I respectfully dissent, finding City of Los Angeles Ordinance No. 177231 is preempted by the National Labor Relations Act (NLRA) (29 U.S.C. § 151 et seq.).
In my view, the ordinance intrudes on the collective bargaining process in an extraordinary and fundamental way, at its very source. It determines the individual workers who will comprise a new employer’s workforce for the first 90 days of its operation. During those 90 days, the new employer must provide employment to a group of workers it did not choose. In addition, the new employer must provide to this mandated workforce specified benefits (such as termination for cause only) for which the workers did not bargain and to which the new employer did not agree. I recognize that governments, *212including states and municipalities, have the authority to impose minimum employment standards of general application—including restrictions on hiring and firing. Consequently, an employer has no absolute right to choose its employees by, for example, discriminating on the basis of some group characteristic or union membership. But federal labor law does not permit a government mandate that an employer hire either a particular worker or a specific group of workers based on a group characteristic (or on anything else). That fundamental choice is left under federal law to the employer and employee, that is, to the free play of economic forces. This has been clear since the Supreme Court first upheld the constitutionality of the NLRA in Labor Board v. Jones & Laughlin (1937) 301 U.S. 1, 45 [81 L.Ed. 893, 57 S.Ct. 615] (Jones & Laughlin), when the high court said that the NLRA “does not interfere with the normal exercise of the right of the employer to select its employees or to discharge them.”
The point—that the NLRA does not interfere with an employer’s selection of its workforce—has been reiterated many times by the high court: in NLRB v. Burns Security Services (1972) 406 U.S. 272, 294 [32 L.Ed.2d 61, 92 S.Ct. 1571] (Burns) (an employer “is ordinarily free to set initial terms on which it will hire the employees of a predecessor”); in Howard Johnson Co. v. Hotel Employees (1974) 417 U.S. 249, 262 [41 L.Ed.2d 46, 94 S.Ct. 2236] (Howard Johnson) “Clearly, Bums establishes that [the new employer] had the right not to hire any of the former [employer’s] employees, if it so desired”; and in Fall River Dyeing & Finishing Corp. v. NLRB (1987) 482 U.S. 27, 40 [96 L.Ed.2d 22, 107 S.Ct. 2225] (Fall River Dyeing) (“[w]e further explained [in Burns] that the successor is under no obligation to hire the employees of its predecessor”).
The majority says these high court precedents establish only a principle of federal common law, and have no bearing on the authority of state and local governments to require the hiring for 90 days of a particular bloc of workers. With this I cannot agree. Under the preemption doctrine established in Machinists v. Wisconsin Emp. Rel. Comm’n (1976) 427 U.S. 132 [49 L.Ed.2d 396, 96 S.Ct. 2548] (Machinists) and its progeny, the salient inquiry is “whether Congress intended that the conduct involved be unregulated because left ‘to be controlled by the free play of economic forces.’ ” (Id. at p. 140.) The implicit right of the employer to select its employees in the first instance—recognized by the high court since 1937—is, it seems to me, as fundamental to the structure of the NLRA as is the correlative express right of those selected employees to organize to secure the redress of grievances and to promote agreements with the employer on working conditions. (See Jones & Laughlin, supra, 301 U.S. at pp. 43-14.) State or municipal regulation that directly interferes with that right is preempted by the NLRA under Machinists doctrine as surely as is regulation that is directed at the collective bargaining process itself. And, even if we ignore the employer’s *213right to select its workforce as a fundamental premise of the NLRA, the impact of the ordinance on the determination whether the new employer is a successor—and thus bound to bargain with the union selected by the predecessor’s employees—likewise is an unpermitted intrusion on the collective bargaining process.
1. The ordinance
The City of Los Angeles (City) adopted the Grocery Worker Retention Ordinance in 2005. (L.A. Ord. No. 177231, adding ch. XVIII, § 181.00 et seq. to the L.A. Mun. Code.) The ordinance is described fully in the majority opinion. In brief, it applies to grocery stores exceeding a specified size that undergo a change of ownership. It gives nonmanagerial employees of the former owner who have worked for that owner for at least six months the right to demand employment by the new owner: the new employer must select its employees from among that group, by seniority, and must retain them for 90 days, discharging them only for cause. After 90 days, the new employer must prepare a written performance evaluation for each such employee and consider offering continued employment to those with satisfactory evaluations. As the majority notes, other municipalities in California and elsewhere have adopted similar ordinances.
2. The NLRA
The NLRA “is a comprehensive code passed by Congress to regulate labor relations in activities affecting interstate and foreign commerce. As such it is of course the law of the land which no state law can modify or repeal.” (Nash v. Florida Industrial Comm’n. (1967) 389 U.S. 235, 238 [19 L.Ed.2d 438, 88 S.Ct. 362].) The NLRA declares the policy of the United States: “to eliminate . . . obstructions to the free flow of commerce ... by encouraging the practice and procedure of collective bargaining and by protecting the exercise by workers of full freedom of association, self-organization, and designation of representatives of their own choosing, for the purpose of negotiating the terms and conditions of their employment or other mutual aid or protection.” (29 U.S.C. § 151.)
The need for the NLRA flowed from a free market economy in which equality of bargaining power did not exist. Employers were (and are) “organized in the corporate or other forms of ownership association” and employees “[did] not possess full freedom of association or actual liberty of contract. . . .” (29 U.S.C. § 151.) Congress found it unnecessary to say in the NLRA that the composition of an employer’s workforce is a matter of the employer’s choice, but in our country, hiring has always occurred on an individual basis, one employee at a time: the employer advertises, prospective workers apply, and the employer selects.
*214The Supreme Court recognized this underlying premise of the NLRA almost immediately, when it said in 1937 that the NLRA “does not prevent the employer ‘from refusing to make a collective contract and hiring individuals on whatever terms’ the employer ‘may by unilateral action determine.’ ” (Jones & Laughlin, supra, 301 U.S. at p. 45.) The NLRA prohibits specified unfair labor practices by employers and unions (29 U.S.C. § 158) and empowers the National Labor Relations Board (NLRB) to prevent anyone from engaging in those unfair labor practices. (29 U.S.C. § 160.) The point of the NLRA was to “restor[e] equality of bargaining power between employers and employees” (29 U.S.C. § 151) by requiring an employer to bargain with “representatives of [the workers’] own choosing” (ibid.)—and not to “interfere with the normal exercise of the right of the employer to select its employees or to discharge them.” (Jones & Laughlin, supra, 301 U.S. at p. 45.)
Thus, the very foundation of the NLRA lies in a workplace composed of individuals selected by the employer. The NLRA protects the rights of those individuals whom the employer has chosen to hire, to associate with each other, to organize themselves, and to designate representatives of their choosing to negotiate the terms and conditions of their employment. (29 U.S.C. § 151.) If the workers vote to select a union to represent them, the employer must negotiate with the union in a good faith effort to reach agreement on wages, hours and other terms of employment. The NLRA does not mandate that the parties reach agreement, only that they try to agree. If they cannot agree, the NLRA establishes the rules of the battleground, protecting certain pressure tactics and prohibiting others, to keep the process fair.
The NLRA specifies the single circumstance under which the NLRB may interfere with the employer’s selection of its workers: the employer cannot hire or fire based on union membership. The NLRA specifies that an employer commits an unfair labor practice “by discrimination in regard to hire or tenure of employment or any term or condition of employment to encourage or discourage membership in any labor organization . . . .” (29 U.S.C. § 158 (a)(3).) Jones & Laughlin, after stating that the NLRA does not interfere with the employer’s right to select or discharge its employees, tells us that “[t]he employer may not, under cover of that right, intimidate or coerce its employees with respect to their self-organization and representation, and, on the other hand, the Board is not entitled to make its authority a pretext for interference with the right of discharge when that right is exercised for other reasons than such intimidation and coercion.” (Jones & Laughlin, supra, 301 U.S. at pp. 45-46.) Twelve years after deciding Jones & Laughlin, the high court announced the corollary principle that states may prohibit discrimination in hiring and firing against nonunion workers. (Lincoln Union v. Northwestern Co. (1949) 335 U.S. 525, 537 [93 L.Ed. 212, 69 S.Ct. *215251] [“Just as we have held that the due process clause erects no obstacle to block legislative protection of union members, we now hold that legislative protection can be afforded non-union workers.”].)
The majority is mistaken in saying I have expressed here the view that the NLRA was founded on an employer’s absolute right to select its employees. Plainly, the NLRA was not founded on an employer’s liberty to discriminate against either union or nonunion members because they are such. In my view, regulations prohibiting unlawful discrimination and imposing minimum employment standards in no way undermine the foundation of the NLRA that each workplace may be composed of individuals selected by the employer for reasons other than discrimination.
The NLRA does not transform the fundamentally individual nature of the employment relationship. Its focus is on the right of individual workers to band together, if they so choose, and to select representatives of their own choosing, to bargain with the employer who has hired them. This is the base upon which the NLRA is constructed and from which all of its provisions flow. Just as Jones & Laughlin made it clear that Congress did not intend in the NLRA to change the fundamental, individual nature of the employer-employee relationship, including “the right of the employer to select its employees” (Jones & Laughlin, supra, 301 U.S. at p. 45), I conclude it necessarily follows that Congress did not intend to allow any other governmental entity to do so either. To permit a city to mandate that a new employer hire the predecessor’s employees as a group violates the fundamental structure of the NLRA and necessarily hands weapons of economic power to a group the employer did not choose to hire. It is preempted under the doctrine established in Machinists and its progeny, to which I now turn.
3. Machinists preemption
Machinists examined the history of labor law preemption under the NLRA and identified two lines of preemption analysis. The first is based on the primary jurisdiction of the NLRB to regulate conduct that is a protected right under section 7 (29 U.S.C. § 157) or conduct that is an unfair labor practice in violation of section 8 of the NLRA (29 U.S.C. § 158). State regulation of conduct that is protected under section 7 or an unfair labor practice under section 8 is preempted under what is now called the Garmon preemption. (San Diego Unions v. Garmon (1959) 359 U.S. 236, 244 [3 L.Ed.2d 775, 79 S.Ct. 773].) Because the City’s ordinance does not “ ‘regulate activity that the NLRA protects, prohibits, or arguably protects or prohibits’ ” (Chamber of Commerce of United States v. Brown (2008) 554 U.S. 60, 65 [171 L.Ed.2d 264, 128 S.Ct. 2408] (Brown)) under sections 7 and 8, there is no Garmon preemption.
*216But Machinists recognized “a second line of pre-emption analysis . . . developed in cases focusing upon the crucial inquiry whether Congress intended that the conduct involved be unregulated because left ‘to be controlled by the free play of economic forces.’ ” (Machinists, supra, 427 U.S. at p. 140.) As Brown put it, Machinists preemption forbids both the NLRB and states from regulating conduct Congress intended be unregulated: “Machinists pre-emption is based on the premise that ‘ “Congress struck a balance of protection, prohibition, and laissez-faire in respect to union organization, collective bargaining, and labor disputes.” ’ ” (Brown, supra, 554 U.S. at p. 65; see id. at p. 66 [holding statutes preempted under Machinists “because they regulate within ‘a zone protected and reserved for market freedom’ ”].)
Under Machinists, “congressional intent to shield a zone of activity from regulation is usually found only ‘implicitly] in the structure of the Act,’ [citation], drawing on the notion that ‘ “[w]hat Congress left unregulated is as important as the regulations that it imposed,” ’ [citation].” (Brown, supra, 554 U.S. at p. 68.) This is just such a case. An employer’s right to select the members of the workforce with whose representatives it will be required to bargain, if they so choose, is the foundation on which the NLRA was built, unstated but implicit in its structure. This is a paradigm of the principle that what Congress left unregulated is fully as important as the regulations it imposed. In my view, this alone is reason enough to conclude the City’s ordinance, which stands in direct contradiction to one of the building blocks of the NLRA, cannot stand.
Machinists arose from the collective bargaining process itself and the pressure tactics employed by both sides during that process. In Machinists, the court found the state could not enjoin a union and its members from refusing to work overtime as part of a union effort to put economic pressure on the employer during contract negotiations. (Machinists, supra, 427 U.S. at p. 133.) The court observed that many of its past decisions concerning conduct “left by Congress to the free play of economic forces” (id. at p. 147) involved union and employee activities, but that “self-help is of course also the prerogative of the employer because he, too, may properly employ economic weapons Congress meant to be unregulable.” (Ibid.) Thus, “ ‘[r]e-sort to economic weapons’ ” was the right of both employer and employee, “and the State may not prohibit the use of such weapons or ‘add to an employer’s federal legal obligations in collective bargaining’ any more than in the case of employees.” (Ibid.) So, “[w]hether self-help economic activities are employed by employer or union, the crucial inquiry regarding preemption is the same: whether ‘the exercise of plenary state authority to curtail or entirely prohibit self-help would frustrate effective implementation of the Act’s processes.’ ” (Id. at pp. 147-148.) Congress meant that these activities “were not to be regulable by States any more than by the NLRB” (id. at *217p. 149), and sanctioning state regulation of economic pressure “deemed by the federal Act ‘desirabl[y] . . . left for the free play of contending economic forces’ ” amounts to “ ‘denying one party to an economic contest a weapon that Congress meant him to have available.’ ” (Id. at p. 150.) Machinists concluded that such regulation by the state was “impermissible because it ‘ “stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.” ’ ” (Id. at p. 151.)
The majority concludes from Machinists and subsequent cases that the areas Congress intended to leave free of local regulation are confined to those relating to the process by which an agreement is reached on the terms of employment and not on the substance of those terms. Certainly, it is true that the NLRA is not concerned with the substance of the parties’ agreement, and cases since Machinists have made it clear that state or federal regulations establishing minimum terms of employment that protect union and nonunion workers alike “ ‘neither encourageQ nor discourage[] the collective-bargaining processes that are the subject of the NLRA’ ” and are not preempted. (Fort Halifax Packing Co. v. Coyne (1987) 482 U.S. 1, 21, 23 [96 L.Ed.2d 1, 107 S.Ct. 2211] (Fort Halifax) [Me. statute requiring employers to provide a one-time severance payment to employees in the event of a plant closing was not preempted; “establishment of a minimum labor standard does not impermissibly intrude upon the collective-bargaining process’’]; see also Metropolitan Life Ins. Co. v. Massachusetts (1985) 471 U.S. 724, 757, 758 [85 L.Ed.2d 728, 105 S.Ct. 2380] (Metropolitan Life) [“[w]hen a state law establishes a minimal employment standard not inconsistent with the general legislative goals of the NLRA, it conflicts with none of the purposes of the Act”; state law requiring that minimum mental health benefits be provided under certain insurance policies was not preempted].)
But in my view, the ordinance profoundly interferes with the collective bargaining process. We are not here concerned with local regulation “establishing minimum terms of employment.” (Metropolitan Life, supra, 471 U.S. at p. 754.) Nor are we concerned with state regulations “prescribing the minimum wages, maximum hours, and standard conditions of employment . . . .” (Industrial Welfare Com. v. Superior Court (1980) 27 Cal.3d 690, 698 [166 Cal.Rptr. 331, 613 P.2d 579].) The ordinance does not merely impose a term or condition of employment, in the sense of requiring mental health insurance coverage as in Metropolitan Life, or a severance payment in the event of a plant closing as in Fort Halifax, or minimum wages and maximum hours as in Industrial Welfare Com.—benefits that apply across the board to every individual employee among the employer’s selected workforce.
We are concerned with a local regulation that creates employment by dictating which individuals the new grocery owner must hire, and then *218establishes minimum standards for those workers. It is the initial requirement to hire, not the minimum standards applied to those hired, that causes the ordinance to be preempted under Machinists. The requirement to hire, of course, is not a part of the collective bargaining process at all. But it necessarily subverts that process, because it determines the members of the workforce who will decide the matters of “self-organization and collective bargaining” (Metropolitan Life, supra, 471 U.S. at p. 751) that are concededly the exclusive province of the NLRA.
As the majority points out, the ordinance does not, on its face, regulate organizing or bargaining. But the ordinance does indeed regulate the economic contest, in at least two basic ways. (See Machinists, supra, 427 U.S. at p. 150 [state regulation of economic pressure, deemed by the NLRA to be left to the free play of contending economic forces, is “ ‘denying one party to an economic contest a weapon that Congress meant him to have available’ ”].)
First, the fact that the employer’s choice of employees who will comprise its workforce necessarily precedes any “ ‘ “union organization, collective bargaining, and labor disputes” ’ ” (Brown, supra, 554 U.S. at p. 65) does not reduce the significance of that choice to the regulatory scheme Congress fashioned in the NLRA, which was created to provide “a framework for self-organization and collective bargaining.” (Metropolitan Life, supra, 471 U.S. at p. 751.) The people who comprise the workforce are an integral part of that framework, and the employer’s right to select its employees, one by one, was left undisturbed in the NLRA, as Jones & Laughlin explicitly tells us. So, I cannot accept the notion that states are left free to do what the NLRA does not do simply because the conduct in question—hiring the employees who will launch the new employer’s grand opening—precedes the collective bargaining process.
Second, while the City’s ordinance is not expressly aimed at the collective bargaining process, I think it is fair to say that the ordinance itself is an economic weapon—not a weapon wielded by either of the parties to a labor dispute, but one superimposed by the City that necessarily “ ‘upset[s] the balance that Congress has struck between labor and management in the collective-bargaining relationship.’ ” (Metropolitan Life, supra, 471 U.S. at p. 751.) An ordinance that dictates whom a new employer must hire denies that new employer the ability to choose the employees it deems most suited to ensuring the success of its enterprise, and does so during the critical first 90 days of its operation. This is a matter of considerable moment to new owners with their own management styles, the implementation of which depends on selecting employees one by one.
The trial testimony demonstrates this point, showing that the first 90 days of a new grocery owner’s operation are the most important to establish the *219new owner’s image and to “deliver” to new customers. One witness said, “The supermarket business is a very competitive business, and if you don’t deliver to the customers in the first 90 days, you’ve probably lost them.” Another witness testified it was critical to his company, when acquiring a new store, to bring in experienced personnel from the grocer’s other stores who understand the company’s business philosophy. The company tries to hire from the community (and if a predecessor’s employee were the right fit for the company, it would hire him or her to work in one of its other stores), but the company had significant concerns about retaining a bloc of employees without the option “to determine whether they would fit, have the work ethic, the work history that would fit our organizational style as a collective group”; “we would have more difficult[y] running that store from day one in the style and fashion that we require.” One witness indicated his company’s need to “hire our own crew” because the company wanted bilingual employees to fully serve the shoppers to whom it catered. In short, new owners of previously faltering grocery stores have sound reasons for the “normal exercise” of their right under federal law to select their own workforces. (Jones & Laughlin, supra, 301 U.S. at p. 45.) The ordinance, stripping away the new employer’s choices in hiring during the critical first 90 days, operates as an economic weapon and directly affects the economic activities of the employer. To say that it is not a form of economic pressure that alters the collective bargaining relationship is, in my view, fundamentally erroneous.
The economic pressure is not imposed by the employees as union members during collective bargaining, but it is government imposed on their behalf during a time when they are free to engage in the process of self-organization governed by the NLRA. The ordinance mandates which individuals are to comprise the workforce that will decide how or whether to bargain with the employer. To deny that an economic weapon is in play between contending economic forces as a result of the ordinance is to ignore the fundamental relationship between labor and management upon which the NLRA is constructed. The requirement to hire a specific bloc of employees, it seems to me, necessarily “has altered the economic balance between labor and management.” (New York Tel. Co. v. New York Labor Dept. (1979) 440 U.S. 519, 532 [59 L.Ed.2d 553, 99 S.Ct. 1328]; see also Burns, supra, 406 U.S. at p. 288 [“The congressional policy manifest in the Act is to enable the parties to negotiate for any protection either deems appropriate, but to allow the balance of bargaining advantage to be set by economic power realities.”].) Workforce selection is perforce one of the most basic “economic power realities.”
The majority’s view is that the ordinance is of a piece with other state and local regulations broadly regulating hiring and firing, and that it does not regulate bargaining or speak directly to the process of organizing; instead the ordinance temporarily preserves the status quo, whatever that *220may be (whether the workforce is unionized or not). (Maj. opn., ante, at pp. 199-200.) But the flaw in the City’s ordinance lies in the mandated selection of an employer’s workforce in the first instance, a sui generis form of regulation. And, in my view, the temporary nature of the ordinance is entirely irrelevant; either Congress intended that the employer’s right to select its workforce be unregulated, or it did not; I do not see how there can be any middle ground on congressional intent. (See Metropolitan Life, supra, 471 U.S. at p. 749 [the “second [Machinists] pre-emption doctrine protects against state interference with policies implicated by the structure of the Act itself, by pre-empting state law . . . concerning conduct that Congress intended to be unregulated”; while “[a]n appreciation of the State’s interest in regulating a certain kind of conduct may still be relevant in determining whether Congress in fact intended the conduct to be unregulated” (id. at pp. 749-750, fn. 27), such preemption “does not involve in the first instance a balancing of state and federal interests, . . . but an analysis of the structure of the federal labor law to determine whether certain conduct was meant to be unregulated” (ibid., italics added & citation omitted)].)
In the end, the majority frames the question as whether there is evidence that Congress affirmatively intended to leave the subject of employee “retention” during ownership transitions unregulated by states and municipalities, and finds the NLRA “resoundingly silent.” (Maj. opn., ante, at p. 198.) In my view, this misstates the question, which is whether Congress intended to leave unregulated the employer’s initial selection—not the retention—of its workforce. And the NLRA is indeed silent on both questions; there is nothing in the text of the NLRA about hiring or firing (except to prohibit doing either based on union affiliation). But the silence is hardly surprising; the high court has told us that congressional intent to shield a zone of activity from regulation in most cases is found “only ‘implicit[ly] in the structure of the Act’ ” (Brown, supra, 554 U.S. at p. 68), not in its text. The NLRA was founded on, and everything in it assumes, the existence of a freely formed employer-employee relationship. (Jones & Laughlin, supra, 301 U.S. at p. 45.) It would be anomalous to conclude that, despite this foundation, Congress did not intend to prevent states and localities from doing the very thing that it declined to do: interfere with an employer’s selection of the people who will conduct its business.
For the foregoing reasons alone, I would find the City’s ordinance preempted. But there is more, because the ordinance does have a direct impact on the collective bargaining process. The ordinance’s requirement that the new owner hire the predecessor’s workforce necessarily influences whether or not the new employer will be considered a successor to the former employer, and thus bound to bargain with the union selected by the predecessor’s employees—a direct intrusion on the collective bargaining process.
*2214. The successorship doctrine
If a new employer is deemed a successor of the old employer and hires a majority of its employees from the predecessor’s workforce, the new employer has a duty to recognize and bargain with the union representing the predecessor’s employees. This “successorship doctrine” developed through several high court cases. The first of these, John Wiley & Sons v. Livingston (1964) 376 U.S. 543 [11 L.Ed.2d 898, 84 S.Ct. 909] (John Wiley), involved the disappearance by merger of a corporate employer and the wholesale transfer of its employees to the company into which the corporate employer merged. The union representing the predecessor’s employees asserted that the new employer (Wiley) was obligated to recognize certain rights of the employees under its collective bargaining agreement with Wiley’s predecessor; Wiley asserted the merger terminated the collective bargaining agreement for all purposes; and the union sought to compel arbitration. (Id. at pp. 545-546.)
The high court first observed that “[flederal law, fashioned ‘from the policy of our national labor laws,’ controls.” (John Wiley, supra, 376 U.S. at p. 548.) It then held that the disappearance by merger of a corporate employer that had entered into a collective bargaining agreement did not automatically terminate all rights of the employees covered by the agreement, and that, “in appropriate circumstances, present here, the successor employer may be required to arbitrate with the union under the agreement.” (Ibid.) The court elaborated: “The objectives of national labor policy, reflected in established principles of federal law, require that the rightful prerogative of owners independently to rearrange their businesses and even eliminate themselves as employers be balanced by some protection to the employees from a sudden change in the employment relationship. The transition from one corporate organization to another will in most cases be eased and industrial strife avoided if employees’ claims continue to be resolved by arbitration rather than by ‘the relative strength ... of the contending forces,’ [citation].” (Id. at p. 549.)
The court found Wiley’s obligation to arbitrate the dispute “in the [predecessor’s] contract construed in the context of a national labor policy.” (John Wiley, supra, 316 U.S. at pp. 550-551; see id. at p. 550 [“the impressive policy considerations favoring arbitration are not wholly overborne by the fact that Wiley did not sign the contract being construed”].) But the court cautioned: “We do not hold that in every case in which the ownership or corporate structure of an enterprise is changed the duty to arbitrate survives. . . . [T]here may be cases in which the lack of any substantial continuity of identity in the business enterprise before and after a change would make a duty to arbitrate something imposed from without, not *222reasonably to be found in the particular bargaining agreement and the acts of the parties involved.” (Id. at p. 551.) In John Wiley, the required “similarity and continuity of operation across the change in ownership [was] adequately evidenced by the wholesale transfer of [the predecessor’s] employees to the Wiley plant. . . .” (Ibid.)
Next, the high court decided Burns, supra, 406 U.S. 272, holding that an employer that hired a majority of its predecessor’s employees had an obligation to bargain with the union that represented those employees, but could not be required to observe the terms of the union’s collective bargaining agreement with the predecessor. (Id. at pp. 278, 286-287.) In Burns there was no relationship between the new employer and its predecessor. The new employer replaced the predecessor when it bid for and obtained a service contract to provide plant protection services. (Id. at pp. 274-275.) The new employer’s obligation to bargain with the union “stemmed from its hiring of [the former employer’s] employees and from the recent election and Board certification.” (Id. at pp. 278-279.) “[I]t would be different if Burns had not hired employees already represented by a union certified as a bargaining agent . . . .” (Id. at p. 280.)
The Burns court noted that the NLRB “has never held that the National Labor Relations Act itself requires that an employer who submits the winning bid for a service contract or who purchases the assets of a business be obligated to hire all of the employees of the predecessor though it is possible that such an obligation might be assumed by the employer.” (Burns, supra, 406 U.S. at p. 280, fn. 5.) But while the new employer had a duty to bargain, it could not be required to honor the terms of the previous employer’s collective bargaining agreement. “The source of its duty to bargain with the union is not the collective-bargaining contract but the fact that it voluntarily took over a bargaining unit that was largely intact and that had been certified within the past year. Nothing in its actions, however, indicated that Bums was assuming the obligations of the contract, and ‘allowing the Board to compel agreement when the parties themselves are unable to agree would violate the fundamental premise on which the Act is based—private bargaining under governmental supervision of the procedure alone, without any official compulsion over the actual terms of the contract.’ ” (Id. at p. 287.) And, Burns said, a successor employer “is ordinarily free to set initial terms on which it will hire the employees of a predecessor . . . .” (Id. at p. 294.)
Then came Howard Johnson, supra, 417 U.S. 249, where the court held that the new employer, who hired only a small fraction of the predecessors’ employees, could not be compelled to arbitrate, under its predecessors’ collective bargaining agreements, the extent of its obligations under those agreements to the predecessors’ employees. (Id. at pp. 250, 264.) In Howard *223Johnson, the new employer leased restaurant and motor lodge premises from the old employer, and purchased all the personal property used in their operations. (Id. at p. 251.) The seller notified its employees their employment would terminate when operations were transferred. Upon the sale, the new employer began operations with 45 employees, only nine of whom had been employed by the seller, and the union sued, seeking arbitration, which the trial court granted under John Wiley. (Howard Johnson, at pp. 252-253.)
The high court held that Burns principles applied. The court acknowledged some inconsistencies in reasoning between John Wiley and Burns, but found it unnecessary to decide whether an irreconcilable conflict existed, because the facts in John Wiley were entirely different from those before the court in Howard Johnson. (Howard Johnson, supra, 417 U.S. at pp. 255-256.) In addition to the fact (among others) that the seller remained a viable entity after the asset sale, “[e]ven more important, in Wiley the surviving corporation hired all of the employees of the disappearing corporation,” making no substantial changes in the operation of the business. (Howard Johnson, at p. 258.) “It was on this basis that the Court in Wiley found that there was the ‘substantial continuity of identity in the business enterprise,’ [citation], which it held necessary before the successor employer could be compelled to arbitrate.” (Howard Johnson, at p. 259.)
By contrast, the new employer in Howard Johnson “decided to select and hire its own independent work force . . . .” (Howard Johnson, supra, 417 U.S. at p. 259.) The union’s position was that the new employer was bound by the seller’s collective bargaining agreement to hire all of the seller’s former employees. (Id. at p. 260.) This position, the high court said, was “completely at odds with the basic principles this Court elaborated in Burns.” (Id. at p. 261.) The court said: “Clearly, Burns establishes that Howard Johnson had the right not to hire any of the former [employer’s] employees, if it so desired.” (Id. at p. 262.) The continuity of identity in the business enterprise that John Wiley required “necessarily includes, we think, a substantial continuity in the identity of the work force across the change in ownership.” (Howard Johnson, at p. 263.)
Moreover: “This interpretation of Wiley is consistent also with the Court’s concern with affording protection to those employees who are in fact retained in ‘[t]he transition from one corporate organization to another’ from sudden changes in the terms and conditions of their employment . ... At the same time, it recognizes that the employees of the terminating employer have no legal right to continued employment with the new employer .... This holding is compelled, in our view, if the protection afforded employee interests in a change of ownership by Wiley is to be reconciled with the new employer’s right to operate the enterprise with his own independent labor force.” (Howard Johnson, supra, 417 U.S. at p. 264, italics added.)
*224Finally, there was Fall River Dyeing, where the court held that a successor’s obligation to bargain was not limited to a situation where (as in Burns) the union had recently been certified; instead, where a union has a rebuttable presumption of majority status, that status continues through a change in employers if “the new employer is in fact a successor of the old employer and. the majority of its employees were employed by its predecessor.” (Fall River Dyeing, supra, 482 U.S. at p. 41, italics added.) Fall River Dyeing repeated the principles established in Burns: “We further explained that the successor is under no obligation to hire the employees of its predecessor .... If the new employer makes a conscious decision to maintain generally the same business and to hire a majority of its employees from the predecessor, then the bargaining obligation of § 8(a)(5) is activated. This makes sense when one considers that the employer intends to take advantage of the trained work force of its predecessor.” (Fall River Dyeing, at pp. 40-41, citations omitted.) Thus “[t]he ‘triggering’ fact for the bargaining obligation was this composition of the successor’s work force.” (Id. at p. 46; see id. at p. 47 [adopting NLRB rule that the determination of the composition of the successor’s workforce is to be made when a transitioning successor has employed a “ ‘substantial and representative complement’ ” of its workforce; “[i]f, at this particular moment, a majority of the successor’s employees had been employed by its predecessor, then the successor has an obligation to bargain with the union that represented these employees”].)
The majority sees nothing in these successorship cases (which presented no preemption issue) to suggest that Congress intended in the NLRA to prevent states and localities from interfering with the employer’s right to choose its employees. I cannot agree. From John Wiley to Fall River Dyeing, these cases show that the employer’s free selection of employees has always been a fundamental part of national labor policy. John Wiley and Howard Johnson could hardly have been more clear: John Wiley tells us that the “objectives of national labor policy, reflected in established principles of federal law, require that the rightful prerogative of owners independently to rearrange their businesses” be balanced “by some protection to the employees from a sudden change in the employment relationship” (John Wiley, supra, 376 U.S. at p. 549, italics added), and Howard Johnson tells us that the high court in John Wiley was “concern[ed] with affording protection to those employees who are in fact retained in ‘[t]he transition from one corporate organization to another’ from sudden changes in the terms and conditions of their employment . . . .” (Howard Johnson, supra, 417 U.S. at p. 264, italics added.)
In short, it seems to me incontrovertible that the “objectives of national labor policy,” as stated by the high court, have always included the employer’s prerogative to select its workforce (subject to the prohibition on discrimination on the basis of union membership). As Howard Johnson put it, “the *225employees of the terminating employer have no legal right to continued employment with the new employer . . . .” (Howard Johnson, supra, 417 U.S. at p. 264.) Consequently, states or cities are not free to regulate to the contrary. (See John Wiley, supra, 376 U.S. at p. 548 [“Federal law, fashioned ‘from the policy of our national labor laws,’ controls.”]; cf. Howard Johnson, supra, 417 U.S. at p. 255 [observing that federal common law regarding enforcement of collective-bargaining agreements “must be ‘fashion[ed] from the policy of our national labor laws’ ”].)
The City conceded in oral argument that the ordinance would be preempted by the NLRA if it expressly declared the new owner to be a successor within the meaning of the successorship doctrine, but the City asserted the express 90-day limitation avoids preemption. If the ordinance were unlimited in duration, I think there would be little doubt, under the terms of the ordinance itself, that it would dictate the successorship outcome. The new employer— defined in the ordinance as the person “that owns, controls, and/or operates the Grocery Establishment” (L.A. Mun. Code, § 181.01, subd. I) after the transfer of “all or substantially all of the assets or a controlling interest” (id., § 181.01, subd. B) of the old employer—would necessarily be a successor, and would necessarily be bound to bargain with the union representing the workers that the new employer has been required (indefinitely under this scenario) to retain. This scenario would present an obvious intrusion into the collective bargaining process, requiring the employer to bargain with a workforce it did not choose.1
In my view, the ordinance seeks to do indirectly that which the City acknowledges it cannot do directly, because the practical effect of the ordinance is to visit upon the new owner all the obligations of a successor under the successorship doctrine. The majority, however, concludes, based on existing NLRB decisions by the agency’s administrative law judges (ALJ’s) and a federal district court decision, that the new employer will be obligated *226to bargain “only if the successor employer retains its predecessor’s employees beyond the mandatory employment period . . . .” (Rhode Island Hospitality Assn. v. City of Providence (D.R.I. 2011) 775 F.Supp.2d 416, 432 (Rhode Island Hospitality).)2 3(Maj. opn., ante, at pp. 204-205.) In other words, because the ordinance does not in so many words require the employer to maintain the predecessor’s workforce after 90 days, the ordinance does not dictate the outcome of the successorship inquiry (maj. opn., ante, at p. 205) (and therefore does not intrude on the processes of organization, collective bargaining, or labor disputes).
I cannot agree with this analysis.
First, it is entirely uncertain that the obligation to bargain will not arise unless the new employer voluntarily retains the workforce after the 90 days. As the Rhode Island Hospitality court acknowledged, the two NLRB decisions on the issue are not of a single mind (though both concluded the new employer under a mandatory retention ordinance was a successor with an obligation to bargain). One of them rejected the new employer’s contention that it was forced to hire its predecessor’s employees (and therefore did not “consciously decide[] to take advantage of its predecessor’s trained workforce”); the ALJ observed that the NLRB had “never formally adopted a requirement that a successor employer must consciously decide to avail itself of its predecessor’s trained workforce in order to be considered a Burns’ successor employer . . . .” (U.S. Service Industries, Inc. (NLRB, Dec. 13, 1995, No. 5-CA-24575) 1995 N.L.R.B. Lexis 1151, pp. *11-*12 (Service Industries).)3 In the other case, the ALJ found that successorship obligations attach on the date the new employer makes offers of permanent employment *227to the workers (or, if no offers are made, at a reasonable time after the expiration of the 90-day period). (M&M Parkside Towers LLC (NLRB, Jan. 30, 2007, No. 29-CA-27720) 2007 N.L.R.B. Lexis 27, pp. *15—*17 (M&M Parkside).)4
This uncertainty as to when the obligation to bargain attaches, it seems to me, itself demonstrates that the ordinance impermissibly intrudes on the processes of self-organization and collective bargaining that are the exclusive province of the NLRA. The fact is that under the current state of the law, a “[successor [gjrocery [e]mployer” (defined in the ordinance) (L.A. Mun. Code, § 181.01, subd. I, boldface omitted) in a unionized workplace would not know whether it is obliged to bargain, at the request of the union representing the employees it was forced to hire, during that 90-day retention period. The new employer is, at a minimum, exposed to charges of an unfair labor practice if it refuses to bargain.
An obvious illustration of this point would be the case of an employee who files a grievance with the union because he was fired within the first 90 days. The right to continue employment unless there is good cause for termination is a significant benefit ordinarily conferred, if at all, only after the employer and union have negotiated all other terms of employment, and the union has made concessions to offset the employer’s “termination for cause only” agreement. The ordinance bestows this benefit upon a unit of employees whom the successor has not chosen to hire. The successor did not obtain *228concessions from the predecessor in exchange for the promise to hire the predecessor’s employees, nor did the successor exercise its right to choose which employees to hire from among the predecessor’s workforce. Yet if the successor were to fire any employee in the first 90 days (or any time thereafter), it is likely the employee will file a grievance, and if the successor refuses to go through grievance procedures under the predecessor’s collective bargaining agreement, the successor risks being found guilty of an unfair labor practice. If the successor were to implement new wages, hours, and benefits without bargaining with the union, the employees might file grievances, demand union negotiation and arbitration, and even walk out or call a strike.
Of course no one can foresee what will occur in any particular case. But the exposure to unfair labor practice charges caused by the ordinance illustrates its intrusion into an area reserved for market freedom. Machinists tells us that, just as the state may not prohibit the use of economic weapons such as strikes, lockouts and other “self-help economic activities,” the state “may not . . . ‘add to an employer’s federal legal obligations in collective bargaining’ . . . .” (Machinists, supra, 427 U.S. at p. 147.) It seems to me that the City’s ordinance does just that, effectively “ ‘upset[ting] the balance that Congress has struck between labor and management in the collective-bargaining relationship.’ ” (Metropolitan Life, supra, 471 U.S. at p. 751.)
The ordinance does not on its face prevent the new employer, after 90 days, from terminating the workers it was forced to employ and hiring its own independent workforce. But, once an employment relationship is established, it is not easy to terminate, certainly not without substantial risk. For one thing, it is highly impractical to terminate an entire workforce at once, and virtually impossible to do so without business disruption and damage to the new employer’s reputation in the community. For another, it is easy to imagine the litigation consequences that would ensue if the employees in a previously unionized workplace are terminated en masse after working for the new employer for 90 days. The likelihood of unfair labor practice charges alleging union animus would be high, as would the likelihood of wrongful termination claims. The majority finds that similar consequences, at least in terms of unfair labor practice charges, would ensue if there were no ordinance and the new employer were free to hire an entirely new workforce at the time of acquisition. I disagree. We need not blind ourselves to probable consequences that ordinary experience suggests will occur.
Moreover, it seems obvious that a new employer free to hire as it chooses before going into business does not have to discharge anyone, much less terminate en masse, in order to employ the people it wants to conduct its business, so its exposure to claims necessarily will be significantly different. *229In effect, the City’s ordinance requires a new grocery employer to step into a morass of litigation—and to function during the important initial period of its operations with a workforce it deems, for entirely legitimate reasons, unsuitable for its planned operations—if it wishes to exercise its right, recognized under federal labor policy, to hire its own independent workforce. (Indeed, this impact was demonstrated at trial, at least in part, by the evidence that, since the City’s ordinance was enacted, three different grocers have declined to buy stores in Los Angeles because of the importance to them of choosing their workforce.) This is not a result within the contemplation of the NLRA.
In short, the facially temporary nature of the ordinance’s interference with an employer’s right to select its workforce does nothing, in my view, to change the core facts. It is simply unrealistic to believe the temporary nature of the ordinance will not materially affect the successorship inquiry. In reality, whether the obligation to recognize the union ripens on day one, or on day 91, or on some later date, may depend on the various circumstances of each successor grocer’s workplace. But there is little room for doubt that, under the law as applied by the NLRB in the decades since John Wiley and Burns, many successor grocers will be deemed obligated to recognize and bargain with the union at some point, as a direct consequence of complying with the ordinance.
Thus, under any of the scenarios that may occur under the ordinance, there is an impact, to a greater or lesser degree, on collective bargaining: either the new employer is a successor who must bargain with the union during the 90-day period and/or shortly thereafter, or the employer must terminate employees en masse after the 90-day period, exposing it to business consequences and legal claims that would otherwise not exist, if it wishes to choose its own workforce. The ordinance legislates in an area that federal labor law, clearly articulated by the high court, left unregulated: “the normal exercise of the right of the employer to select its employees” (Jones & Laughlin, supra, 301 U.S. at p. 45), “ ‘the rightful prerogative of owners independently to rearrange their businesses’ ” (Burns, supra, 406 U.S. at p. 301, quoting John Wiley, supra, 376 U.S. at p. 549), and a new employer’s “right not to hire any of the former [employer’s] employees, if it so desire[s]” (Howard, Johnson, supra, 417 U.S. at p. 262). To me, the ineluctable conclusion is that the ordinance is preempted under Machinists because it regulates “within ‘a zone protected and reserved for market freedom.’ ” (Brown, supra, 554 U.S. at p. 66.)
5. Summary and conclusion
The high court in Bums identified the freedom of contract as a fundamental premise on which the NLRA is based—“ ‘private bargaining under *230governmental supervision of the procedure alone, without any official compulsion over the actual terms of the contract.’ ” (Burns, supra, 406 U.S. at p. 287, quoting Porter Co. v. NLRB (1970) 397 U.S. 99, 108 [25 L.Ed.2d 146, 90 S.Ct. 821].) In the nearly 40 years since Burns was decided, the high court has continued to recognize the freedom of every employer and union to reach agreement, or not, in their negotiation of the terms and conditions of employment. This case presents a very different, far more fundamental issue than the freedom of parties who choose to work together to agree on the terms of employment.
The very different question presented here is whether the NLRA evinces an intent to prohibit regulations that interfere with an employer’s freedom to choose whether to enter into any employment relationship at all with a particular employee or bloc of employees. In my view, the absence of high court precedent saying in so many words that the intent of the NLRA is to preserve the freedom of choice for employees and employers alike to decide whether to form an employment relationship does not mean states and municipalities are free to enact so-called worker “retention” ordinances. Rather, the high court has not yet had occasion to address this truly extraordinary intrusion upon the freedom of contract, one that threatens to upend the very foundation of our national labor laws and policies. Perhaps this case will offer the high court the occasion to address the question now.
Associate Justice of the Court of Appeal, Second Appellate District, Division Eight, assigned by the Chief Justice pursuant to article VI, section 6 of the California Constitution.
In Washington Service Contractors v. Dist. of Columbia (D.C. Cir. 1995) 311 U.S. App.D.C. 407 [54 F.3d 811, 816], the court, upholding a retention ordinance of unlimited duration against a preemption claim, stated that this argument contains a “logical flaw,” because the NLRB may decide not to require the new employer to bargain with the union in circumstances where a statute or ordinance requires the employer to retain its predecessor’s employees. According to the court, if the NLRB decides not to require bargaining, then the problem disappears and there is no intrusion in the bargaining process. And if the NLRB does require bargaining, that would mean that in the NLRB’s judgment, the ordinance is congruent with the aims of the NLRA. (Washington Service Contractors, 54 F.3d at pp. 816-817.) In my view, both scenarios miss the mark. The first (not requiring bargaining) would intrude on the employees’ rights to representatives of their own choosing, and the second presumes that the NLRB’s judgment would presage that of the high court, a conclusion I find to be unwarranted. In either case, the ordinance has an impact on, and “ 'upset[s] the balance that Congress has struck between labor and management in the collective-bargaining relationship.’ ” (Metropolitan Life, supra, 471 U.S. at p. 751.)
In Rhode Island Hospitality, the federal district court upheld the enforcement of a municipal ordinance similar to the one challenged here. The ordinance required a new employer in the hospitality business to retain the previous employer’s workers for three months, subject to a “good cause” right to discharge. (Rhode Island Hospitality, supra, 775 F.Supp.2d at p. 423.) The court acknowledged that the NLRB had “not yet developed a consistent position” on when the duty to bargain would accrue when there is a mandatory retention ordinance (id. at p. 432), that “[t]his is a close case” (ibid.), and that the ordinance could not be “simply characterized as a ‘minimum labor standard.’ ” (Id. at p. 433.) But, because the ordinance did not preclude an employer from making its own hiring decisions after 90 days, did not compel a successor employer to honor the terms of a collective bargaining agreement negotiated by its predecessor, and allowed the new employer to set employment terms (ibid.), the district court concluded that the ordinance was “primarily designed to provide temporary job protection to both unionized and nonunionized employees which does not constitute a significant intrusion into the equitable collective bargaining process established by the NLRA.” (Ibid.)
In Service Industries, a union filed a charge with the NLRB and a complaint was issued alleging the employer refused to recognize and bargain with the union. The employer’s predecessor provided janitorial services to buildings in the District of Columbia and had recognized the union as the exclusive bargaining representative of its janitorial employees. The predecessor lost its contract to clean a particular building. The new employer contracted with *227the building’s management, and hired 24 janitorial employees, 15 of whom had been employed by the predecessor. (Service Industries, supra, 1995 N.L.R.B. Lexis 1151 at pp. *8-*9.) The union requested bargaining on the terms and conditions of employment, but the new employer refused. (Id. at p. *10.) The ALJ found substantial continuity in the employing enterprises, but the new employer contended that it was forced to hire its predecessor’s employees by the district’s Displaced Workers Protection Act. (1995 N.L.R.B. Lexis 115 at p. *11.) The ALJ rejected the argument, and found the new employer’s refusal to recognize and bargain with the union was an unfair labor practice. (Id. at p. *14.)
M&M Parkside, decided 11 years after Service Industries, involved a New York City ordinance requiring a purchaser of apartment buildings and commercial property to retain all employees for 90 days (and to offer permanent employment to employees who received a satisfactory written evaluation after 90 days). (M&M Parkside, supra, 2007 N.L.R.B. Lexis 27 at pp. *3-*4.) After the 90-day period ended, the purchaser in M&M Parkside, while claiming not to be a successor, retained all the seller’s employees, established new terms and conditions of employment, and declined the union’s request to negotiate and execute a collective bargaining agreement. (Id. at pp. *8-*9.) In the ensuing proceeding alleging unfair labor practices, the ALJ found the new employer’s business was unchanged and the predecessor’s employees were the entirety of the workforce, so the new employer had a legal obligation to recognize and bargain with the union. (Id. at pp. *10-* 11.) The ALJ rejected the notion that the local ordinance compelling the employer to retain the employees for 90 days should mitigate against a successorship finding, viewing the argument as a nonsequitur, since the issue was whether the employer had an obligation to bargain after it chose, for whatever reason, to hire the predecessor’s employees. (Id. at pp. *12—*14.)