their counterpart would have final settlement authority without some further client manifestation to that effect. This analysis is fine as far as it goes, but it does not go far enough, because it does not take into account all of the applicable ethical rules. According to the certified question, the attorney “led the opposing party to believe that the client had agreed.” That opposing counsel relied on that representation is not unusual: when lawyers are involved, lawyers deal with lawyers. This not only is common practice, but also is mandated by the ethical prohibition on direct contact with a represented party. See Rule of Prof. Conduct 4.2(a). In this context, it is imperative that lawyers be able to rely upon the facially reasonable representations of opposing counsel. That expectation is particularly sound when a representation is made in the course of a judicial proceeding, as here, not only to opposing counsel but to the presiding judicial officer. See Rule 3.3(a)(1) (“A lawyer shall not knowingly: (1) make a false statement of material fact or law to a tribunal.”). A lawyer’s operating assumption must be to expect—and our opinions place decisive significance upon—candor from counsel in their dealings with each other and with the court consistent with the rules of professional conduct we have promulgated. See Miranda v. Contreras, 754 A.2d 277, 281 (D.C.2000) (“As colleagues at bar and officers of the court, and to ensure the efficient, accurate and just operation of judicial proceedings, counsel must be able to reasonably rely on representations made by fellow counsel in the context of litigation”). I see nothing in the facts of record that should have alerted the District’s attorneys to suspend that customary assumption and instead suspect the bona fides of Mr. Harrison’s representation that settlement had been reached satisfactory to his client. To the contrary, the certified question’s formulation that counsel “led the opposing party to believe the client had agreed” carries the implication that the District’s attorneys’ (and the magistrate’s) reliance was reasonable. What we hold is that this reliance did not legally bind the client in this case because it was not based on sufficient manifestations from the client, who merely retained counsel. If the majority means to argue that the District’s continued reliance on counsel’s representations was unreasonable once it turned to enforce the settlement on an apparent authority theory which requires a manifestation by the client, I fail to see how that posture could be deemed unreasonable when the very question has been certified to us by our federal appellate colleagues, and it took our court, sitting en banc, to settle the matter.
On this limited basis, I concur in the full court’s negative answer to the certified question.