The notion that the acquiescence of parties to an administrative proceeding sanctions constitutionally suspect jurisdiction ignores the reason structural restraints, like the judicial powers clause, were built into constitutions in the first instance. The idea was to create a complex system of checks and balances that would operate to “counteract the effects of faction despite the inevitability of the factional spirit” (Sunstein, Interest Groups in American Public Law (1985) 38 Stan. L.Rev. 29, 44) and “prevent both majorities and minorities from usurping government power to distribute wealth or opportunities in their favor.” (Ibid.) These days, however, the Madisonian world has gone “topsy turvy” as factions, defined as groups “ ‘activated by some common interest . . . ,’ have become sectors of policy.” (Golembiewski & Wildavsky, The Cost of Federalism, Bare Bones: Putting Flesh on the Skeleton of American Federalism (1984) p. 73.) “Indeed,” as Aaron Wildavsky notes, “government now pays citizens to organize, lawyers to sue, and politicians to run for office. Soon enough, if current *760trends continue, government will become self-contained, generating (apparently spontaneously) the forces to which it responds.” (Ibid.) When the courts aid and abet such insularity, the implications give rise to all the more concern.
The most disturbing aspect of this case is not that the court should cede constitutional ground with such alacrity; its willingness to do so is by now well established. (See Obrien v. Jones (2000) 23 Cal.4th 40 [96 Cal.Rptr.2d 205, 999 P.2d 95]; In re Rose (2000) 22 Cal.4th 430 [93 Cal.Rptr.2d 298, 993 P.2d 956]; Leone v. Medical Board (2000) 22 Cal.4th 660 [94 Cal.Rptr.2d 61, 995 P.2d 191].) Rather, it is the encouragement given this agency—and any others that are quick studies—to do what the agency deems best for its constituency or its consequence, even if that results in arrogating core judicial functions the Legislature itself has implicitly recognized are beyond reach. I respectfully dissent.
I
In Walnut Creek Manor v. Fair Employment & Housing Com. (1991) 54 Cal.3d 245, 267 [284 Cal.Rptr. 718, 814 P.2d 704] (Walnut Creek Manor), this court determined an award by the Fair Employment and Housing Commission (Commission) of compensatory damages—for nonquantifiable emotional distress—violated the judicial powers clause of the California Constitution. (Cal. Const., art. VI, § 1; cf. id., art. XIV, § 1 [authorizing the Legislature to confer judicial powers on a commission for the general welfare of employees].) In looking-glass fashion, the majority purports to find that the 1992 and 1993 amendments to the California Fair Employment and Housing Act (FEHA; Gov. Code, § 12900 et seq.)1 have overcome and displaced the constitutional objections identified in our opinion and that the Legislature so intended. This interpretation finds no substance in either the statutory history or the text.
As the majority itself explains, the impetus for the changes was not to abrogate Walnut Creek Manor but to make the FEHA substantially equivalent to federal law to ensure certification by the United States Department of Housing and Urban Development (HUD). (Maj. opn., ante, at pp. 749-750.) Indeed, the legislation designed to accomplish this goal was initiated prior to the court’s decision. (See Dept, of Fair Employment and Housing, Request for Approval of Proposed Legislation (Feb. 21, 1991) p. 1 [requesting amendment of the FEHA “to bring the housing discrimination provisions of California law into conformity with the equivalent federal statute”].) Subsequently, the Legislature acknowledged Walnut Creek Manor (see Cal. Fair *761Employment and Housing Com., Enrolled Bill Rep. on Sen. Bill No. 1234 (1991-1992 Reg. Sess.) July 2, 1992, p. 2) but took no direct steps to address it. (Cf. Gov. Code, § 12970, subd. (a)(3); Stats. 1992, ch. 911, § 6, p. 4245 [expressly authorizing award of compensatory damages for acts of employment discrimination under FEHA after Walnut Creek Manor].)
To the extent the 1992 and 1993 amendments contained any reference to the court’s invalidation of Commission authority to award compensatory damages, they suggest legislative acquiescence: Section 12980, subdivision (d), requires that once a person files a housing discrimination complaint with the Department of Fair Employment and Housing (DFEH), the DFEH must advise the complainant of time limits, the parties’ rights, and the choice of forums and provide “a written explanation that informs the complainant that, if an accusation is issued, the complainant may only be able to recover damages for emotional distress or other intangible injuries through a civil action filed under Section 12989.” (Italics added; Stats. 1993, ch. 1277, § 9, p. 7519.) Likewise, section 12981, subdivision (g), provides that should the DFEH include in an accusation or amended accusation “a prayer for damages for emotional distress or other intangible injuries as a component of actual damages, the department shall advise the complainant, in writing, that he or she may only be able to recover damages for emotional distress or other intangible injuries through a civil action filed under Section 12989.” (Italics added; Stats. 1993, ch. 1277, § 10, p. 7522; see also Cal. Code Regs., tit. 2, § 7408, subd. (h)(10) [accusation must contain statement that complainant may recover damages for emotional distress or other intangible injury through a civil action].)
It is equally clear, as the majority further acknowledges (maj. opn., ante, at p. 751), that these changes were sufficient for HUD to consider the FEHA substantially equivalent to the Fair Housing Act (42 U.S.C. § 3601 et seq.): HUD officials found any inconsistency between state and federal law with respect to the Commission’s inability to award compensatory damages “not problematic.” (HUD Asst. Gen. Counsel Harry L. Carey, mem. to Asst. Sect. Gordon H. Mansfield, U.S. Off. of Fair Housing and Equal Opportunity, Aug. 27,1992, p. 5.) “[T]his shortcoming does not appear significant in light of the provisions empowering the agency to represent aggrieved persons, at agency expense, in a civil action which affords an opportunity for recovery of the full range of damages obtainable pursuant to the [Fair Housing] Act. [Sections] 12989 and 12989.2 afford aggrieved persons such an opportunity.” (HUD Asst. Gen. Counsel Harry L. Carey, mem. to Director Jacquelyn J. Shelton, Off. of Fair Housing Enforcement, Oct. 28, 1991, p. 4.)
Accordingly, even though the Commission recommenced awarding compensatory damages following passage of the 1992 and 1993 amendments, it *762did not do so pursuant to any legislative intent to resurrect that authority in the wake of Walnut Creek Manor. It did so strictly on its own unsanctioned initiative.
II
Even if the majority has correctly interpreted the Legislature’s intent in amending the FEHA, the result here remains incompatible with the reasoning of Walnut Creek Manor and the limitations imposed by the judicial powers clause. In McHugh v. Santa Monica Rent Control Bd. (1989) 49 Cal.3d 348 [261 Cal.Rptr. 318, 777 P.2d 91] (McHugh), the court delineated a two-part analysis for assessing the validity of an administrative agency’s award of damages: “An administrative agency may constitutionally hold hearings, determine facts, apply the law to those facts, and order relief— including certain types of monetary relief—so long as (i) such activities are authorized by statute or legislation and are reasonably necessary to effectuate the administrative agency’s primary, legitimate regulatory purposes, and (ii) the ‘essential’ judicial power (i.e., the power to make enforceable, binding judgments) remains ultimately in the courts, through review of agency determinations.” (Id. at p. 372, italics in original.)
Applying this analysis, the court in Walnut Creek Manor found the award of compensatory damages for housing discrimination ran afoul of the substantive prong. “[T]he primary regulatory purpose of the [FEHA] is to prevent discrimination in housing before it happens and, when it does occur, to offer a streamlined and economical administrative procedure to make its victim whole in the context of the housing [citation]. The award of unlimited general compensatory damages is neither necessary to this purpose nor merely incidental thereto; its effect, rather, is to shift the remedial focus of the administrative hearing from affirmative actions designed to redress the particular instance of unlawful housing discrimination and prevent its recurrence, to compensating the injured party not just for the tangible detriment to his or her housing situation, but for the intangible and nonquantifiable injury to his or her psyche suffered as a result of the respondent’s unlawful acts, in the manner of a traditional private tort action in a court of law. [Citations.]” (Walnut Creek Manor, supra, 54 Cal.3d at p. 264, italics in original.)
In reaching this conclusion, the court did not question the Commission’s contention that compensatory damages “serve to deter discrimination and compensate its victim .... Under McHugh . . . , the issue, rather, is whether the award of substantial emotional distress compensatory damages *763is ‘reasonably necessary’ to accomplish the commission’s legitimate regulatory purposes and ‘merely incidental’ to its primary regulatory purposes, or in reality transfers to the agency the judicial function of determining traditional common law claims. [Citation.]” (Walnut Creek Manor, supra, 54 Cal.3d at pp. 258-259.) The court found the award of unlimited nonquantifiable damages plainly constitutes the latter, emphasizing that “ ‘[t]he power to award compensatory and punitive tort damages to an injured party is a judicial function.’ [Citations.]” (Id. at p. 262; see McHugh, supra, 49 Cal.3d at pp. 366-367; Curtis v. Loether (1974) 415 U.S. 189, 196 [94 S.Ct. 1005, 1009, 39 L.Ed.2d 260].) In relation to the judicial powers clause, the problem the court identified was thus structural, which no amount of judicial review could salvage. (Walnut Creek Manor, at p. 265; see McHugh, at p. 364; cf. Commodity Futures Trading Comm’n v. Schor (1986) 478 U.S. 833, 851 [106 S.Ct. 3245, 3256-3257, 92 L.Ed.2d 675] (CFTC).)
Neither can legislative enactment. Regardless of statutory changes that now permit either complainant or respondent to opt out of administrative proceedings (§ 12989, subd. (a)), the immutable fact remains that authorizing the Commission to award compensatory damages invests it with a fundamental judicial prerogative. While this constitutional obstacle lies at the heart of the decision in Walnut Creek Manor, the court identified several others as well, none of which have been overcome by the 1992 and 1993 amendments. Thus, the fact remains that while restitutive damages “are quantifiable amounts of money due an injured private party,” “[g]eneral compensatory damages for emotional distress, by contrast, are not pecuniarily measurable, defy a fixed rule of quantification, and are awarded without proof of pecuniary loss. [Citations.]” (Walnut Creek Manor, supra, 54 Cal.3d at p. 263.) “As the commission itself has recognized, in seeking to place a dollar value on a complainant’s mental and emotional injuries there is little in legal authority to guide it, for the reason that ‘[i]t has traditionally been left to the trier of fact to assess the degree of harm suffered and to fix a monetary amount as just compensation therefor. [Citation.]’ [Citations.]” (Ibid.) Based on representations at oral argument, the Commission apparently now finds such “legal authority” in its own past decisions. Nevertheless, this tautology does not transmute the essential nonquantifiable character of compensatory damages. Nor has the Commission explicated the nature of its superior agency expertise in an area “traditionally” left to a trier of fact. (Ibid.; cf. CFTC, supra, 478 U.S. at pp. 844-845 [106 S.Ct. at pp. 3253-3254].)
The Commission also fails to explain why it is no longer the case that “[t]he availability of unlimited damages . . . risks converting the focus of *764the commission’s remedial decision from one of fashioning equitable remedies directed to making the injured party whole in the context of housing, to one of compensating him or her for the psychic harm suffered.” (Walnut Creek Manor, supra, 54 Cal.3d at p. 261.) Why is there no longer a concern that “[a]s the commission seeks to assess and evaluate the extent of the complainant’s injury, what once was an alternative or incidental adjunct to the primary relief of securing the same or comparable housing, has assumed an independent importance that potentially threatens to dominate the administrative hearing”? (Id. at pp. 261-262.)
In response, the majority and the Commission rely substantially, if not exclusively, on CFTC, supra, 478 U.S. 833, in which the United States Supreme Court upheld the authority of the Commodity Futures Trading Commission to adjudicate a state law counterclaim in a reparation proceeding based on the parties’ consent to the agency’s assumption of jurisdiction. (See maj. opn., ante, at pp. 751-755.) Because the FEHA did not previously contain a bilateral opt-out provision and the question of consent and its impact on the judicial powers analysis was not at issue in Walnut Creek Manor, the court is supposedly at liberty to reconsider its prior conclusions.
As the Court of Appeal below correctly understood, however, the CFTC rationale does not—and cannot—obtain when the judicial powers concern is one of substance rather than procedure. The Supreme Court expressly recognized this distinction: “Article III, § 1, safeguards the role of the Judicial Branch in our tripartite system by barring congressional attempts ‘to transfer jurisdiction [to non-Article III tribunals] for the purpose of emasculating’ constitutional courts, [citation], and thereby preventing ‘the encroachment or aggrandizement of one branch at the expense of the other.’ [Citations.] To the extent that this structural principle is implicated in a given case, the parties cannot by consent cure the constitutional difficulty for the same reason that the parties by consent cannot confer on federal courts subject-matter jurisdiction beyond the limitations imposed by Article III, § 2. [Citation.] When these Article III limitations are at issue, notions of consent and waiver cannot be dispositive because the limitations serve institutional interests that the parties cannot be expected to protect.” (CFTC, supra, 478 U.S. at pp. 850-851 [106 S.Ct. at pp. 3256-3257].)
What the high court characterized as a “structural” limitation in CFTC, this court identified as a “substantive” one in Walnut Creek Manor. (See Walnut Creek Manor, supra, 54 Cal.3d at pp. 256, 265-266; McHugh, supra, 49 Cal.3d at p. 372.) Whatever the label, the principle remains the same: the parties cannot confer upon an administrative agency authority that contravenes constitutional constraints. (Cf. Walnut Creek Manor, at pp. 257, 265; *765McHugh, at p. 364.) Regardless of consent, the award of compensatory damages remains a “ ‘judicial function’ ” (Walnut Creek Manor, at p. 262) that “ ‘has traditionally been left to the trier of fact.’ ” (Id. at p. 263.) Nor can consent shift the remedial focus of the administrative hearing back to “affirmative actions designed to redress the particular instance of unlawful housing discrimination and prevent its recurrence.” (Id. at p. 264.) For the majority to hold otherwise misperceives not only the content of the judicial powers clause but, more importantly, its constitutional significance to the very structure of our governmental system.
Ill
There is a qualitative difference between a bureaucratic process and a judicial decision and that difference remains, even when an administrative agency takes on all the attributes of a court. The preserving virtue of the judiciary is its independence. Not only are administrative agencies not immune to political influences, they are subject to capture by a specialized constituency. (See, e.g., Macey, Separated Powers and Positive Political Theory: The Tug of War Over Administrative Agencies (1992) 80 Geo. L.J. 671, 675 [“administrative agencies, like legislatures, are subject to substantial interest group influence”].) Indeed, an agency often comes into existence at the behest of a particular group—the result of a bargain between interest groups and lawmakers. (Ibid)
Administrative processes are often touted as quicker and less costly. Here, counsel argued that the Commission’s expertise makes it more efficient than the judicial system and that it could not successfully compete with the courts without authority to award compensatory damages. In this context, however, the streamlined and economical administrative procedure is efficient precisely because it is limited to making the victim whole in the context of housing and not trying to offer a broad array of judicial remedies. (Walnut Creek Manor, supra, 54 Cal.3d at pp. 261-262, 264.) In any event, these arguments are irrelevant to the constitutional question. Many constitutional mandates are inefficient, but neither efficiency nor convenience should have determinative weight when the structural integrity of the Constitution is at stake. A single branch of government that arrogated all power to itself would unquestionably be efficient. It would merely be unconstitutional.
Clearly, the administrative state is with us to stay. The question is all the more incumbent, therefore, how the courts will take the measure of administrative forays into territory assigned to the judicial branch. I would not *766abandon well-founded precedent that preserves the appropriate balance. Accordingly, I respectfully dissent.
All undesignated statutory references are to the Government Code.