This direct appeal under the Expediting Act, 15 U.S.C. § 29, is taken by the United States from a judgment of the District Court for the District of New Jersey dismissing, after full hearing, the Government's complaint seeking to enjoin as a violation of § 7 of the Clayton Act, 15 U.S.C. § 18,1 the proposed merger of appellees, Phillipsburg National Bank and Trust Co. (PNB) and the Second National Bank of Phillipsburg (SNB), both located in Phillipsburg, New Jersey. The Comptroller of the Currency, also an appellee here, approved the merger in December 1967 and intervened in this action to defend it, as he was authorized to do by the Bank Merger Act of 1966, 12 U.S.C. § 1828(c)(7)(D) (1964 ed., Supp. V).2 The Bank Merger Act required that the District Court engage in a two-step process, United States v. First City National Bank of Houston, 386 U.S. 361, 87 S.Ct. 1088, 18 L.Ed.2d 151 (1967); United States v. Third National Bank in Nashville, 390 U.S. 171, 88 S.Ct. 882, 19 L.Ed.2d 1015 (1968), the first of which was to decide whether the merger would violate the antitrust prohibitions of § 7 of the Clayton Act. If the court found that § 7 would be violated, then the Bank Merger Act required that the District Court decide whether 'the anticompetitive effects of the proposed transaction are clearly outweighed in the public interest by the probable effect of the transaction in meeting the convenience and needs of the community to be served.' 12 U.S.C. § 1828(c)(5)(B). The District Court found that the United States 'failed to establish by a preponderance of the evidence that the proposed merger would have any anti-competitive effect and, further, that even if there were de minimis anticompetitive effect in the narrowly drawn market proposed by the government, such effect is clearly outweighed by the convenience and needs of the community to be served by the merged bank.' 306 F.Supp. 645, 667, (1969). We noted probable jurisdiction. 397 U.S. 933, 90 S.Ct. 945, 25 L.Ed.2d 114 (1970). We reverse. We have concluded from our examination of the record that the District Court erred in its definitions of the relevant product and geographic markets and that these errors invalidate the court's determination that the merger would have no significant anticompetitive effects.

* The Factual Setting

Phillipsburg is a small industrial city on the Delaware River in the southwestern corner of Warren County, New Jersey. Its population was 18,500 in 1960, 28,500 counting the population of its bordering suburbs. Although the population of the suburbs is and has been increasing, Phillipsburg itself has not grown. Easton, Pennsylvania, lies directly across the river. It had a population of 32,000 in 1960, 60,000 counting its bordering suburbs. Its population growth pattern has paralleled that of Phillipsburg. The cities are linked by two bridges and the testimony was that they are 'in effect * * * one town.'

This 'one town' has seven commercial banks, four in Easton and three in Phillipsburg. PNB and SNB are respectively the third and fifth largest in overall banking business. All seven fall within the category of small banks, their assets in 1967 ranging from $13,200,000 to $75,600,000.3 PNB, with assets then of approximately $23,900,000, and SNB with assets of approximately $17,300,000, are the first and second largest of the three Phillipsburg banks. The merger would produce a bank with assets of over $41,100,000, second in size of the six remaining commercial banks in 'one town.'

PNB and SNB are direct competitors. Their main offices are opposite one another on the same downtown street. SNB's only branch is across a suburban highway from one of PNB's two branches. Both banks offer the wide range of services and products available at commercial banks, including, for instance, demand deposits, COMPARISON OF PHILLIPSBURG—EASTON BANKS*

BANK No. of Amt. Phill Phill Amt. Phill Phill Amt. Phill Phill Amt. Phill Phill

Offices East. Only East. Only East. Only East. Only

Phillipsburg Nat.

Bank 3 $23.9 11.2 44.0 $22.4 13.7 44.3 $6.5 11.3 45 $14.5 15.8 48.9

Second Nat. Bank 2 17.3 8.1 31.8* 16.0 9.8 31.7 4.6 7.9 31.3 10.5 11.4 35.2

(Resulting Bank) 5 41.1 19.3 75.8 38.4 23.4 76.1 11.1 19.2 76.4 24.9 27.3 84.1

Co. 2 13.2 6.2 24.2 12.1 7.4 23.9 3.4 6.0 23.6 4.7 5.2 15.9

& Trust 5 75.6 35.5 — 67.7 41.4 — 25.4 44.1 32.6 35.7 —

Northampton Nat.

Bank 1 23.2 10.9 — 19.0 11.6 — 6.4 11.0 — 6.5 7.1 —

Lafayette Trust Bank 2 27.7 13.0 — 24.3 14.8 — 10.8 18.8 — 11.8 12.9 —

Nazareth Nat. Bank 1 32.0 15.0 — 2.3 1.4 — 0.5 0.9 — 10.8 11.9 —

Totals 16 $212.8 100 100 $163.7 100 100 $57.7 100 100 $91.5 100 100 savings and time deposits, consumer loans, commercial and industrial loans, real estate mortgages, trust services, safe deposit boxes and escrow services. As is characteristic of banks of their size operating in small communities, PNB and SNB have less of their assets in commercial and industrial loans than do larger banks. They emphasize real estate loans and mortgages, and they have relatively more time and savings deposits than demand deposits. Similarly, their trust assets are quite small. In short, both banks are oriented toward the needs of small depositors and small borrowers. Thus, in 1967 75% of PNB's number of deposits and 73% of SNB's were $1,000 or less; 98% of PNB's number of deposits and 97% of SNB's were $10,000 or less. Similarly, 75% of PNB's number of loan accounts and 59% of SNB's were $2,500 or less, and 93% and 87% respectively were $10,000 or less.

Both banks serve predominantly Phillipsburg residents. In 1967, although 91.6% of PNB's and 92% of SNB's depositors were residents of 'one town,' only 5.3% of PNB's and 9% of SNB's depositors lived in Easton. And, although 78.6% of PNB's and 87.2% of SNB's number of loans were made to residents of 'one town,' only 14.8% and 11.6% respectively went to persons living in Easton. A witness testified that all of the approximately 8,500 Phillipsburg families deal with one or another of the three commercial banks in that city. The town's businessmen prefer to do the same. The preference for local banks was strikingly evidenced by the fact that PNB and SNB substantially increased their savings deposit accounts during 1962—1967, even though their passbook savings rates were lower than those being paid by other readily accessible banks. At a time when Phillipsburg banks were paying 3.5% interest and Easton banks only 3%, other banks within a 13-mile radius were offering 4%.

Phillipsburg-Easton is in the northeastern part of the Lehigh Valley, a region of approximately 1,000 square miles, with a population of 492,000 in 1960 and 38 commercial banks in June 1968. There is considerable mobility among residents of the area for social, shopping, and employment purposes. Customer preference and conservative banking practices, however, have tended to limit the bulk of each commercial bank's business to its immediate geographic area. Neither PNB nor SNB has aggressively sought business outside 'one town.' Similarly, most other banks in the Lehigh Valley have shown little interest in seeking customers in Phillipsburg-Easton. The District Court found that '(t) here is an attitude of complacency on the part of many banks (in the Valley). They are content to continue outmoded banking practice and reluctant to risk changes which would improve service and extend services over a greater area to a larger segment of the population.' 306 F.Supp., at 661.

The merger would reduce the number of commercial banks in 'one town' from seven to six, and from three to two in Phillipsburg. The merged bank would have five of the seven banking offices in Phillipsburg and its environs and would be three times as large as the other Phillipsburg bank; it would have 75.8% of the city's banking assets, 76.1% of its deposits, and 84.1% of its loans. Within Phillipsburg-Easton PNB—SNB would become the second largest commercial bank, having 19.3% of the total assets, 23.4% of total deposits, 19.2% of demand deposits, and 27.3% of total loans. This increased concentration would give the two largest banks 54.8% of the 'one town' banking assets, 64.8% of its total deposits, 63.3% of demand deposits, 63% of total loans, and 10 of the 16 banking offices.

We entertain no doubt that this factual pattern requires a determination whether the merger passes muster under the antitrust standards of United States v. Philadelphia National Bank, 374 U.S. 321, 83 S.Ct. 1715, 10 L.Ed.2d 915 (1963), which were preserved in the Bank Merger Act of 1966. United States v. First National Bank of Houston, supra; United States v. Third National Bank in Nashville, supra. Mergers of directly competing small commercial banks in small communities, no less than those of large banks in large communities, are subject to scrutiny under these standards. Indeed, competitive commercial banks, with their cluster of products and services, play a particularly significant role in a small community unable to support a large variety of alternative financial institutions. Thus, if anything, it is even more true in the small town than in the large city that 'if the businessman is denied credit because his banking alternatives have been eliminated by mergers, the whole edifice of an entrepreneurial system is threatened; if the costs of banking services and credit are allowed to become excessive by the absence of competitive pressures, virtually all costs, in our credit economy, will be affected * * *.' Philadelphia Bank, 374 U.S., at 372, 83 S.Ct., at 1746.

When PNB and SNB sought the Comptroller's approval of their merger, as required by the Bank Merger Act, 12 U.S.C. § 1828(c), independent reports on the competitive factors involved were obtained, as required by § 1828(c)(4), from the Federal Reserve Board, the Federal Deposit Insurance Corporation, and the Attorney General. All three viewed the problem as involving commercial banking in Phillipsburg-Easton and reported that the merger would have a significantly harmful effect upon competition in that area. The Comptroller nevertheless approved the merger, finding that the agencies had defined the product and geographic markets too narrowly. He treated not Phillipsburg-Easton but most of the Lehigh Valley as the relevant geographic area, and evaluated competition from 34 finance companies and 13 savings and loan institutions, as well as from the more than 30 commercial banks in the area. The Comptroller concluded that the merger would have no significant anticompetitive effect and, further, that it would enable the resultant bank to serve more effectively the convenience and needs of the community.

In Philadelphia Bank we said that the 'cluster of products (various kinds of credit) and services (such as checking accounts and trust administration) denoted by the term 'commercial banking' * * * composes a distinct line of commerce.' 374 U.S., at 356, 83 S.Ct., at 1737. As indicated, PNB and SNB offer the wide range of products and services customarily provided by commercial banks. The District Court made no contrary finding, and, in its actual evaluation of the effect of the merger upon competition, the court looked only to commercial banking as the relevant product market. See 306 F.Supp., at 655—661.

Earlier in its opinion, however, the District Court appeared to reject commercial banking as the appropriate line of commerce. Rather than focusing its attention upon the effect of the merger in diminishing competition among commercial banks, the court emphasized the competition between PNB—SNB and other types of financial institutions—for example, savings and loan associations, pension funds, mutual funds, insurance, and finance companies. The court expressed its view that '(i)n terms of function the defendant banks are more comparable to savings institutions than to large commercial banks,' 306 F.Supp., at 648, and continued: 'So, while the term 'commercial banking' may be used to designate the general line of commerce embracting all bank services, attention must be given in analysis of competition to different groupings within the line of commerce separating those products and services where absence of competition may be significant from those in which competition from many sources is so widespread that no question of significant diminution of competition by the merger could be raised.' 306 F.Supp., at 650—651.

The District Court erred. It is true, of course, that the relevant product market is determine by the nature of the commercial entities involved and by the nature of the competition that they face. See, e.g., United States v. Continental Can Co., 378 U.S. 441, 456—457, 84 S.Ct. 1738, 1746—1747, 12 L.Ed.2d 953 (1964). Submarkets such as the District Court defined would be clearly relevant, for example, in analyzing the effect on competition of a merger between a commercial bank and another type of financial institution. But submarkets are not a basis for the disregard of a broader line of commerce that has economic significance. See, e.g., Brown Shoe Co. v. United States, 370 U.S. 294, 326, 82 S.Ct. 1502, 1524, 8 L.Ed.2d 510 (1962).

Philadelphia Bank emphasized that it is the cluster of products and services that full-service banks offer that as a matter of trade reality makes commercial banking a distinct line of commerce. Commercial banks are the only financial institutions in which a wide variety of financial products and services—some unique to commercial banking and others not—are gathered together in one place. The clustering of financial products and services in banks facilitates convenient access to them for all banking customers. For some customers, full-service banking makes possible access to certain products or services that would otherwise be unavailable to them; the customer without significant collateral, for example, who has patronized a particular bank for a variety of financial products and services is more likely to be able to obtain a loan from that bank than from a specialty financial institution to which he turns simply to borrow money. In short, the cluster of products and services termed commercial banking has economic significance well beyond the various products and services involved.4

Customers of small banks need and use this cluster of services and products no less than customers of large banks. A customer who uses one service usually looks to his bank for others as well, and is encouraged by the bank to do so. Thus, as was the case here, customers are likely to maintain checking and savings accounts in the same local bank even when higher savings interest is available elsewhere. See also Philadelphia Bank, supra, 374 U.S., at 357 n. 34, 83 S.Ct., at 1738. This is perhaps particularly true of banks patronized principally by small depositors and borrowers for whom the convenience of one-stop banking and the advantages of a good relationship with the local banker—and thus of favorable consideration for loans—are especially important. See id., at 358 n. 35, 369, 83 S.Ct., at 1738.

Moreover, if commercial banking were rejected as the line of commerce for banks with the same or similar ratios of business as those of the appellee banks, the effect would likely be to deny customers of small banks—and thus residents of many small towns the antitrust protection to which they are no less entitled than customers of large city banks. Indeed, the need for that protection may be greater in the small town since, as we have already stated, commercial banks offering full-service banking in one institution may be peculiarly significants to the economy of communities whose population is too small to support a large array of differentiated savings and credit businesses.

In determining the relevant geographic market, we held in Philadelphia Bank, supra, at 357, 83 S.Ct., at 1738, that '(t)he proper question to be asked * * * is not where the parties to the merger do business or even where they compete, but where, within the area of competitive overlap, the effect of the merger on competition will be direct and immediate. * * * This depends upon 'the geographic structure of suppliercustomer relations." More specifically we stated that 'the 'area of effective competition in the known line of commerce must be charted by careful selection of the market area in which the seller operates, and to which the purchaser can practicably turn for supplies. * * *" Id., at 359, 83 S.Ct., at 1739.

The District Court selected as the relevant geographic market an area approximately four times as large as Phillipsburg-Easton, with a 1960 population of 216,000 and 18 banks. The area included the city of Bethlehem, Pennsylvania. 306 F.Supp., at 652—653, 656 658. The court explicitly rejected the claim of the United States that Phillipsburg-Easton constitutes the relevant market. We hold that the District Court erred.

Commercial realities in the banking industry make clear that banks generally have a very localized business. We observed in Philadelphia Bank, supra, at 358, 83 S.Ct., at 1738, that '(i)n banking, as in most service industries, convenience of location is essential to effective competition. Individuals and corporations typically confer the bulk of their patronage on banks in their local community; they find it impractical to conduct their banking business at a distance. * * * The factor of inconvenience localizes banking competition as effectively as high transportation costs in other industries.' In locating 'the market area in which the seller operates,' it is important to consider the places from which it draws its business, the location of its offices, and where it seeks business. As indicated, the appellee banks' deposit and loan statistics show that in 1967 they drew over 85% of their business from the Phillipsburg-Easton area and, of that, only about 10% from Easton. It has been noted that nearly every family in Phillipsburg deals with one of the city's three banks, and the town's businessmen prefer to do the same. All of PNB and SNB's banking offices are located within Phillipsburg or its immediate suburbs; although the city is sufficiently small that there is easy access to its downtown area where the banks have their main offices, the banks found it necessary to open branches in the suburbs because, as a witness testified, that is 'where the customers are.' See also Philadelphia Bank, supra, at 358 n. 35, 83 S.Ct., at 1738. The 'one town' banks generally compete for deposits within a radius of only a few miles.

The localization of business typical of the banking industry is particularly pronounced when small customers are involved. We stated in Philadelphia Bank, supra, at 361, 83 S.Ct., at 1740, that 'in banking the relevant geographical market is a function of each separate customer's economic scale'—that 'the smaller the customer, the smaller is his banking market geographically,' id., at 359 n. 36, 83 S.Ct., at 1739. Small depositors have little reason to deal with a bank other than the one most geographically convenient to them. For such persons, geographic convenience can be a more powerful influence than the availability of a higher rate of interest at a more distant, though still nearby, bank. The small borrower, if he is to have his needs met, must often depend upon his community reputation and upon his relationship with the local banker. PNB, for instance, has made numerous unsecured loans on the basis of character, which are difficult for local borrowers to get elsewhere. And, as we said in Philadelphia Bank, supra, at 369, 83 S.Ct., at 1745, '(s)mall businessmen especially are, as a practical matter, confined to their locality for the satisfaction of their credit needs. * * * If the number of banks in the locality is reduced, the vigor of competition for filling the marginal small business borrower's needs is likely to diminish.' Thus, the small borrower frequently cannot 'practicably turn for supplies' outside his immediate community; and the small depositor—because of habit, custom, personal relationships, and, above all, convenience—is usually unwilling to do so. See id., at 357 n. 34, 83 S.Ct., at 1738. The patrons of PNB and SNB, of course, are small customers: almost 75% of the banks' deposits are for amounts less than $1,000, and virtually all of their loans are for less than $10,000, most falling below $2,500.

We observed in Philadelphia Bank, supra, at 361, 83 S.Ct., at 1740, that we were helped to our conclusion regarding geographic market 'by the fact that the three federal banking agencies regard the area in which banks have their offices as an 'area of effective competition." Here the Federal Reserve Board, the Federal Deposit Insurance Corporation, and the Attorney General found that a relevant banking market exists in the Phillipsburg-Easton area and that the proposed merger's competitive effect should be judged within it.5 We agree. We find that the evidence shows that Phillipsburg-Easton constitutes a geographic market in which the proposed merger's effect would be 'direct and immediate.' It is the market area in which PNB and SNB operate, and, as a practical matter, it is the area in which most of the merging banks' customers must, or will, do their banking. Thus, we hold that the District Court mistakenly rejected the Government's contention that Phillipsburg-Easton is an appropriate 'section of the country' under § 7.

Appellee banks argue that Phillipsburg-Easton 'cannot conceivably be considered a 'market' for antitrust purposes,' on the ground that it is not an 'economically significant section of the country.' They cite our language in Brown Shoe, supra, 370 U.S., at 320, 82 S.Ct., at 1521, that '(t)he deletion of the word 'community' in the original (Clayton) Act's description of the relevant geographic market is another illustration of Congress' desire to indicate that its concern was with the adverse effects of a given merger on competition only in an economically significant 'section' of the country.' In Brown Shoe, however, we found 'relevant geographic markets' in cities 'with a population exceeding 10,000 and their environs.' Id., at 339, 82 S.Ct., at 1531. Phillipsburg-Easton and their immediate environs had a population of almost 90,000 in 1960. Seven banks compete for their business. This market is clearly an economically significant section of the country for the purposes of § 7.

We turn now to the ultimate question under § 7: whether the effect of the proposed merger 'may be substantially to lessen competition.' We pointed out in Philadelphia Bank, supra, 374 U.S., at 362, 83 S.Ct., at 1741, that a prediction of anticompetitive effects 'is sound only if it is based upon a firm understanding of the structure of the relevant market; yet the relevant economic data are both complex and elusive. * * * And unless businessmen can assess the legal consequences of a merger with some confidence, sound business planning is retarded. * * * So also, we must be alert to the danger of subverting congressional intent by permitting a too-broad economic investigation. * * * And so in any case in which it is possible, without doing violence to the congressional objective embodied in § 7, to simplify the test of illegality, the courts ought to do so in the interest of sound and practical judicial administration.' We stated in Brown Shoe, supra, 370 U.S., at 315, 82 S.Ct., at 1518, that '(t)he dominant theme pervading congressional consideration of the 1950 amendments (to § 7) was a fear of what was considered to be a rising tide of economic concentration in the American economy.' In Philadelphia Bank, supra, 374 U.S., at 363, 83 S.Ct., at 1741, we held that '(t)his intense congressional concern with the trend toward concentration warrants dispensing, in certain cases, with elaborate proof of market structure, market behavior, or probable anticompetitive effects. Specifically, we think that a merger which produces a firm controlling an undue percentage share of the relevant market, and results in a significant increase in the concentration of firms in that market is so inherently likely to lessen competition substantially that it must be enjoined in the absence of evidence clearly showing that the merger in not likely to have such anticompetitive effects.' That principle is applicable to this case.

The commercial banking market in Phillipsburg-Easton is already concentrated. Of its seven banks, the two largest in 1967 Easton National Bank and Lafayette Trust Co.—had 49% of its total banking assets, 56% of its total deposits, 49% of its total loans and seven of its 16 banking offices. Easton National is itself the product of the merger of two smaller banks in 1959. The union of PNB—SNB would, in turn, significantly increase commercial banking concentration in 'one town.' The combined bank would become the second largest in the area, with assets of over $41,100,000 (19.3% of the area's assets), total deposits of $38,400,000 (23.4%), and total loans of $24,900,000 (27.3%). The assets held by the two largest banks would then increase from 49% to 55%, the deposits from 56% to 65%, the loans from 49% to 63%, and the banking offices from seven to 10. The assets held by the three largest banks would increase from 60% to 68%, the deposits from 70% to 80%, the loans from 64% to 76%, and the banking offices from 10 to 12. In Phillipsburg alone, of course, the impact would be much greater: banking alternatives would be reduced from three to two; the resultant bank would be three times larger than the only other remaining bank, and all but two of the banking offices in the city would be controlled by one firm. Thus, we find on this record that the proposed merger, if consummated, 'is * * * inherently likely to lessen competition substantially.' Cf. Philadelphia Bank, supra; Nashville Bank, supra; United States v. Von's Grocery Co., 384 U.S. 270, 86 S.Ct. 1478, 16 L.Ed.2d 555 (1966); United States v. Pabst Brewing Co., 384 U.S. 546, 86 S.Ct. 1665, 16 L.Ed.2d 765 (1966).

Appellee banks argue that they are presently so small that they lack the personnel and resources to serve their community effectively and to compete vigorously. Thus, they contend that the proposed merger could have procompetitive effects: by enhancing their competitive position, it would stimulate other small banks in the area to become more aggressive in meeting the needs of the area and it would enable PNB—SNB to meet an alleged competitive challenge from large, outside banks. Although such considerations are certainly relevant in determining the 'convenience and needs of the community' under the Bank Merger Act, they are not persuasive in the context of the Clayton Act. As we said in Philadelphia Bank, supra, 374 U.s., at 371, 83 S.Ct., at 1745, for the purposes of § 7, 'a merger the effect of which 'may be substantially to lessen competition' is not saved because, on some ultimate reckoning of social or economic debits and credits, it may be deemed beneficial.'

The District Court stated: 'Ease of access to the market is also a factor that deserves consideration in evaluating the anticompetitive effects of a merger. It is not difficult for a small group of business men to raise sufficient capital to establish a new small bank when the banking needs of the community are sufficient to warrant approval of the charter.' 306 F.Supp., at 659. Appellees, however, made no attempt to show that a group of businessmen would move to start a new bank in Phillipsburg-Easton, should the proposed merger be approved. The banking laws of New Jersey and Pennsylvania severely restrict the capacity of existing banks to establish operations in 'one town.' Relying on a recent New Jersey banking statute, N.J. Stat. Ann. § 17:9A—19 (Supp.1969), appellees contend that '(t)here is no doubt that the three banks in Phillipsburg * * * are fair game for attractive merger proposals by the large banks from Bergen, Passaic, Essex, Hudson and Morris Counties.' But, as the District Court pointed out, 'Large city banks in Newark and in other well populated cities in the counties mentioned can now establish branch banks in Warren County (only) in any municipality in which no banking institution has its principal office or a branch office and in any municipality which has a population of 7,500 or more where no banking institution has its principal office * * *.' 306 F.Supp., at 660. Thus, mergers under § 17:9A—19 are possible in Phillipsburg only with the three banks now in existence there. Accordingly, mergers under this statute would not bear upon the anticompetitive effects in question, because they could not increase the number of banking alternatives in 'one town.' Since the decision below, the Court of Appeals for the Third Circuit has held that a national bank may avoid the New Jersey bar against branching, N.J. Stat. Ann. § 17:9A—19(B)(3) (Supp.1969), by moving its headquarters into a protected community, such as Phillipsburg, while simultaneously reopening its former main office as a branch. Ramapo Bank v. Camp, 3 Cir., 425 F.2d 333 (1970). We intimate no view upon the correctness of that decision. We do observe, however, that the District Court decision in Ramapo Bank, affirmed in the recent Court of Appeals ruling, was handed down almost five months before the present District Court decision. Both opinions were written by the same District Judge. Accordingly, had an outside national bank been interested in moving its main office to Phillipsburg, no doubt this fact would have been made known to the District Court or to this Court. Nothing in the present record suggests that any national bank now located outside Phillipsburg will apply to move its main office to that city; therefore, on the record before us, that possibility does not bear on the anticompetitive effects of the merger.

Meeting the Convenience and Needs of the Community

The District Court's errors necessarily require reexamination of its conclusion that any anticompetitive effects caused by the proposed merger would be outweighed by the merger's contribution to the community's convenience and needs. The District Court's conclusion, moreover, is undermined by the court's erroneous application of the convenience-and-needs standard. In the balancing of competitive effect against benefit to community convenience and needs, '(t)o weigh adequately one of these factors against the other requires a proper conclusion as to each.' Nashville Bank, supra, 390 U.S., at 183, 88 S.Ct., at 890.

The District Court misapplied the convenience-and-needs standard by assessing the competitive effect of the proposed merger in the broad, multi-community area that it adopted as the relevant geographic market, while assessing the merger's contribution to community convenience and needs in Phillipsburg alone. Appellees argue that '(n)owhere does the district court equate 'community' with Phillipsburg.' We disagree. In determining convenience and needs, the court stated that '(t)here are two banking services which must be improved in the area to satisfy present and rapidly increasing need. Lending limits of the small banks are not sufficient to satisfy loan requirements for substantial industrial and commercial enterprise. * * * There is a definite lack of competent trust service and * * * servicing of substantial trust accounts must be obtained outside the community * * *. If the merger is approved, the merged bank can establish (loan and trust) departments and staff them with personnel capable of the kind of loan and trust service that patrons must, in large part, now seek outside the community.' 306 F.Supp., at 661. The court then cited examples of persons in Phillipsburg who found the existing loan and trust services in that city inadequate. Id., at 662—666. Since several Easton banks already provide appreciable trust services and have legal lending limits greater than those of PNB—SNB combined, it is obvious that the court was primarily concerned with loan and trust possibilities in Phillipsburg. We hold, however, that evaluation must be in terms of the convenience and needs of Phillipsburg-Easton as a whole.

Section 1828(c)(5)(B) provides that 'any * * * proposed merger transaction whose effect in any section of the country may be substantially to lessen competition * * * (shall not be approved by the responsible banking agency) unless it finds that the anticompetitive effects of the proposed transaction are clearly outweighed in the public interest by the probable effect of the transaction in meeting the convenience and needs of the community to be served.' Representative Reuss explained during debate on the Bank Merger Act that '(w)hat is meant by (§ (c)(5)(B)) and what counts is the effect of the transaction in meeting the needs and conveniences of the community which that particular sought-to-be merged bank serves.' 112 Cong.Rec. 2457. He indicated that 'in a community having say, 10 banks of relatively equal size, and where one of the banks was in difficulty—say with regard to a problem of management succession the 'convenience and needs of the community' would be best served if that bank were permitted to merge with one of the other 9 banks despite some resulting anti-competitive effects.' Id., at 2445.

These comments support our conclusion that the geographic market—the 'community which that particular sought-to-be merged bank serves'—is the area in which convenience and needs must be evaluated. Commercial realities, moreover, make clear that the 'community to be served' is virtually always as large, or larger, than the geographic market. Although the area in which merging banks compete while they are still separate entities is often smaller than the area in which the resultant bank will compete, it is rate that the community served by a merged bank is smaller than that served by its constituent firms prior to their merger. Further, evaluation of convenience and needs in an area smaller than the geographic market could result in the approval of a merger that, though it has anticompetitive effects throughout the market, has countervailing beneficial impact in only part of the market. Under the approach taken by the District Court, anticompetitive effects in some parts of a relevant geographic market could be justified by community benefits in other parts of it. Such a result would subvert the clear congressional purpose in the Bank Merger Act that convenience and needs not be assessed in only a part of the community to be served, and such a result would unfairly deny the benefits of the merger to some of those who sustain its direct and immediate anticompetitive effects.6 Cf. Philadelphia Bank, supra, 374 U.S., at 370, 83 S.Ct., at 1745. Accordingly, we hold that the District Court erred in failing to assess the proposed merger's effect in terms of the convenience and needs of the relevant geographic market.

We held in Nashville Bank, supra, 390 U.s., at 190, 88 S.Ct., at 893, that 'before a merger injurious to the public interest is approved, a showing (must) be made that the gain expected from the merger cannot reasonably be expected through other means.' Thus, before approving such a merger, a district court must 'reliably establish the unavailability of alternative solutions to the woes' faced by the merging banks. Ibid. Accordingly, on remand, the District Court should consider in concrete detail the adequacy of attempts by PNB and SNB to overcome their loan, trust, and personnel difficulties by methods short of their own merger. Beyond careful consideration of alternative methods of serving the convenience and needs of Phillipsburg-Easton, the court should deal specifically with whether the proposed merger is likely to benefit all seekers of banking services in the community, rather than simply those interested in large loan and trust services.

The judgment of the District Court is reversed and the case is remanded for further proceedings consistent with this opinion. No costs shall be assessed against appellee banks.

It is so ordered.

Judgment reversed and case remanded.

Mr. Justice STEWART took no part in the decision of this case, and Mr. Justice BLACKMUN took no part in its consideration or decision.

Mr. Justice HARLAN, with whom The CHIEF JUSTICE joins, concurring in part and dissenting in part.

My first reaction to this case, from the vantage point of what is depicted in the record and briefs, was wonderment that the Department of Justice had bothered to sue. How could that agency of government, I asked myself, be efficiently allocating its own scarce resources if it chose to attack a merger between two banks as small as those involved in this case? When compared with any of the 10 prior cases in which a bank merger was contested, the total assets of the bank that would result from this merger are minuscule.1 Moreover, measured by trust assets, the Phillipsburg National Bank in 1968 ranked 1346th and the Second National Bank of Phillipsburg 2429th out of the approximately 3100 banks with trust powers in the United States. If the two banks were merged, the resulting bank would have ranked 1323d—only 23 places ahead of the Phillipsburg National alone.2 With tigers still at large in our competitive jungle, why should the Department be taking aim at such small game?

The Court's disposition of this case provides justification enough from the Department's point of view. After today's opinion the legality of every merger of two directly competing banks—no matter how small—is placed in doubt if a court, through what has become an exercise in 'antitrust numerology,' United States v. First National Bank & Trust Co. of Lexington, 376 U.S. 665, 673, 84 S.Ct. 1033, 1037, 12 L.Ed.2d 1 (1964) (Harlan, J., dissenting), concludes that the merger 'produces a firm controlling an undue percentage share of the relevant market,' ante, at 366.

* Under the Bank Merger Act it is now settled that a court must engage in a two-step process in order to decide whether a proposed merger passes muster. First, the effect of the merger upon competition must be evaluated, applying the standards under § 7 of the Clayton Act, United States v. Third National Bank in Nashville, 390 U.S. 171, 181—183, 88 S.Ct. 882, 889—890, 19 L.Ed.2d 1015 (1968). If there would be a violation, the court must then proceed to decide whether 'the anticompetitive effects of the proposed transaction are clearly outweighed in the public interest by the probable effect of the transaction in meeting the convenience and needs of the community to be served.'3

For the first stage of the analysis, the Court appears to decide whether the effect of this proposed merger 'may be substantially to lessen competition' by the following process: First, the Court defines the relevant product market as commercial banking. Second, it defines the geographic market as Phillipsburg-Easton.4 The Court next calculates the percentage share of this market that would be held by the proposed merged bank,5 and the resulting changes in 'concentration,' as measured by the percent of market held by the two largest6 and three largest banks.7 It appears that from the magnitude of these figures alone, the Court concludes that the proposed merger would 'significantly increase commercial banking concentration' in an 'already concentrated' market.8 On the basis of the magnitude of these figures alone the Court concludes that this merger would violate § 7 of the Clayton Act.

I have voiced my disagreement before, particularly in the banking field, with the "numbers game' test for determining Clayton Act violations', United States v. Third National Bank, supra, at 193, 88 S.Ct., at 895 (Harlan, J., concurring in part and dissenting in part); see United States v. First National Bank, supra, 376 U.S., at 673, 84 S.Ct., at 1037 (Harlan, J., dissenting). Although I consider myself bound by the Court's decision in Philadelphia Bank, see United States v. Third National Bank, supra, 390 U.S., at 193, 88 S.Ct., at 895, I cannot concur in the simplistic way in which the Court applies the numbers test here.

Philadelphia Bank did not hold that all bank mergers resulting in an 'undue percentage share of the relevant market' and 'in a significant increase in the concentration of firms in that market,' 374 U.S., at 363, 83 S.Ct., at 1741, necessarily violated § 7 of the Clayton Act. Instead that case established a rule by which the percentage figures alone do no more than 'raise an inference,' id., at 365, 83 S.Ct., at 1742, that the merger will significantly lessen competition. Philadelphia Bank left room, however, for the merging companies to show that the 'merger is not likely to have such anti-competitive effects,' id., at 363, 83 S.Ct., at 1741. In short, under the Philadelphia Bank test, the percentage figures create a rebuttable presumption of illegality.

In this case there are two aspects of market structure, each largely ignored by the Court, that I think might well rebut the presumption raised by the percentage figures that the merger will have a significant effect on competition. Consequently, I think the appellees should on remand be given an opportunity to show by 'clear evidence' that despite the percentage figures, the anticompetitive effects of this merger are not significant.

The first of these aspects of the market structure concerns 'entry.' The percentage figures alone tell nothing about the conditions of entry in a particular market. New entry can, of course, quickly alleviate 'undue' concentration. And the possibility of entry can act as a substantial check on the market power of existing competitors.

Entry into banking is not simply governed by free market conditions, of course, for it is also limited by regulatory laws. When the complaint in this case was filed, entry into the Phillipsburg-Easton market was very much restricted by both the Pennsylvania and New Jersey banking statutes.9 However, a recent change in the New Jersey statute10 together with a new opinion of the Court of Appeals for the Third Circuit rendered since the trial in this case,11 appears to increase considerably the possibility of new entry into Phillipsburg. For the first time it may be possible for any national bank already operating anywhere in the northern region of New Jersey to open, under certain circumstances, a new office in Phillipsburg.12

If one assumes the regulatory barriers to entry have been permanently lowered, it would seem that the competitive significance of this merger may well be considerably overstated by the percentage figures alone. Certainly new entry into the market involved in this case would be both easier and of much greater competitive significance than in the Philadelphia Bank market. In a market dominated by banks of enormous absolute size, with assets of hundreds of millions and even billions of dollars, it is of course unlikely that a new entrant will quickly become a substantial competitive force. The same is not true, however, of a market in which the largest competitor is, in absolute terms, rather small.

In short, I think the significance of the percentage figures recited in the Court's opinion can only be fully evaluated after consideration of the present entry conditions in the Phillipsburg-Easton area. Because of the new developments in the New Jersey regulation of banking that have occurred since the trial of this case, I think it inexcusable of the majority not to give the appellee banks an opportunity on remand to demonstrate whether there is now a substantial possibility of new entry, and if so, what effect that possibility would have on the market power of the combined bank.13

Quite apart from entry, there is another aspect of the market structure relevant here that affects the significance of the percentage figures cited by the Court. Relying on Philadelphia Bank, the Court concludes that the 'cluster of products * * * and services * * * denoted by the term 'commercial banking' * * * composes a distinct line of commerce' for purposes of this case. The Court eschews all analysis of the composition of the products and services offered by appellee banks, however. The Court thus manages to ignore completely the extent to which competition from savings and loan companies, mutual savings banks, and other financial institutions that are not commercial banks affects the market power of the appellee banks.

A closer analysis of what the merging banks here do, plainly shows that they have more in common with savings and loan institutions and mutual savings banks than with the big city commercial banks considered in Philadelphia Bank. In particular, a much higher percentage of the total deposits of the banks here comes from savings accounts as opposed to demand deposits than is true of big city commercial banks.14 Moreover, a much larger proportion of the total loans of these small banks is in the form of real estate or personal loans as opposed to commercial loans.15 Savings and loan companies, savings banks, credit unions, etc., are of much greater competitive significance in this market than in the market analyzed in Philadelphia Bank. For in this market, these nonbank financial institutions offer close substitutes for the products and services that are most important to the appellee banks.

In choosing its product market, the Court largely ignores these subtleties and instead emphasizes the cluster of services and products which in the Court's words 'makes commercial banking a distinct line of commerce.' Because the Court does not explain why that combination has any substantial synergistic cf. Anderson's-Black Rock, Inc. v. Pavement Salvage Co., Inc., 396 U.S. 57, 61, 90 S.Ct. 305, 308, 24 L.Ed.2d 258 (1969), the Court's choice of a product market here can be seriously questioned. Certainly a more discriminating conclusion concerning the antitrust implication of this merger could be made if separate concentration percentages were calculated for each of the important products and services provided by appellee banks, and then an overall appraisal made of the effect of this merger on competition.

In any event, even assuming that for purposes of a preliminary analysis one were to use commercial banking as the line of commerce for the antitrust analysis—if only for the sake of convenience—that does not excuse the majority's failure to consider the competitive realities of the case in appraising the significance of the concentration percentages thus calculated, see United States v. First National Bank of Maryland, 310 F.Supp. 157, 175 (D.C.Md.1970). The bare percentages themselves are not affected by the presence or absence of significant competition for important bank products or services from firms outside commercial banking. By treating these percentages as no different from those found in Philadelphia Bank, the Court blithely assumes that percentages of the same order of magnitude represent the same degree of market power, irrespective of the amount of competition from neighboring markets.

Seen another way, the Court's mode of analysis makes too much turn on the all-or-nothing determination that the relevant product market either includes or does not include products and services of savings and loan companies, and other competition. A far better approach would be to recognize the fact that a product or geographic market is at best an approximation—necessary to calculate some percentage figures. In evaluating such figures, however, the Court should not decide the case simply by the magnitude of the numbers alone—it should give the appellees on remand an opportunity to demonstrate that the numbers here significantly 'overstate' the competitive offects of this merger because of the approximate nature of the assumptions underlying the Court's definition of the relevant market.

In short, I think that this case should be remanded to the District Court so that it might re-evaluate whether, in light of the entry conditions and existing competition from savings and loan and similar financial institutions, the merger can fairly be said to threaten a substantial loss of competition in the Phillipsburg-Easton area. Cf. White Motor Co. v. United States, 372 U.S. 253, 83 S.Ct. 696, 9 L.Ed.2d 738 (1963). If the District Court concludes that the merger would so threaten competition, it should then, in the manner the Court's opinion suggests, proceed to decide whether there are countervailing public interest advantages.