dissenting.
I agree that the judgment below should be remanded, but I do not agree that the remand should be limited to *586reshaping the decree. Because I believe that the definition of the relevant market here cannot be sustained, I would reverse and remand for a new determination of this basic issue, subject to proper standards.
We have here a case under both § 1 and § 2 of the Sherman Act, which proscribe combinations in restraint of trade, and monopolies and attempts to monopolize. The judicial task is not difficult to state: Does the record show a combination in restraint of trade or a monopoly or attempt to monopolize? If so, what are its characteristics, scope and effect? And, finally, what is the appropriate remedy for a court of equity to decree?
Each of these inquiries depends upon two basic referents: definition of the geographical area of trade or commerce restrained or monopolized, and of the products or services involved. In § 1 cases this problem ordinarily presents little difficulty because the combination in restraint of trade itself delineates the “market” with sufficient clarity to support the usual injunctive form of relief in those cases. See, e. g., United States v. Griffith, 334 U. S. 100. In the present case, however, the essence of the offense is monopolization, achieved or attempted, and the major relief is divestiture. For these purposes, “market” definition is of the essence, just as in § 7 cases 1 the kindred definition of the “line of commerce” is fundamental. We must define the area of commerce that is allegedly engrossed before we can determine its engrossment; and we must define it before a decree can be shaped to deal with the consequences of the monopoly, and to restore or produce competition. See United States v. du Pont & Co. (the Cellophane Case), 351 U. S. 377, *587389-396; United States v. Aluminum, Co. of America, 148 F. 2d 416 (C. A. 2d Cir. 1945).
In § 2 cases, the search for “the relevant market” must be undertaken and pursued with relentless clarity. It is, in essence, an economic task put to the uses of the law. Unless this task is well done, the results will be distorted in terms of the conclusion as to whether the law has been violated and what the decree should contain.
In this case, the relevant geographical and product markets have not been defined on the basis of the economic facts of the industry concerned. They have been tailored precisely to fit defendants’ business. The Government proposed and the trial court concluded that the relevant market is not the business of fire protection, or burglary protection, or protection against waterflow, etc., or all of these together. It is not even the business of furnishing these from a central location. It is the business, viewed nationally, of supplying “insurance accredited central station protection services” (CSPS)— that is, fire, burglary and other kinds of protection furnished from a central station which is accredited by insurance companies. The business of defendants fits neatly into the product and geographic market so defined. In fact, it comes close to filling the market so defined.2 This Court has now approved this Procrustean definition.
The geographical market is defined as nationwide. But the need and the service are intensely local — more local by far, for example, than the market which this Court found to be local in United States v. Philadelphia Nat. Bank, 374 U. S. 321, 357-362.3 The premises pro*588tected do not travel. They are fixed locations. They must be protected where they are. Protection must be provided on the spot. It must be furnished by local personnel able to bring help to the scene within minutes. Even the central stations can provide service only within a 25-mile radius. Where the tenants of the premises turn to central stations for this service, they must make their contracts locally with the central station and purchase their services from it on the basis of local conditions.
But because these defendants, the trial court found, are connected by stock ownership, interlocking management and some degree of national corporate direction, and because there is some national participation in selling as well as national financing, advertising, purchasing of equipment, and the like,4 the court concluded that the competitive area to be considered is national. This Court now affirms that conclusion.
This is a non sequitur. It is not permissible to seize upon the nationwide scope of defendants’ operation and to bootstrap a geographical definition of the market from this. The purpose of the search for the relevant geographical market is to find the area or areas to which a potential buyer may rationally look for the goods or services that he seeks. The test, as this Court said in United States v. Philadelphia Nat. Bank, is “the geographic structure of supplier-customer relations,” 374 U. S. 321, 357, quoting Kaysen & Turner, Antitrust Policy 102 (1959). And, as Mr. Justice Clark put it in Tampa Electric Co. v. Nashville Coal Co., 365 U. S. 320, 327, the definition of the relevant market requires *589“careful selection of the market area in which the seller operates, and to which the purchaser can practicably turn for supplies.” 5 The central issue is where does a potential buyer look for potential suppliers of the service— what is the geographical area in which the buyer has, or, in the absence of monopoly, would have, a real choice as to price and alternative facilities? This depends upon the facts of the market place, taking into account such economic factors as the distance over which supplies and services may be feasibly furnished, consistently with cost and functional efficiency.
The incidental aspects of defendants’ business which the court uses cannot control the outcome of this inquiry. They do not measure the market area in which buyer and sellers meet. They have little impact upon the ascertainment of the geographical areas in which the economic and legal questions must be answered: have defendants “monopolized” or “restrained” trade; have they eliminated or can they eliminate competitors or prevent or obstruct new entries into the business; have they controlled or can they control price for the services? 'These are the issues; and, in defendants’ business, a finding that the “relevant market” is national is nothing less than a studied failure to assess the effect of defendants’ position and practices in the light of the competition which exists, or could exist, in economically defined areas — in the real world.
Here, there can be no doubt that the correct geographic market is local. The services at issue are intensely local: they can be furnished only locally. The business as it is done is local — not nationwide. If, as might well be the case on this record, defendants were found to have violated the Sherman Act in a number of these local areas, a proper decree, directed to those markets, as well as to *590general corporate features relevant to the condemned practices, could be fashioned. On the other hand, a gross definition of the market as nationwide leads to a gross, nationwide decree which does not address itself to the realities of the market place. That is what happened here: The District Court’s finding that the market was nationwide logically led it to a decree which operated on the only national aspect of the situation, the parent company nexus, instead of on the economically realistic areas — the local situations. This Court now directs the trial court to require “some [unspecified] divestiture” locally by the alarm companies. This is a recognition of the economic reality that the relevant competitive areas are local. In plain terms, the Court’s direction to the trial court means a “market-by-market” analysis for the purpose of breaking up defendants’ monopoly position and creating competitors and competition wherever feasible in particular cities. In my view, however, by so directing, the Court implies that which it does not command: that the case should be reconsidered at the trial court level because of the improper standard it used to define the relevant geographic markets.
The trial court’s definition of the “product” market even more dramatically demonstrates that its action has been Procrustean — that it has tailored the market to the dimensions of the defendants. It recognizes that a person seeking protective services has many alternative sources. It lists “watchmen, watchdogs, automatic proprietary systems confined to one site, (often, but not always,) alarm systems connected with some local police or fire station, often unaccredited CSPS [central station protective services], and often accredited CSPS.” The court finds that even in the same city a single customer seeking protection for several premises may “exercise its option” differently for different locations. It may choose *591accredited CSPS for one of its locations and a different type of service for another.
But the court isolates from all of these alternatives only those services in which defendants engage. It eliminates all of the alternative sources despite its conscientious enumeration of them. Its definition of the “relevant market” is not merely confined to “central station” protective services, but to those central station protective services which are “accredited” by insurance companies.
There is no pretense that these furnish peculiar services for which there is no alternative in the market place, on either a price or a functional basis. The court relies solely upon its finding that the services offered by accredited central stations are of better quality, and upon its conclusion that the insurance companies tend to give “noticeably larger” discounts to policyholders who use accredited central station protective services. This Court now approves this strange red-haired, bearded, one-eyed man-with-a-limp classification.
The unreality of the trial court’s market definition may best be illustrated by an example. Consider the situation of a retail merchant in Pittsburgh who wishes to protect his store against burglary. The Holmes Electric Protective Company, a subsidiary of Grinnell, operates an accredited central station service in Pittsburgh. It provides only burglary protection.
The gerrymandered market definition approved today totally excludes from the market consideration of the availability in Pittsburgh of cheaper but somewhat less reliable local alarm systems, or of more expensive (although the expense is reduced by greater insurance discounts) watchman service, or even of unaccredited central station service which virtually duplicates the Holmes service.
Instead, and in the name of “commercial realities,” we are instructed that the “relevant market” — which totally *592excludes these locally available alternatives — requires us to look only to accredited central station service, and that we are to include in the “market” central stations which do not furnish burglary protection and even those which serve such places as Boston and Honolulu.6
Moreover, we are told that the “relevant market” must assume this strange and curious configuration despite evidence in the record and a finding of the trial court that “fringe competition” from such locally available alternatives as watchmen, local alarm systems, proprietary systems, and unaccredited central stations has, in at least 20 cities, forced the defendants to operate at a “loss” even though defendants have a total monopoly in these cities of the “market” — namely, the “accredited central station protective services.” And we are led to this odd result even though there is in the record abundant evidence that customers switch from one form of property protection to another, and not always in the direction of accredited central station service.
I believe this approach has no justification in economics, reason or law. It might be supportable if it were found that the accredited central stations offer services which are unique in the sense that potential buyers — or at least a substantial, identifiable part of the trade — look only to them for the services in question, and that neither cost, type, quality of service nor other factors bring competing services into the market. The findings here and the record do not permit this conclusion.
The Government’s market definition, accepted by the trial court, is a distortion which inevitably leads to a superficial and distorted result even in the hands of a highly skilled judge. As this Court held in Brown Shoe, supra, the “reasonable interchangeability of use or the *593cross-elasticity of demand,” determines the boundaries of a product market. 370 U. S., at 325. See also the Cellophane Case, 351 U. S., at 380. In plain language, this means that the court should have defined the relevant market here to include all services which, in light of geographical availability, price and use characteristics, are in realistic rivalry for all or some part of the business of furnishing protective services to premises. In the present situation, however, the court’s own findings show that practical alternatives are available to potential users— although they vary from market to market and possibly from user to user. These have been arbitrarily excluded from the court’s definition.
I do not suggest that wide disparities in quality, price and customer appeal could never affect the definition of the market. But this follows only where the disparities are so great that they create separate and distinct categories of buyers and sellers. The record here and the findings do not approach this standard. They fall far short of justifying the narrowing of the market as practiced here. I need refer only to the exclusion of non-accredited central stations, which the court seeks to justify by reference to differentials in insurance discounts. These differentials may indeed affect the relative cost to the consumer of the competing modes of protection. But, in the absence of proof that they result in eliminating the competing services from the category of those to which the purchaser “can practicably turn” for supplies,7 they do not justify such total exclusion. This sort of exclusion of the supposedly not-quite-so-attractive service from the basic definition of the kinds of business and service against which defendants’ activity will be measured, is entirely unjustified on this record.8
*594The importance of this kind of truncated market definition vividly appears if we are to say, as the trial court here held, that if defendant has so large a fraction of the market as to constitute a “predominant” share, a rebut-table presumption of monopolization follows. The fraction depends upon the denominator (the “market”) as well as the numerator (the defendants’ volume). Clearly, this “presumption” is unwarranted unless the “market” is defined to include all competitors. The contrary is not supported by this Court’s decisions in either the Cellophane Case, supra, or United States v. du Pont & Co. (General Motors), 353 U. S. 586. The latter case defined the market in terms of the total products which could be used for the defined purposes: automobile fabrics and finishes. This embraces the total range of options for customers seeking these products. On the contrary, as the record here shows and as the findings, candidly read, imply, substantial options exist for services other than through accredited central stations providing protective services. Those options, whether for all or a part of the services in issue, must be included in the assessment of the market.
In the opinion which this Court hands down today, there is considerable discussion of defendants’ argument that the market should be “broken down” by different *595type of service: e. g., burglar protection, fire protection, etc. The Court rejects this on the ground that it is appropriate to evaluate a “cluster” of services as such. It points to Philadelphia Nat. Bank, supra, for support for its approach. In that case, Mr. Justice Brennan’s opinion, for the Court carefully set out the distinctive characteristics of banking services: that some of these services (e. g., checking accounts) are virtually free of competition from other types of institutions, and that other services are distinctive in cost or other characteristics. 374 U. S., at 356-357. See also United States v. First Nat. Bank, 376 U. S. 665, 668 (per Douglas, J.). Similarly, in United States v. Paramount Pictures, 334 U. S. 131, and International Boxing Club v. United States, 358 U. S. 242, 249-252, “first-run” moving pictures and championship boxing matches were held sufficiently distinctive in terms of demand in the market place to warrant consideration as separate markets.
But no such distinctiveness exists here. As I have discussed, neither this record nor the trial court’s findings show either a distinctive demand or a separable market for “insurance accredited central station protective services.” The contrary is evident. None of the services furnished by accredited central stations is unique, as I have discussed. Nor is there even a common or predominant “cluster” of services offered by the central stations. One of the defendants, Holmes, is engaged only in the burglary alarm business. Another, AFA, furnishes only fire and waterflow service. Only ADT among the defendants makes available to its customers the full “cluster.”
I do not mean to suggest that the Government must prove its case, service by service. But in defining the market, individual services, even if furnished in isolation, ought to be specified and here, as distinguished from the conclusion impelled by the circumstances in *596Philadelphia Nat. Bank, supra, competitors for individual services ought to be taken into account.
I do not intend by any of the foregoing to suggest that, on this record, the relief granted by the trial court and the substantially more drastic relief ordered by this Court would necessarily be unjustified. It is entirely possible that monopoly or attempt to monopolize may be found— and perhaps found with greater force — in local.situations. Relief on a pervasive, system-wide, national basis might follow, as decreed by the trial court, as well as divestiture in appropriate local situations, as directed by this Court. It is impossible, I submit, to make these judgments on the findings before us because of the distortion due to an incorrect and unreal definition of the “relevant market.” Now, because of this Court’s mandate, the market-by-market inquiry must begin for purposes of the decree. But this should have been the foundation of judgment, not its superimposed conclusion. This inquiry should— in my opinion, it must — take into account the total economic situation — all of the options available to one seeking protection services. It should not be limited to central stations, and certainly not to “insurance accredited central station protective services” which this Court sanctions as the relevant market. Since I am of the opinion that defendants and the courts are entitled to a reappraisal of the liability consequences as well as the appropriate provisions of the decree on the basis of a sound definition of the market, I would reverse and remand for these purposes.
United States v. Continental Can Co., 378 U. S. 441, 447-458; United States v. Alcoa, 377 U. S. 271, 273-277; United States v. Philadelphia Nat. Bank, 374 U. S. 321, 356; Brown Shoe Co. v. United States, 370 U. S. 294, 324.
The defendants constitute 87% of the market as defined. One of the defendants alone, ADT, has 73%.
See also United States v. First Nat. Bank, 376 U. S. 665, 668 (per Douglas, J.); American Crystal Sugar Co. v. Cubara American Sugar Co., 152 F. Supp. 387, 398 (D. C. S. D. N. Y. 1957), aff’d, 259 F. 2d 524 (C. A. 2d Cir. 1958).
There is a danger that this Court’s opinion, ante, at 575-576, will be read as somewhat overstating the case. There is neither finding nor record to support the implication that rates are to any substantial extent fixed on a nationwide basis, or that there are nationwide contracts with multistate businesses in any significant degree, or that insurers inspect or certify central stations on a nationwide basis.
See also Brown Shoe Co. v. United States, 370 U. S. 294, 336-337.
None of the stations operated by defendant Automatic Fire Alarm Company offers burglary protection, just as none of Holmes’ stations protects against the risk of fire.
Tampa Electric Co. v. Nashville Coal Co., 365 U. S., at 327.
The example used by the court in its findings is illuminating and disturbing. In explanation of its narrow market definition, *594the court says that the difference between the accredited central station protective services and all others “could be compared” to the difference between a compact six-cylinder car and a chauffeur-driven sedan. It is probably true that the degree of direct competition between luxury automobiles and compacts is slight. But it is by no means as clear-cut as the trial court seems to suggest. The question would require careful analysis in light of the total facts and issues. For example, if the antitrust problem at hand involved an acquisition of the business of a manufacturer of compacts by a maker of luxury cars, it is by no means inconceivable that sufficient competitive overlap would be found to place both products in the “relevant market.”