United States v. Provident National Bank, 280 F. Supp. 1 (E.D. Pa. 1968)
February 12, 1968
PROVIDENT NATIONAL BANK and Central-Penn National Bank of Philadelphia, Defendants,
William B. Camp, Comptroller of the Currency, Intervenor.
United States District Court E. D. Pennsylvania.
*2 *3 John W. Neville, Julius J. Hollis, Arthur I. Cantor, John W. Clark, Dept. of Justice, Washington, D. C., Donald G. Balthis, Dept. of Justice, Drew J. T. O'Keefe, U. S. Atty., Philadelphia, Pa., for plaintiff.
Frederic L. Ballard, Charles I. Thompson, Jr., Tyson W. Coughlin, Matthew M. Strickler, Richard C. Bull, Philadelphia, Pa., for defendants; Ballard, Spahr, Andrews & Ingersoll, White & Williams, Philadelphia, Pa., of counsel.
Philip L. Roache, Jr., Joseph J. O'Malley, Charles H. McEnerney, Eugene J. Metzger, Gilbert L. Amyot, Office of The Comptroller of the Currency, Washington, D. C., for intervenor.
CLARY, Chief Judge.
This case involves a proposed merger between the Provident National Bank and the Central-Penn National Bank, both of the City of Philadelphia, Commonwealth of Pennsylvania. It is being challenged by the Department of Justice as being in violation of the antitrust laws of the United States. The case was tried to the Court without a jury. In making a decision in this case, the Court is faced with the dilemma so ably stated by that noted legal scholar, John H. Wigmore, the late Professor and Dean of Northwestern University School of Law, and author of "Wigmore on Evidence", that "law and justice are from time to time inevitably in conflict."
*4 The dilemma confronting the Court is that the Court firmly believes that a jury would have no difficulty in finding qualitatively that the proposed merger would be a good merger; good for the banks in that they would complement each other in both location and services; good for the community in that it would create another billion dollar bank in Philadelphia with its attendant advantages, particularly in the field of international finance; put real depth into the management of the surviving bank; generate more intense competition in the oligopolistic banking structure of Philadelphia, and might well make some improvement in the economic status of business in the Philadelphia area generally, and specifically its port facilities and operation. However, because of the quantitative (mechanical rules) approach to the question of antitrust violations declared as a policy by the Supreme Court and its decisions supporting that policy, this Court is constrained, albeit reluctantly, to declare that this merger may not be consummated.
On December 6, 1965, Provident National Bank (hereinafter Provident) and Central-Penn National Bank (hereinafter Central-Penn) filed with the Comptroller of the Currency an application to merge. Pursuant to Title 12 U.S.C. § 1828(c) (4), the Comptroller requested advisory reports from the other Government agencies, Federal Deposit Insurance Corporation (hereinafter F.D. I.C.) and Federal Reserve Board (hereinafter F.R.B.), and the Department of Justice (hereinafter Justice). The F.D. I.C. filed no report, but both F.R.B. and Justice reported that the merger's effect on competition would be significantly adverse and therefore should not be approved.
Prior to the Comptroller's decision on this merger, the Bank Merger Act of 1960 (hereinafter B.M.A.1960) was amended by Congress on February 21, 1966. The amended Act (hereinafter B.M.A.1966) created a new defense for anticompetitive mergers in that if the anticompetitive effects of the proposed merger 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, the merger is saved. Title 12 U.S.C. § 1828(c) (5) (B). The amended Act also deleted the words "line of commerce" which were present in B.M.A.1960, thus leaving open the question whether commercial banking is still to be considered as the area of effective competition. See United States v. First City National Bank of Houston, 386 U.S. 361, 369, f. n. 1, 87 S. Ct. 1088, 18 L. Ed. 2d 151 (1967).
On March 4, 1966, soon after B.M.A. 1966 was enacted into law, the Comptroller approved the proposed merger between Provident and Central-Penn. A formal decision containing the reasons for such approval followed on March 31, 1966.
Justice, on April 1, 1966, then instituted suit under Section 7 of the Clayton Act (15 U.S.C. § 18), alleging that the proposed merger was anticompetitive under traditional Section 7 standards and therefore should be proscribed. The Comptroller was allowed to intervene in this suit by virtue of Title 12 U.S.C. § 1828(c) (7) (D).
The suit was dismissed by this Court on December 29, 1966 on the basis that Justice chose to rely solely on Section 7 of the Clayton Act without regard to the new standards enunciated by B.M.A. 1966. United States v. Provident National Bank et al., 262 F. Supp. 397 (E. D.Pa.1966).
*5 The Supreme Court reversed this decision in United States v. First City National Bank of Houston, supra. The Court ruled that the Government's failure to rely on B.M.A.1966 was not a fatal defect in its pleadings since anticompetitive effects of such mergers were still to be assessed by traditional antitrust criteria. The only effect the amended Act had on these traditional criteria was that of creating a new defense or justification to a merger in the convenience and needs test previously mentioned. United States v. First City National Bank of Houston, supra, 386 U.S. at 363, 87 S. Ct. 1088. It also ruled that the burden of proving this new defense was upon the proponents of the merger the banks in this instance. The case was then remanded to this Court.
Before discussing the legal questions presented by this merger, it would be wise to have the respective positions and histories of the merging banks in view.
Provident is presently the fifth largest bank in the four-county area comprising Bucks, Montgomery, Delaware, and Philadelphia counties. As of June 30, 1965, Provident controlled approximately $683,000,000 or 9% of the total assets, $398,000,000 or 9% of the total loans, $475,000,000 or 9% of the total IPC deposits, and 32 or 9% of the banking offices in the four-county area.
Since 1947, Provident has acquired seven banks in the four-county area. As of April 5, 1966, these acquired banks' assets constituted 58% of Provident's present asset position.
As of June 30, 1965, Central-Penn, the seventh largest commercial bank in the four-county area, controlled approximately $369,000,000 or 5% of the total assets, $210,000,000 or 5% of the total loans, $263,000,000 or 5% of the total IPC deposits, $170,000,000 or 5% of the total IPC demand deposits, and 24 or 6% of the banking offices.
Central-Penn has acquired six commercial banks in the four-county area since 1950. No less than 24% of its present asset position is attributable to these mergers and acquisitions.
These facts must be viewed in the over-all setting of commercial banking in the four-county area. In 1945, there were 115 commercial banks in the four-county area. In 1960, there were 45. In 1965, the number of commercial banks was reduced to 37.
On December 31, 1945, the five largest commercial banks held 51% of the total assets, 54% of the total loans, 52% of the total deposits, and had 15% of the total banking offices. In 1965, the five largest banks controlled approximately 73% of the assets, 74% of the total loans, 71% of IPC deposits, and 57% of all banking offices. Thus, there has been a substantial reduction of independent commercial banks, most of which have disappeared through the merger route. The trend is continuing as evidenced by Girard Trust Bank's recent announcement of intent to merge with Doylestown National Bank, and Continental Bank's announcement of intent to merge with Sonsitaly Bank and Trust Company. Since these acquired banks represent only a small percentage of area assets, loans, and banking offices, it may be questioned whether these mergers will be challenged by Justice.
If this merger were approved, the resultant bank would become the fourth largest bank in the four-county area, controlling approximately 14% of the total assets, loans and deposits in the four-county area, and 15% of the commercial banking offices in the four-county area. The five bank concentration ratio previously mentioned will increase from 73.1% to 77.7% in terms of assets, 73.1% to 77.8% in terms of deposits, 73.9% to 78.6% in terms of loans, and from 57% to 63.3% in terms of total offices.
With these facts in view, we shall now turn to our first level of inquiry, which is whether the proposed merger between Provident and Central-Penn is anticompetitive, when measured by traditional Clayton Section 7 standards.
Before assessing the competitive effects of this merger, however, it is necessary to delineate a geographic market *6 within which to assess the impact of this merger.
THE RELEVANT GEOGRAPHIC MARKET
The four-county area of Bucks, Montgomery, Delaware, and Philadelphia, has been proposed by the Government as the relevant "section of the country" for this case. The Government's contention is grounded on the following facts: The defendant banks may not establish branch offices in any other area. The convenience factor localizes the area of competition, at least for retail customers. The Government also introduced statistics showing that by number, 95% of Provident's total deposit accounts are within the four-county area, and 98% of Central-Penn's total deposit accounts are located within the same area. In terms of dollar volume, 82% of Provident's total deposits are within the four-county area, while Central-Penn has 88% of its dollar volume for total deposits within the same area. So by number and by value, the four-county area appears to be a relevant section of the country. This evidence indicates to us that the four-county area is the most appropriate section of the country where the effect of the merger will be direct and immediate. United States v. Philadelphia National Bank, 374 U.S. 321, 357, 83 S. Ct. 1715, 10 L. Ed. 2d 915 (1963) (hereinafter PNB).
The defendant banks accede to the Government's geographic market since it is a permissible "section of the country" for purposes of B.M.A.1966. In so acceding, the banks feel that the case of United States v. Pabst Brewing Co., 384 U.S. 546, 86 S. Ct. 1665, 16 L. Ed. 2d 765 (1966) controls. This is undoubtedly correct. In Pabst, the Supreme Court did away with the requirement that Courts, or the Department of Justice, had to decide upon an appropriate geographic market as a preliminary step in discovering whether a merger was anticompetitive. The Government need prove only that there has been a merger between two corporations engaged in commerce, and that the merger may have a substantial anticompetitive effect somewhere in the United States. United States v. Pabst Brewing Co., supra, at 549, 86 S. Ct. 1665. It (the Department of Justice) is no longer forced to prove what constitutes the relevant economic or geographic market.
The Comptroller of the Currency disagrees with the assertion that the four-county area is a relevant geographic market. He contends that the Standard Metropolitan Statistical Area (SMSA), which comprises the three New Jersey counties of Gloucester, Camden, and Burlington, and the five Pennsylvania counties of Philadelphia, Delaware, Montgomery, Bucks, and Chester, is more relevant. The Comptroller states that he is not bound by the Supreme Court's ruling in PNB, supra, as to what constitutes the relevant geographic market on the basis that in this case, a different factual record has been developed. This is true, but the Comptroller has neglected to point to any compelling facts which would necessitate a change from the PNB standard. One of the mainstays of the Comptroller's argument is the Bureau of the Budget's definition of a metropolitan area as "an integrated economic and social unit with a recognized large population nucleus." (N.T. 3119). Yet, Frank A. Cisar, of the Bureau of the Budget, and Chairman of the Committee on Area Definitions, admitted upon cross-examination that the SMSA in itself cannot be equated with a marketing area, since no two market areas are the same. (N.T. 3127).
While it is true that many persons from the excluded counties of Chester, Burlington, Camden, and Gloucester, work and shop in the four-county area, the statistics introduced by the Comptroller proved only that the banking done in the four-county area is minimal.
*7 Therefore, we find that the four-county area is the appropriate section of the country for the purposes of this case. The only caveat to our finding that the four-county area is the appropriate geographic market is that within this geographic market, there must be a distinction made between wholesale accounts, which, in some cases, should be relegated to the national market, and retail accounts which essentially are relegated to the local community. This awareness of the so-called national market is not a new concept. The Supreme Court referred to such a market in PNB when it scaled down the Department of Justice's concentration figures by approximately 16% because of accounts which should have been allocated to the national, rather than the four-county market. PNB, supra, 374 U.S. at 364, 83 S. Ct. 1715.
This problem was also in the forefront when the House was considering the amended B.M.A. In the Report which accompanied the House Bill to the Senate, the concept of a national market was placed in clear view. Under the heading "The relevant market", the House Report stated:
"A recurrent bone of contention in the development of general antitrust law is the problem of defining the relevant market. Sometimes this is easy, as in the case of automobile manufacturers whose market is clearly national, or laundries whose market is clearly local. It is not so easy in the case of the larger banks, which serve a local market of individuals and small businesses, but which are also competing in the regional or national financial markets for the opportunity to place large loans with major corporations which are shopping the country for funds." 112 Cong.Rec. 2440 (Feb. 8, 1966).
Therefore, in finding the four-county area as the appropriate geographic market, we do so with an awareness that not all accounts found and assessed in this market are allocable to it. Some adjustment must and will be made when we discuss the effect of this dichotomy (i. e., national and local markets) under our "Anticompetitive Effects" heading.
THE RELEVANT PRODUCT MARKET
Our next level of inquiry before delving into the competitive aspects of this merger is the ascertainment of the appropriate line of commerce within which this merger is to be viewed.
Justice maintains that commercial banking is still the appropriate product market in which to assess this merger's competitive effects. In essence, it relies on the Supreme Court's ruling in PNB where the unique cluster of products and services offered by commercial banks was held sufficiently distinctive to constitute a relevant product market. However, since the PNB decision, as stated above, B.M.A.1960 has been amended and the phrase "line of commerce" has been deleted. The Supreme Court, in United States v. First City National Bank of Houston, supra, 386 U.S. at 369, f. n. 1, 87 S. Ct. 1088, expressly left open the question whether this omission was significant.
Two lower Courts have split on the significance of this omission. In United States v. Third National Bank of Nashville, 260 F. Supp. 869, 878, n. 5 (M.D. Tenn.1966), prob. juris. noted, 388 U.S. 905, 87 S. Ct. 2111, 18 L. Ed. 2d 1345 (1967), the lower Court noted that such an important change in established antitrust law cannot be presumed to have been made by a mere omission, and also cites the lack of any significant legislative history in support thereof. It held that commercial banking was still the appropriate "line of commerce".
Judge Zirpoli, in United States v. Crocker-Anglo National Bank, D.C.D.Cal. 277 F. Supp. 133, filed October 30, 1967, took the opposite view and ruled that a *8 broader test was intended by Congress in amending B.M.A.1960. Relying on the canons of statutory construction, and a statement by Senator Robertson in the legislative history as to the effect of the omission in the statutory language, the District Court in California construed a broader line of commerce which comprises all institutions competing either for the savings or investment dollar or for the extension of credit.
This Court is persuaded by Judge Zirpoli's lucid analysis of this vexing problem presently before this Court. We are convinced that Congress did intend a change to be effected by the omission. A thorough examination of the legislative history and the House Committee Report reveals that the effect of this omission was not discussed in the House, but Senator Robertson dealt with this omission on several occasions during the Senate debates.
"The text of paragraph (B) of the new bill follows the terms of section 1 of the Sherman Act and section 7 of the Clayton Act, with the exception that the reference to `any line of commerce' in the Clayton Act is not carried over into the new bill. In this respect the new bill resembles the Bank Merger Act of 1960, and calls for an appraisal of the overall effects of the merger on competition, weighing increases of competition in one field against decreases in competition in another field. The banking agencies and the courts, in other words, are not intended and are not permitted to select some single, perhaps minor aspect of the banks' business and to say that, because there is some lessening of competition in this element of the business, the overall effects of the merger the increase of competition in the entire field of banking and in the broader field of financial institutions which may result from other aspects of the merger are irrelevant and may not be considered." See Statements of Senator Robertson, 112 Cong.Rec., 2541, 2549-50, (Feb. 9, 1966).
In so broadening the line of commerce, this Court is very much aware of the uniqueness of commercial banks in that they are the only financial institutions permitted by law which can accept demand deposits and service them. It is clear from this unique function that certain other financial services necessarily follow, such as payroll and lock box services to the corporation customer. It is also evident that commercial banks are insulated for the most part from direct competition in the area of trust services, corporate pension funds, etc. However, the inter-relationship of all these services is not so tightly bound or localized to preclude meaningful and direct competition from other financial institutions within or without the four-county area.
It was admitted by Doctor Donald R. Hodgman, one of plaintiff's expert witnesses, upon cross-examination, that in the area of mortgages, loans and savings, there are other alternative, and sometimes better, sources than commercial banks, such as mutual savings banks and saving and loan associations. (N.T. 48-49).
It was also pointed out by several witnesses that the competition for the savings dollar has intensified among commercial banks, mutual savings banks and saving and loan associations in the four-county area, and the nation as a whole. This competition was evidenced particularly by the Certificate of Deposit see-saw which took place in the four-county area during the latter part of 1965 and 1966. This competition for the savings dollar is of particular significance since the deposit base of commercial banks has undergone a drastic change in recent years. In 1960, the proportion between demand and time deposits nationally was 70:30. Today, it is 50:50. Thus, one-half of a bank's deposits are subject to competition, not only from other commercial banks, but mutuals and saving and loan associations as well.
Since there is a reasonable interchangeability and meaningful competition for the savings dollar and mortgage loans, such competition from other financial *9 institutions must be considered in order to get a true competitive picture. The factor of convenience, which predominates the retail banking business, does not alter this conclusion, since the offices of these other financial institutions are as widespread and convenient as the commercial banking offices.
In extending the line of commerce only to these financial institutions (mutuals and saving and loan associations) because they alone offer direct and meaningful competition to commercial banks, our reasoning for so doing is buttressed by the Supreme Court's ruling in United States v. Continental Can Co. et al., 378 U.S. 441, 84 S. Ct. 1738, 12 L. Ed. 2d 953 (1964). In that case, the nation's second largest producer of cans desired to merge with the third largest producer of glass containers. There the Court found that Section 7 is not limited to identical products and lumped together the glass and container industries into one product market. It did so realizing the only actual competition between these two industries was primarily in the beer market, although potentially, they were competitors in broader markets. The Court emphasized cross-elasticity of demand and interchangeability of use as its reason for doing so. United States v. Continental Can Co. et al., supra, at 453, 84 S. Ct. 1738. This reason exists, to a pronounced degree, in the situation at hand, only the competition is more direct and meaningful. If such interchangeability can be used negatively in antitrust jurisprudence, surely the same concept can be used positively. The underlying question remains the same in each instance, i. e., to construct reasonable product markets in which to measure probable competitive effects realistically.
Therefore, since it is no longer true that commercial banks enjoy a "settled consumer preference" for the savings dollar. PNB, supra, 374 U.S. at 357, 83 S. Ct. 1715, the line of commerce must be broadened in this respect. Competition from life insurance companies, finance companies, lawyers for trust services, etc., should not be considered since the competition is not direct, and such a consideration would extend the product market to an unrealistic range.
Furthermore, this Court has heard no compelling evidence and has observed no noted changes from the PNB situation during the trial of this case in competition from these other sources to extend the product market any further.
THE ANTICOMPETITIVE EFFECTS OF THIS MERGER
We can now turn to the crucial question in this case, which is whether the proposed merger between Provident and Central-Penn is anticompetitive.
Various approaches have been suggested and used in judging whether a proposed merger is anticompetitive. The approach adopted by Justice in this case and accepted by the Courts traditionally is what is termed the "structural approach" to antitrust analysis. Given a certain industry structure, the structuralists contend that one can reasonably predict a kind of ultimate economic performance the industry is likely to turn in. This prediction is more valid when the composition of the industry remains static and new firms are not entering at a fast rate. [See generally Dixon, "Antitrust Policy: Some `Legal' and `Economic' Considerations", 14 U.C.L.A.L. Rev. 979 (1967); Mueller, "The New Antitrust: A `Structural' Approach", 12 Vill.L.Rev. 764 (1967)]. Adopting this structural position, some legal commentators have tried to discern at which concentration *10 level a merger should be outlawed.
The exact purpose of Section 7 is hard to pin down since the word "competition", under the statute, has been given various shades of meaning by the Courts and the Governmental agencies. The predominant view was that Congress, in enacting Section 7, was concerned with maintaining the "little guy" in business on the premise that the dispersion of economic power in the hands of many is salutary. [See generally Bok, "Section 7 of the Clayton Act and the Merging of Law and Economics", 74 Harv.L.Rev. 226 (1960)].
However, recent Supreme Court decisions in the merger area indicate that a pronounced shift in emphasis has occurred. Although Congress is still concerned with protecting small businesses from the overreaching grasps of corporate giants, it has extended its concern to the consumer. Thus, mergers are no longer considered solely as some kind of wrong against the small businessman, but also as a welfare loss to the consumer. United States v. El Paso Natural Gas Co., 376 U.S. 651, 84 S. Ct. 1044, 12 L. Ed. 2d 12 (1964); see also Dixon, supra, generally.
In contending that the proposed merger between Provident and Central-Penn is anticompetitive, Justice adopts a structural position pointing to the rising tide of concentration among commercial banks which is taking place in the four-county area. Basically, the approach of the Supreme Court and Justice has been one focusing on the structure of the respective markets. If the structure of the market examined is highly concentrated, Justice contends inferentially in positing the structure that the incentive for independent decisions and competitive behavior is destroyed. The advantages of such an approach are apparent. It lends itself to easy application since positing a certain market share or concentration ratio, anticompetitive effects can be affirmed or denied. It is clear to us that the antitrust standards utilized in merger cases have not been changed by BMA 1966 and that therefore the structural *11 analysis posited by Justice in this case is valid. But it doesn't necessarily follow that the structure to be examined by this standard must remain the same.
Our disagreement with Justice's analysis lies not with its initial approach, but on the data and statistics it has poured into this accepted structural analysis. We admit under current antitrust standards enunciated by the Supreme Court that if certain market shares are proved and are realistic, anticompetitive effects can be deduced from such shares. It has been stated by Justice that the merged bank would control 14% of the appropriate market, a share which must be proscribed. Additionally, it has been pointed out that the concentration ratio among the five largest banks in the four-county area will be increased from 73% to 78%. This factor, coupled with the entry barriers to commercial banking, especially in the Philadelphia market, will produce a highly oligopolistic banking structure, which will stifle competition and lead to a further reduction of individual competitors in the commercial banking market.
However, the Clayton Act and the Bank Merger Act are concerned with arresting mergers with anticompetitive effects. In so construing these Acts, this Court cannot close its eyes to competition from other sources or other areas which operate in the four-county area.
In shying away from the simplistic approach applied in PNB and other merger cases, we feel the need to do so is compelled by the uniqueness of commercial banking as an industry. A brief illustration might make this point more clearly.
Unlike two individual steel producers from different counties, or states, two individual commercial banks from the same counties do not always compete for the same retail or wholesale customer. This is so because of the convenience factor on the retail level and the differences in services offered by large and small commercial banks on the wholesale level. Thus, concentration ratios for non-banking industries are more realistic in that competition is usually more even and equal and all firms dealing in that one or several products are justifiably lumped together. This is not true with commercial banks. A case in point are the saving and loan associations and mutual savings banks, which offer the same product as a commercial bank in time and savings accounts, yet are excluded from Justice's market structure.
In order to get a true picture of the competition for the dollar in the four-county area, it is useful to break down commercial banking into two sub-products wholesale and retail banking. Wholesale banking is that which deals with a number of large accounts, primarily with large corporations, governmental bodies, financial institutions, and wealthy individuals, while retail banking caters "to the mass needs of the general public and small business." United States v. Manufacturers Hanover Trust Co., 240 F. Supp. 867, 896 (S.D.N.Y. 1965).
With this breakdown of commercial banking, it becomes clear why the Government's concentration ratios and market shares are invalid. We are sure the Government would not contend that the Bank of King of Prussia competes with First Pennsylvania Banking and Trust Company or Philadelphia National Bank for international or large corporate accounts. Yet, all of these figures and accounts are lumped together in one hodgepodge of percentages from which we are to deduce anticompetitive effects.
*12 This breakdown of commercial banking into subproducts does produce some analytical problems. The question invariably arises as to what accounts within the two banks before us fall into the wholesale market level and secondly, whether such wholesale accounts are subject to competition from without the four-county area.
We do not think it can be questioned that some accounts of a large commercial city bank should be placed in a national market. Put another way, it is clear that the corporations and individuals presently maintaining large accounts in both Provident and Central-Penn have alternative and convenient sources for deposits or loans outside the four-county area.
The defendant banks, in support of this national market for wholesale accounts, point us to a Federal Reserve System report which refers to the composition of the national credit market. While agreeing that not all corporate accounts are national in nature, the banks contend that a bulk of a corporation's large loans and related deposit balances can be relegated to this market. Also, some bank services are so specialized, i. e., international business and pension trust services, that they can be obtained only from large sophisticated banks in the national market. Such specialized or large accounts are sought eagerly by banks across the nation as evidenced by advertisements published by leading banks in periodicals of national circulation. (See defendants' Exhibit 49). Aside from this fact, we also heard direct testimony from corporate officials in the Philadelphia area that they bank in New York City, Boston, or elsewhere, to obtain unusually large loans or sophisticated and expert advice and are solicited by banks across the country for their banking business.
Upon arriving at a national market for certain accounts, the problem which then confronts us is just what accounts should be placed in the national market. Defendants assert that the following accounts should be included:
"(a) Demand deposits over $100,000;
(b) All time open accounts and corporate certificates of deposit;
(c) Savings accounts and certificates over $10,000;
(d) Time deposits of municipalities and correspondent banks over $100,000;
(e) Real estate loans over $100,000;
(f) All securities loans and loans to financial institutions;
(g) Commercial and industrial loans over $100,000;
(h) Consumer loans purchased from large dealers;
(i) Single payment loans over $25,000; and
(j) Other loans over $100,000."
Admittedly, the above accounts are substantially those adopted in United States v. Manufacturers Hanover Trust Co., supra, at 921. However, the same problem which existed in that case exists here to a greater degree. The defendant banks have given us the national market shares of Provident and Central-Penn, but they have not carried their analysis further to include the national market shares of the other large commercial banks in the four-county area.
"Determination of the local market shares of the constituent banks, however, is not a simple matter of subtracting what we have determined to be the percentage of their deposits and loans which are allocable to the national market from their total deposits and loans. Manifestly the true local market is the total deposits and loans of all the banks located in the metropolitan area, less the aggregate of their deposits and loans which originate in the national market. We cannot deduct all national business *13 from the area's total deposits and loans, however, without knowing the percentage of business which each local bank originates in the national market. Except for the constituent banks, there is no evidence before us on that subject." United States v. Manufacturers Hanover Trust Co., supra, at 924.
So even if we totally accept defendants' national market allocation with its resulting percentages of national business, we still find ourselves back in the position from which we started a search for realistic market shares or concentration ratios.
Faced with this insurmountable task of formulating realistic concentration ratios, and cognizant of the fact that many of these so-called national accounts are tied to the Philadelphia market by virtue of convenience, loyalty, etc., this Court finds that the most practicable solution to this problem is to adopt the "shading procedure" used by the Supreme Court in PNB.
Any other type of analysis would fall within the Supreme Court's proscription in PNB, supra, 374 U.S. at 360-361, 83 S.Ct. at 1740:
"Large borrowers and large depositors the record shows, may find it practical to do a large part of their banking business outside their home community; very small borrowers and depositors may, as a practical matter, be confined to bank offices in their immediate neighborhood; and customers of intermediate size, it would appear, deal with banks within an area intermediate between these extremes * * * But that in banking the relevant geographical market is a function of each separate customer's economic scale means simply that a workable compromise must be found: some fair intermediate delineation which avoids the indefensible extremes of drawing the market either so expansively as to make the effect of the merger upon competition seem insignificant, because only the very largest bank customers are taken into account in defining the market, or so narrowly as to place appellees in different markets, because only the smallest customers are considered."
Thus, in PNB the Supreme Court scaled down the aggregate share of the market held by PNB and Girard as a merged bank from 36-30%, or by approximately 16%. If we apply the same reduction formula between Provident and Central-Penn, we reduce their market share from 14% to 11.7%.
This reduction might seem somewhat arbitrary and non-realistic, but it is the fairest market analysis of which this Court can conceive. The defendants have asked us to include only the following statistics in our four-county analysis, based upon their breakdown of what accounts are national or local in scope. Of total deposits for Provident, approximately $245,000,000 out of $635,000,000 are attributed to the four-county area on the basis of the MacMahon breakdown adopted by the banks. Of total loans for Provident, approximately $86,000,000 of $427,000,000 are attributable to the four-county area. (See defendants' Exhibits 64 and 64A). On the Central-Penn side of the ledger, the banks ask this Court to include only approximately $152,000,000 of some $332,000,000 in deposits as emanating from the four-county area. They also state that only $79,000,000 of Central-Penn's $227,000,000 in total loans should be included in our four-county analysis, based again on the MacMahon criteria. (See defendants' Exhibits 46 and 46A).
For reasons stated before, a total acceptance of these proposed figures would be as unrealistic as the ones offered by Justice. Rather than becoming involved in a morass of detail and statistics, or involving ourselves in a construction of concentration ratios, which would be based on somewhat inadequate statistics and evidence, we believe the "shading procedure" adopted is the fairest and most realistic approach under the circumstances.
On the retail level, the deposit base examined must be expanded to take into *14 account saving and loan associations and mutual savings banks. Competition for the savings dollar is intense and emanates from these non-commercial bank sources, as well as from the commercial banks themselves. The retail business of the banks in question can be restricted to the four-county area realistically because such business is restricted by the factor of convenience. The product offered tends to be "relatively standardized in form, quality and cost." Since the two banks in question can only branch by law into the four-county area [Pa.Stat.Ann.Tit. 7, § 904 (1965)], there is no question that substantially all of its retail business takes place within these four counties. Unlike the competition for wholesale accounts, all commercial banks, plus saving and loan associations and mutual savings banks, compete for the savings dollar, and residential mortgages.
By branching, the large commercial banks in the city compete with the small individual suburban banks for the savings dollar and the demand deposit accounts. Specialized personnel, large lending limits, or reserve city bank status, are not essentials for this retail business. Thus, on the retail level, the figures offered by the Government are appropriate and realistic with the one exception that they should have included saving and loan associations and mutual savings banks.
The defendant banks, during the trial of this case and in their briefs, have attempted to prove that since retail accounts are sufficiently localized by the factor of convenience, there is not much competitive interplay between them. More specifically, they presented in evidence detailed maps illustrating the branch locations of both Provident and Central-Penn showing that retail customers are concentrated in proximity to the branches. While agreeing that a Central-Penn branch in Bucks County does not compete with a Provident branch in Montgomery County, they at least have the potentiality of competing in that area by de novo branching. The Supreme Court's rulings on this subject are clear. One purpose of Section 7 is to arrest the trends toward concentration or monopoly, before the consumer's alternatives disappear by mergers. United States v. Philadelphia National Bank, supra, 374 U.S. at 367, 83 S. Ct. 1715, 10 L. Ed. 2d 915; United States v. El Paso Natural Gas Co., 376 U.S. 651, 659, 84 S. Ct. 1044, 12 L. Ed. 2d 12 (1963). Potential competition must also be preserved. It is not sufficient to prove only that the banks do not actually compete. It must also be proven that they probably will never compete or have the potential to compete.
The banks' proposed plotting techniques fall short on a second ground. Such a market characterization is a perfect illustration of what the Supreme Court condemned as an "indefensible extreme" in PNB. Here, the proposed market is drawn so narrowly as to place (the merging banks) in different markets, because only the smallest customers are considered. PNB, supra, 374 U.S. at 361, 83 S. Ct. 1715. If we were to adopt the banks' plotting techniques, no banks or financial institutions would be in competition for the retail customers since such customers would be clustered around the financial institution of their choice. The mere fact that a customer chooses this one bank has nothing to do with competition, since the purpose of Section 7 and the Courts in enforcing this statute, is to preserve competition between the successful and unsuccessful providers of these banking services. United States v. El Paso Natural Gas Co., supra, 376 U.S. at 661, 84 S. Ct. 1044. The choice to bank where the customer pleases must be preserved and the gradual elimination of potential or actual competitors takes away competitive options which the customer should have. These competitive banking options, if preserved, will also require the banks or financial institutions to maintain competitive *15 services and interest rates which are all to the benefit of the customer. The desire to innovate new services to attract the retail customer is also present where there are many competitive options.
Therefore, we reject the defendants' proposed localized markets for the above reasons and shall consider the retail aspects of this merger within the four-county area. We shall, however, include competition from saving and loan associations and mutual savings banks where relevant.
The inclusion of saving and loan associations and mutual savings banks broadens the structural base to be examined even further. As already noted, we have reduced the market share of the merging banks from 14% to 11.7% because of accounts which are more properly allocable to the national market. But the problems which faced us there in arriving at realistic market shares reoccur in this instance. The statistics offered are incomplete and again would require the Court to delve into statistical problems which it is not fully competent to solve. The inclusion of these other institutions noticeably affects the examined deposit base. In recent years, commercial banks have become increasingly interested in securing time deposits, and as a result of this interest, the ratio between time and demand deposits is fast reaching an equal parity.
In 1961, the ratio of demand to time deposits in relation to total deposits was 74.6%-25.4% in the four-county area. In 1966, demand deposits increased only 14.4%, while time deposits increased 112.2%. This increase in time deposits changed the ratio to 61.2%-38.8%. Thus, time deposits now comprise approximately 40% of the total deposits held by commercial banks in the four-county area. (See defendants' Exhibit 19). Among the four-county commercial banks, Provident controls 8.3% of the IPC (Individauls, Partnerships and Corporations) time and savings deposits, while Central-Penn controls 4.6%. The resultant bank would control 12.9% of IPC time and savings deposits in the four-county area. (See plaintiff's Exhibit 15). The five largest commercial banks, as of December 31, 1965, controlled 73.2% of the total time and savings deposits held by commercial banks in the four-county area. The merger would increase this percentage to 77.8%.
When we consider competition from saving and loan associations and mutual savings banks, these above figures are considerably reduced. As of December 31, 1965, commercial banks held only 19.49% of the total savings dollar in the four-county area. Mutual savings banks held 45.73%, saving and loan associations 32.95%, and credit unions 1.88%. Provident and Central-Penn held 1.63% and 1.11%, respectively. (See defendants' Exhibit 20).
The resultant bank would control 2.74% of the total savings dollar in the four-county area. Since savings constitute some 40% of the total deposit base, some further adjustment must be made.
After shading down Justice's concentration figures from 14% to 11.7% because of accounts allocable to the national market, we shall further shade this 11.7% figure by including competition from the saving and loan associations. Since 2.74% represents the resultant bank's share of all time and savings deposits in the four-county area, we arrive at a final aggregate share of 9.74%.
We realize these figures are not totally accurate, and that our method in arriving at them is mathematically crude. However, this figure of 9.74% is more realistic in that it represents as nearly as possible competition for only those *16 accounts for which financial institutions compete in the four-county area.
After arriving at a more realistic concentration ratio, we can now consider whether a merger, which produces a firm controlling some 9.8% of the relevant market, is anticompetitive. Here, there has been no merger which involves shares of the market so large that it approaches monopolistic proportions. See Brown Shoe Co., Inc. v. United States, 370 U.S. 294, 82 S. Ct. 1502, 8 L. Ed. 2d 510 (1962); PNB, supra. Nor does it involve a merger which forecloses only a de minimis share of the market, since Central-Penn does not fall within that category. This merger lies somewhere in the middle of these two extremes and therefore an examination of the various economic and historical factors is necessary. Brown Shoe Co. Inc. v. United States, supra, at 329, 82 S. Ct. at 1526. Such an analysis must involve a prognosis of the probable future effect of the merger and not merely its immediate effect.
There is a distinct difference between competition and the number of competitors, especially in commercial banking. When Philadelphia had many more commercial banks than it does today, the banks then were not, in any sense, full service banks. The larger banks usually concentrated on a wholesale business with little or no regard to retail business, such as savings or trust accounts. More specifically, national banks were concerned with commercial lending, trust companies with trusts, and State banks concentrated on the retail accounts. By the merger process, the large commercial banks extended their reach into the retail area until they became what they are today, full service banks.
The banks, in their brief, contend that if this merger is not approved, the leading banks in the Philadelphia area will become insulated from competition. They cite this merger as a case where two small firms are proposing to merge in order to compete more successfully with the leading firm in that market. To support this contention, the banks point out that this merger will not produce the largest bank in the Philadelphia area, as was the case in PNB, supra, but will merely place the resultant bank on an even parity with the other four large reserve city banks in Philadelphia. To reach this parity, the merger is necessary to increase the lending limits of the respective banks to a competitive $10,000,000 level. It will also allow Provident and Central-Penn to solicit more correspondent banking business of which First Pennsylvania Banking and Trust Company and Philadelphia National Bank have the lion's share. In the area of computer-based services, neither Provident nor Central-Penn presently have the resources to make substantial computer-based innovations. The merged bank will be able to function as a "provider" of these computer services, rather than a purchaser. Another benefit *17 of the merger is that the respective banks can increase their share of international business in which the bank's size plays a determining role. It would be able to attract corporate funds by competing for corporate Certificates of Deposit.
In summary, the banks contend that the merger between the fifth and seventh largest banks in the four-county area will produce a bank better able to compete with the four leading commercial banks on a qualitative basis, since it will have more resources to be innovators in commercial banking, rather than mere takers or followers.
Similarly, Provident and Central-Penn maintain that this merger will also be pro-competitive in the retail market since the improvement of services resulting therefrom should intensify the competition with other banks and financial institutions. There is no question that a larger bank will be better equipped to serve retail customers, but it doesn't necessarily follow that improvements in quality of service will produce more competition.
However, the banks go one step further in their analysis. In retail banking, it is undisputed that convenience is a major factor in the location and competition for these accounts. Thus, as long as the number of independent banking offices in the community have remained the same, the competitive options for the retail customer will remain the same. Despite the substantial reduction of independent commercial banks in the four-county area, the competitive options for the retail customer have increased, and the services offered have increased quantitatively and qualitatively. For example, in thirty-five of the forty communities in the four-county area analyzed by the banks, there has been an increase in the number of bank and other financial institution offices. In seven, the number has remained the same, and in only two have the number of bank offices declined. (See defendants' Exhibits 89, 92). This fact is not so startling when one goes to the source of the increase. The large reserve city banks have branched into these areas, giving the retail customers the benefit of a large city bank with its concomitant services at minimal inconvenience. In doing so, more often than not, the large commercial banks have often merged with a smaller independent bank, but these mergers never affected competition per se on the retail level, since the number of banks in that community has remained the same.
This distinction between competition and the number of competitors is also true on the wholesale level of commercial banking. A large number of independent banks in the four-county area does not necessarily guarantee that competition for wholesale accounts will be meaningful or intense. Most of the present thirty-seven commercial banks in the four-county area are limited severely by their size, location and staff, and thus cannot compete seriously for wholesale accounts. Wholesale transactions usually require large lending limits, reserve city status, and a large and expert staff in diversified areas. Most small independent banks do not fit into this category.
The merger in question will increase the quality of competition among commercial banks for the wholesale customer. Testimony adduced at trial attests to this fact. Central-Penn is not at present a major competitive factor in the wholesale market because of its low lending limit (3.2 million) and its relatively small staff. The merger would produce another bank on par with the other leading wholesale banks in the city, thus giving a wholesale customer in the Philadelphia market a broader range from which to choose.
Our observation that competition on a qualitative basis will increase or remain the same on the wholesale and retail level is not determinative of our Section 7 analysis. We have determined generally that the immediate effect of this merger will not decrease competition in the four-county area. However, before a merger can be approved, this *18 Court must not look only to its immediate effect, but also the probable effect this merger will have in the future.
As stated before, the number of independent commercial banks in the four-county area has been substantially reduced. At present, the number stands at thirty-seven, with at least four mergers pending. Approval of this merger might well encourage more mergers which would reduce the present number substantially. Defendants counter this assertion by maintaining that the proposed merger involves unique considerations which will not be duplicated in the future. These unique considerations are that the two merging banks are the two smallest reserve city banks in Philadelphia and the resultant bank will be on parity only with the other reserve city banks. Any other merger within this group of reserve city banks would resemble the PNB-Girard merger, since it would inevitably create the largest bank in the city. Secondly, since Provident and Central-Penn have complementary branch systems, the effect on competition in the retail area will be minimal. Lastly, they argue, that the merger of Provident and Central-Penn is the culmination of the evolution of full service reserve city banks in Philadelphia.
These considerations are not compelling ones, however, since both banks, by their own admission, are healthy and progressive and presently are meaningful competitors in the four-county area. Although the two branching areas of the respective banks are somewhat complementary, both banks have the potential to branch into other areas. As a result, the retail customers of the outlying counties will be losing a potential entrant into their market if we allow the merger. In all respects, Provident is a full service bank, a status it has attained by the merger route. Central-Penn also is a full service bank, but admittedly, it is not on par with the other leading banks in the city. However, it is developing and strengthening its weaknesses, such as its trust and international departments, and always will be able to find some place in community banking.
But even assuming that this merger would not trigger more of the same, this merger is anticompetitive when viewed in light of the Supreme Court's pronouncements in interpreting Section 7 of the Clayton Act. The proposed merger is desired in a market which is becoming rapidly concentrated. Barriers to entry in this market are severe. These barriers are borne out by the fact that only three independent banks have been formed in the four-county area since World War II. (See defendant's Exhibit 54). Faced with this rigid market structure, even slight increases in concentration must be viewed with askance. United States v. Aluminum Co. of America, 377 U.S. 271, 84 S. Ct. 1283 (1964).
The aggregate share of total deposits which the resultant bank will hold in the four-county area is approximately 10%. Although this is relatively a small concentration ratio, the merger still runs counter to the Supreme Court's precept that competition will be most vital when there are many sellers, none of which has any significant market share. United States v. Aluminum Co. of America, supra; United States v. Von's Grocery, 384 U.S. 270, 86 S. Ct. 1478, 16 L. Ed. 2d 555 (1966). It becomes clear that if a market can be characterized as oligopolistic, any merger which strengthens this oligopolistic structure must be struck down. This is the dilemma which confronts this Court. Philadelphia concededly has an oligopolistic banking structure. The purpose of Section 7 is to preserve and maintain competition, but in so preserving, it is also possible for the oligopolistic structure to be strengthened. Barring significant new entries into commercial banking in the four-county area, and the prospect of divestiture of the large banking institutions seemingly remote, it seems the *19 necessary result of such a policy will only strengthen the present leaders while the lesser banking institutions are correspondingly weakened. Justice and the Supreme Court have offered solutions to this dilemma and in essence, these solutions amount to an absolute bar of all horizontal mergers which would increase the concentration ratio of the merging firms in question, especially in a market which is already highly concentrated.
The merging banks' percentage share of the market (10%) certainly is not as foreboding as the large percentage of the market which PNB and Girard would have controlled had that merger been approved. Since that time, the Supreme Court has indicated that any merger producing a firm controlling 25% of the relevant market, is presumptively bad. United States v. Continental Can Co. et al., 378 U.S. 441, 84 S. Ct. 1738, 12 L. Ed. 2d 953 (1964).
We do not find that this merger which would produce a bank controlling approximately 10% of the relevant market is presumptively anticompetitive. This percentage figure is not undue in relation to the composition of the market as a whole. The resultant bank certainly will not have market power which would enable it to persistently behave over substantial periods of time in a non-competitive fashion. (Dirlam, N.T. 414-15). The converse is true. The defendant banks desire this market power so that they can behave competitively.
Yet, Congress and the Supreme Court, in carrying out the mandate of Section 7, have stated that one purpose of Section 7 is to arrest a trend toward concentration or a tendency toward monopoly, before the consumers' alternatives disappeared through mergers. United States v. El Paso Natural Gas Co., supra. Thus, we must not focus only on the effect of this merger in the present, but must scan the horizon to see what lies ahead. The thrust of the El Paso decision, supra, extends to the fact situation presently before us. Provident and Central-Penn are presently competing for the retail and wholesale dollar. This competition (between these two banks) is not as keen as it could be, but potentially this competition can become more intense. In the retail area, de novo branching can bring these two banks into the private domain of each other. In the wholesale area, both banks presently offer sufficient alternatives for the commercial customer. This Court accepts the fact that both Provident and Central-Penn would become a better and more competitive bank after the merger, but this fact does not offset the loss of Central-Penn which is a growing and healthy bank in an oligopolistic market.
The case which we feel controls the merger before us is United States v. Von's Grocery, supra, which involved a merger between the third largest and the sixth largest retail supermarkets in the Los Angeles area. The resultant company would control only 7.5% of the total value of retail groceries sold in the Los Angeles market each year. The Los Angeles food market was characterized as one where there was a long and continuous trend toward fewer owner-competitors. In an eleven year span, the number of chain stores (two or more) increased from 96 to 150, while the number of single owner-operated stores was reduced from 5,365 to 3,818. On these facts alone, the merger was condemned without a functional inquiry into the nature and quality of the competition. The basic purpose of the Sherman and Clayton Acts (as construed by Justice Black) was to prevent economic concentration in the American economy by keeping a large number of small competitors in business. The only exceptions to this rule are where one of the companies is about to fail, or where two companies had to merge to save themselves from destruction by some larger and more powerful competitors. This view, which shifts the emphasis of Section 7 from protecting competition to protecting competitors (see Justices Stewart's and Harlan's dissent in Von's), was reiterated in United States v. Pabst Brewing Co., supra.
*20 The facts in this case are parallel to those in Von's. Here, the merger involves the fifth and the seventh largest banks in the four-county area. The resultant bank would control approximately 10% of the relevant market, thus exceeding the percentage figure in Von's by approximately 2%. The Philadelphia banking market unqualifiedly can be characterized as one where there has been a long and continuous trend toward concentration.
Therefore, even if we did find that competition would be improved by this merger, as it probably was in Von's, the inescapable fact is that the number of competitors in the four-county area has been steadily decreasing over the years, and necessarily increasing the concentration ratios among the leading banks. Since this merger falls within this trend toward merger and toward more concentration, it must be ruled anticompetitive under existing antitrust laws. This bank merger clearly does not fall within the two exceptions referred to in Von's. Central-Penn is not a failing bank. Nor is this an attempt by two banks to save themselves from oppressive competition. These banks are not being preyed on by the larger banks, nor are they losing appreciable ground in the struggle for market power. They are merely asking this Court to put them in a position where they can fight on equal terms and we cannot put them there because it is not absolutely necessary.
THE CONVENIENCE AND NEEDS OF THE COMMUNITY
Our disposition of this case does not end with the conclusion that this merger is anticompetitive under current Section 7 standards. B.M.A.1966 has carved out a new exception or defense to such anticompetitive mergers in the form of the convenience and needs test. 12 U.S.C. § 1828(c) (7) (B). This defense or justification is available to the banks once it has been determined that the proposed merger would be anticompetitive as judged by normal antitrust standards. United States v. First City National Bank of Houston, supra, 386 U.S. at 366, 87 S. Ct. 1088, 18 L. Ed. 2d 151.
Again, the lower Courts have taken opposite views on the meaning of this new test. The District Court, in United States v. Crocker-Anglo National Bank, supra, felt that the merger between Crocker and Citizens met the convenience and needs test because (a) it caused an immediate increase in the number of statewide banks competing within the State; (b) produced a bank having increased ability to mobilize financial resources in a capital deficit area; (c) created a bank which is better equipped and therefore more likely to engage in credit gap financing; (d) created another substantial competitor in the underwriting of municipal bonds, and (e) enhanced the merged bank's ability to compete more vigorously in international banking. A quick appraisal of these reasons reveals that the Crocker Court adopted a broad public interest approach rather than a restricted narrow exception test in analyzing the scope of this new exception.
The District Court, in United States v. Third National Bank of Nashville, supra, 260 F. Supp. at 877, did not reach the question of convenience and needs directly since it found that the merger between Third National Bank and the Trust Company was not anticompetitive. However, the Court referred to the Supreme Court's characterization of the convenience and needs test. In the "somewhat larger contours of the failing company doctrine", the District Court placed the problem of the "floundering" or "stagnating" bank. Such a bank is one which has reached a point of deterioration or stagnation, but has not reached the status of a failing bank which is an admitted exception in all merger cases. These floundering or stagnating banks do fall within the convenience and needs exception because they are no longer able to serve the public on a competitive basis with other banks *21 in the market area. United States v. Third National Bank of Nashville, supra, at 884. The Tennessee Court's reference to the convenience and needs test is at best a restricted one, but it places in view the two conflicting interpretations of this new exception which has been debated in Congress and is now being argued in the Courts.
The banks, in their brief, state that the merger between Provident and Central-Penn will serve the convenience and needs test in the following ways: (1) The merged bank will be more comparable in resources, services and prestige to the larger banks in other cities. As such, it will help the City of Philadelphia in its competition with those cities to attract and retain job-creating enterprises. It will also aid the Port of Philadelphia in its competition with other ports and help to attract and retain foreign commerce; (2) the merged bank will provide more competition for the larger banks in Philadelphia; (3) the merged bank will provide better service to individuals and small businesses in the communities served by its branches.
Before attempting any in-depth analysis of the defendant banks' contentions as to what the convenience and needs test embraces, this Court finds, at the outset, that it is necessary to make some general observations concerning the new exception. Since the community's primary need is for a competitive banking market, an anticompetitive merger, if approved, eliminates one aspect of the community's need for banking services. Thus, anticompetitive mergers should be approved only in those few instances where these anticompetitive effects are overwhelmed by the more compelling needs of the community.
In United States v. First City National Bank of Houston, supra, 386 U.S. at 369, 87 S. Ct. at 1093, the Supreme Court alluded to the convenience and needs test exception as one "related although * * * remotely, to the failing-company doctrine." This language does not mean that the defense of convenience and needs must be equated with the failing bank doctrine because this concept was already firmly engrained into our antitrust jurisprudence. Congress, in amending the Bank Merger Act of 1960, intended a wider embrace under the new exception. Just what this embrace involves is the problem before this Court.
In construing the meaning of the convenience and needs test, this Court has turned for guidance to the legislative history of the amended Bank Merger Act. Before doing so, we are reminded at the outset by Congressman Reuss' remarks that the words "convenience and needs of the community" are capable of some meaning, but they will require some Court interpretation. [See 112 Cong. Rec., 2345 (Feb. 8, 1966)].
Realizing that we must interpret this new defense, we must of necessity seek enlightenment from the legislative history. In examining the convenience and needs of the community, the question arises as to what community should be examined. Is the test limited to the appropriate geographic market utilized in determining anticompetitive effects, or can our analysis extend beyond this territorial limitation? The legislative history on this subject reveals that the sponsors of the House Bill intended that "conditions in the regional market or national market are relevant only insofar as they affect the convenience and needs of the community to be served." [See remarks of Congressman Reuss, 112 Cong.Rec., 2350 (Feb. 8, 1966); see also Senator Robertson's remarks, 112 Cong.Rec., 2542 (Feb. 9, 1966)]. With this statement, we agree. In particular cases, a community might need a larger bank which can handle larger loans and provide more sophisticated services. A merger, albeit anticompetitive, might be allowed in this instance if the community's need and convenience for a larger bank clearly outweigh the anticompetitive effects. *22 Another instance for allowing a merger on the basis of convenience and needs, by considering the national or regional market, is where an entire region might be without a large commercial bank. A merger creating such a large bank to meet the convenience and needs of this community, whether it be local or regional, might be allowed.
Beyond these general observations, perhaps the most common application of the convenience and needs test, and the one primarily intended by Congress, is the "floundering" or "stagnating" bank. Congress was intensely concerned that under the rigid test propounded in PNB, such floundering banks could never be saved by merging with a stronger and healthy bank. The concern was not without justification since the only announced exception to the strict anticompetitive standard enunciated in PNB was the failing company doctrine.
The House Report which accompanied the House Bill to the Senate dealt with the floundering bank situation quite specifically under the heading, "The floundering bank":
"Under general antitrust law criteria, particularly as they have been developed over the past few years, the banking agencies find it difficult to deal as they would like with the floundering bank problem in medium to smaller sized communities. The problem arises where there is a relatively small number of banks, and one or more of these banks appear to be stagnating. It may be because it is below the economic minimum size to attract capable and vigorous management personnel, it may be because it is closely held by owners who insist on unrealistically conservative policies, or it may be for any other reasons which are discernible only by an examination of that particular bank as an individual institution. The banking agencies, with some differences in degree among themselves, have contended that they should be able to consider a merger application on the basis of such an individual examination, and to approve it if they believe that the result would be a more vigorously competing institution, furnishing better overall service to the community, even though the reduction in the number of competing units, or the concentration in the share of the market in one or more lines of commerce, might result under general antitrust law criteria in a substantial lessening of competition." H.R.Rep. No. 1221, 89th Cong., 2d Sess. 3 (1966).
Congressman Todd, in the legislative debate, clarified this concept of the floundering bank even further:
"The floundering bank case arises when it is not absolutely certain that a bank is failing, but when it may be in a position of stagnation, which might lead, at a later date, to a closing of its doors." 112 Cong.Rec., 2352 (Feb. 8, 1966); see also Congressman Widnall's remarks at 2335.
Senator Robertson, in explaining the effect of the House amendment on convenience and needs to the Senate stated:
"There are times, when there are two banks in a community and one is about to fail because of poor loans or bad management, and the banking agencies say, `It would be bad to have a bank failure; will you see if you can arrange to merge?' Everybody will agree that such a merger would be in the public interest. But unfortunately, Section 7 of the Clayton Act is so harsh that if two banks seek to merge into one, any judge could say, `You have ipso facto lessened competition. If you have lessened competition, you are illegal and we will destroy you.'" 112 Cong.Rec., 2545 (Feb. 9, 1966).
Under the amended Act, a compromise was effected. The anticompetitive standard was made stronger since it was no longer to be diluted by the other banking factors. It was to be considered singly. The other important change was the new convenience and needs exception which takes into account the banking factors that were previously enmeshed with the competitive factor under the old standard. *23 However, the competitive factor does remain pre-eminent and any exceptions to this standard would indeed be rare. [Patman, 112 Cong.Rec., 2334 (Feb. 8, 1966)].
The legislative history confirms our belief that the convenience and needs test must be read restrictively. It is not a mandate for mergers between banks across the board. It merely allows bank mergers in those cases where the needs and convenience of the community are so compelling that competition will be enhanced rather than decreased by the merger. This will occur in only a very few instances.
With these observations in view, we shall now turn to the defendant banks' contentions that they have met their burden of proof as to the convenience and needs test. The banks contend that the primary issue in this case is the convenience and needs of the community to be served. With this, we disagree, since the convenience and needs test has already been characterized by the Supreme Court as a defense or an exception to the primary issue of anticompetitive effects. United States v. First City National Bank of Houston, supra.
Generally, the banks state that the proposed merger would create a more efficient bank offering a broader range of services. This broader based bank will be able to attract and retain jobcreating enterprises. It will also be able to compete with the large Philadelphia and New York banks. Lastly, the merger will provide better services to industries and small businesses.
More specifically, the banks maintain that we are on the threshold of a "checkless society" whereby computer and other automated services will be of prime use. This increased cost and specilization will require large capital outlays for personnel and equipment which can only be met by the larger banks. It is true that the resultant bank would be better able to provide these services because it would simply have more funds at its disposal. However, the fact that improved service to the wholesale and retail customer will result from this merger is not controlling. The question in this area is does the Philadelphia banking community need another billion dollar institution aside from the four which it already has? The reasons proffered by the banks for such a need or convenience are not compelling since improved services will almost invariably result from every merger. Something more direct or unique in the way of convenience and needs is required by the Act.
The banks also argue that the merger will assist Philadelphia in its efforts to solve its economic problems. They point to the fact that Philadelphia's employment growth has lagged behind the national average, citing the loss of textile companies which employed many unskilled workers. In recruiting science-based industries to fill this employment gap, adequate financing and specialized personnel familiar with the technology of the industry are required. They also state that the merger will improve the Philadelphia Port situation. New York City currently finances 70% of all foreign commerce in the United States at a decided disadvantage to the Philadelphia Port.
Again, it is true that the proposed merger may help the above situation, but there are too many other causes and reasons involved in these economic problems, which were testified to during the trial of this case, to convince the Court of the legal validity of the banks' assertions. At most, these assertions are tenuous. It was admitted that the textile companies did not leave Philadelphia because of inadequate or poor financial support, but because the South offered a better deal, tax-wise and employment-wise. Also, the Philadelphia Port does *24 not suffer from inadequate financing, but loses most of its general cargo trade to New York because that city is the hub of commercial activity in the East, with the largest population. The loss of corporate headquarters in the Philadelphia area cannot be attributed to the Philadelphia banks either, since in most cases, the industry or company itself remains, with merely the headquarters shifting. Testimony at trial revealed that corporate headquarters have been moved from Philadelphia to elsewhere for a plethora of reasons, but nowhere, in no case, were the banks blamed. In any event, the fact that Philadelphia presently has four billion dollar banks indicates that the desired expertise or capital can be furnished in a particular situation by these four larger banks.
If specific facts were shown that Philadelphia was losing accounts, corporations, or was suffering in any way from the lack of an additional billion dollar bank, this Court would be of a different disposition. However, there has been no showing or proof of a linkage between such an additional billion dollar bank and Philadelphia's problems. The mere supposition that a larger bank might assist Philadelphia and its economic problems just is not enough under the convenience and needs test which we feel is a narrow and restricted exception.
The banks also have alleged that the proposed merger will be pro-competitive and, in this sense, also meets the convenience and needs of the community. They justify the Provident and Central-Penn merger as a case where two small firms in a market propose to merge in order to be able to compete more successfully with the leading firms in that market. This merger, unlike the PNB-Girard merger, would not produce a bank which is the largest in the four-county area, but one which merely challenges the leaders. If the merger is not allowed to go through, the banks contend that the leading banks will become insulated from competition because of their head start on lending limits, branches, correspondent business, and computer-based services. Only by this merger can the five largest reserve city banks be on an even parity. More or less, this type of analysis goes more to the convenience and needs of the respective banks themselves, rather than to the community. Perhaps the merger would increase competition for the wholesale accounts, but it would decrease one substantial option for some wholesale accounts and definitely for retail accounts. The pro-competitive effects of the merger are definitely offset by the anticompetitive effects which were dealt with under our previous discussion of the anticompetitive effects of this merger.
The banks have declined to justify this merger as one which would join a healthy and prosperous bank with one which is floundering. Central-Penn does have some capitalization and officer replacement problems, but they are not serious. Central-Penn recently declared an extra dividend at Christmas time which does not reflect a bank with capitalization difficulties. Its position in Philadelphia is relatively secure. Perhaps it never will be able to challenge the largest banks in Philadelphia on equal terms, but it will be able to challenge the medium-sized and smaller banks on more than equal terms.
The burden of proof upon the banks to prove that this merger meets the convenience and needs of the community is indeed a heavy one. BMA 1966 requires that the convenience and needs of the community must clearly outweigh any anticompetitive effects before a merger can be approved. This burden has not been met in the instant case. At best, it has resulted in a tie. Justice has shown anticompetitive effects and the banks have pointed to convenience and needs. However, in a tie situation, the banks necessarily lose. As stated before, *25 Congress intended that exceptions to anticompetitive mergers should be rare. In those few instances, the convenience and needs of the community must be so compelling that the merger's anticompetitive effects become only a sidelight to the overall good of the community.
In the case before us, the banks have proved the convenience and needs of the Philadelphia community in a persuasive, but not compelling fashion. Although we are persuaded that this merger would be good for Philadelphia and its banking community, this Court has seen no compelling evidence which would except this merger from our antitrust laws.
In short, the banks have failed to meet their burden of proving that this merger would have the probable effect of meeting the convenience and needs of the community. At best, they have convinced us that this is a possibility. They have failed to establish "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." Title 12 U.S.C. § 1828(c) (5) (B).
The Comptroller of the Currency presented a very appealing jury case. His witnesses, outstanding bankers and businessmen, established by a fair preponderance of the evidence (a) that they were convinced that the merger was a good thing for Philadelphia, as well as the banks; (b) that instead of the fears voiced by the economists for Justice that competition would be lessened, there would be strengthened competition both at the retail and wholesale level; (c) that not entirely as a measure of civic pride, another bank of the size contemplated would redound to the economic benefit of Philadelphia. It is not necessary to delineate in detail the testimony adduced. It was extremely persuasive and most interesting, but where the Judge must declare the law and the findings of fact, he cannot deviate one jot from those requirements. The present posture of the antitrust laws, as declared by the Congress and the Opinions of the Supreme Court, require the result reached. Comptroller Saxon's remarks about the necessity of financing urban renewal, redevelopment of the ghettos, etc., are in this day and age peculiarly appropriate. However, since the Court must adhere to a quantitative, rather than a qualitative theory, the result is inevitable.
It might also be pertinent to comment at this point on the effect of the Government's rebuttal testimony after the banks' and the Comptroller's showing of convenience and needs. The witnesses, a small banker, a city official, and small businessmen, were apparently produced for the purpose of showing that where the larger urban bank moved into suburban territory, it was not a good thing for the community. The banker, a very estimable gentleman, who lives on the perimeter but within the four-county area, clearly indicated that he knew nothing about the problems confronting the Philadelphia banks and could throw no light on whether the merger was good or bad for the community. The other witnesses definitely proved and established beyond peradventure of doubt that the coming of the branches of the urban bank into the suburban community created banking competition where none had been in existence before and definitely added to the well-being generally and economy of the entire community. Instead of proving that branch banking into small towns was harmful, Justice definitely proved the contrary.
The foregoing will constitute the general Findings of Fact and Conclusions of Law of the trial Judge. Each party, Justice, the banks, and the intervenor, have submitted, in different forms, Requests for Findings of Fact and Conclusions of Law. The rulings thereon will be filed as an appendix to this *26 Opinion. In any instance where there may be any lack of clarity in approved Requests for Findings of Fact or Conclusions of Law, the language of the Opinion will govern.
Justice's Requests for Findings of Fact, having been considered, are ruled upon as follows:
AFFIRMED Requests: Numbers 1, 2, 3, 5-14 inclusive, 16-19 inclusive, 21-28 inclusive, 30, 32, 33, 42, 44, 46, 53, 57, 59, 60-70 inclusive, 72, 73, 76, 77, 80, 81-83 inclusive, 85, 86, 88, 89, 92-97 inclusive, 102, 109, 110, 112-116 inclusive, 124, 127, 128, 130-137 inclusive, 139, 142-149 inclusive, 151, 153-156 inclusive, 158, 160-167 inclusive, 169.
AFFIRMED WITH AMENDMENTS Requests: Numbers 20 (89.9 million is substituted for 90.1 million), 50 (first two full sentences are deleted), 51 (last sentence of (c) is deleted), 71 (accepted in its limited and technical sense), 79 (final sentence is deleted), 99 (see Opinion), 100 (last sentence is deleted), 111 (word "rapidly" is deleted), 121 (technically correct as stated), 123 (last sentence is deleted), 138 ("tends to decrease" is substituted for the word "decreases"), 150 (the words "but less than Los Angeles, Detroit, San Francisco, St. Louis and Washington" are added), 152 (the words "but less than Los Angeles, Detroit, San Francisco, St. Louis and Washington" are added).
REFUSED AS STATED Requests: Numbers 4, 15, 31, 34-37 inclusive, 40, 41, 49, 52, 54, 55, 58, 78, 101, 104-108 inclusive, 117-119 inclusive, 122, 125, 126, 129, 141, 157, 159, 168, 170.
DENIED Requests: Numbers 29, 43, 45, 47, 48, 56, 74, 75, 84, 87, 90, 91, 98, 103, 120, 140.
Justice's Requests for Conclusions of Law, having been considered, are ruled upon as follows:
AFFIRMED Requests: Numbers 1, 3, 4 (under the current Supreme Court test, this Conclusion is valid based upon the statistics offered), 5, 6, 7.
REFUSED AS STATED Request: Number 2 (see Opinion).
The banks' Requests for Findings of Fact and Conclusions of Law, having been considered, are ruled upon as follows: Since the defendant banks have elected to make no specific Requests for Findings of Fact and Conclusions of Law, the Court is unable to pass on them seriatim. The Opinion sets forth the findings and conclusions of this Court in sufficient detail to cover this situation.
Intervenor's Requests for Findings of Fact, having been considered, are ruled upon as follows:
AFFIRMED Requests: Numbers 1.01, 1.02, 1.03-1.08 inclusive, 2.01-2.09 inclusive, 3.01-3.07 inclusive, 4.01, 4.03-4.06 inclusive, 4.07-4.11 inclusive, 4.13-4.16 inclusive, 4.19, 5.01, 5.07-5.11, inclusive, 5.14-5.19 inclusive, 5.21, 5.24-5.26 inclusive, 5.28-5.30 inclusive, 5.33, 5.35, 5.37-5.43 inclusive, 7.02, 7.04, 7.05, 8.07, 8.09, 8.12-8.14 inclusive, 8.17-8.22 inclusive, 8.25, 8.31, 8.32-8.35 inclusive, 8.37, 8.38, 8.40-8.47 inclusive, 8.51-8.53 inclusive, 8.58, 8.61, 8.63-8.71 inclusive, 8.73-8.82 inclusive, 8.84-8.90 inclusive, 8.92-8.94 inclusive, 8.97, 8.98, 8.100, 8.102, 8.105-8.111 inclusive, 8.115, 8.116-8.118 inclusive, 8.120-8.128 inclusive, 8.130, 8.131, 8.133-8.137 inclusive, 9.04, 9.09, 9.12, 9.14, 9.15-9.18 inclusive, 10.05-10.07 inclusive, 11.04-11.07 inclusive, 11.12-11.22 inclusive.
AFFIRMED AS AMENDED Requests: Numbers 5.04 (the word "commercial" is inserted at the beginning of the paragraph), 5.23 (see Opinion), 5.27 (see Opinion) 5.44 (as applied to large borrowers).
REFUSED AS STATEDRequests: Numbers 4.17, 4.20, 4.21, 5.02, 5.03, 5.05, 5.06, 5.12, 5.13 (see Opinion), 5.20, 5.22, 5.31, 5.34, 5.36 (see Opinion), 7.01, 7.03, 7.06-7.09 inclusive, 8.01-8.06 inclusive, 8.08, 8.10, 8.11, 8.15, 8.16, 8.23, 8.24, 8.26-8.29 inclusive, 8.36, 8.39, 8.48-8.50 inclusive, 8.54-8.57 inclusive, 8.59, 8.60, 8.62, 8.72, 8.83, 8.91, 8.95, 8.96, 8.99, 8.101, 8.103, 8.104, 8.112-8.114 inclusive, 8.119, 8.129, 8.132, 8.138, 9.01-9.03 inclusive, *27 9.06-9.08 inclusive, 9.10, 9.11, 9.13, 10.01-10.04 inclusive, 11.01-11.03 inclusive, 11.08-11.11 inclusive, 11.23.
DENIED Requests: Numbers 4.02, 4.12, 4.18, 5.32, 8.30, 9.05.
With respect to Requests: Numbers 6.01-6.20 inclusive, the language of the Opinion will control.
Intervenor filed no specific Requests for Conclusions of Law.NOTES
 This quotation is taken from an article by Professor Wigmore in the April, 1929, issue of The Journal of The American Judicature Society, reading in part as follows: "Law and justice are from time to time inevitably in conflict. That is because law is a general rule (even the stated exceptions to the rules are general exceptions); while justice is the fairness of this precise case under all its circumstances. * * * Law the rule MUST be enforced the exact terms of the rule, justice or no justice. * * * Now this is where the jury comes in. The jury, in the privacy of its retirement, adjusts the general rule of law to the justice of the particular case. Thus the odium of inflexible rules of law is avoided, and popular satisfaction is preserved. * * * That is what jury trial does. It supplies that flexibility of legal rules which is essential to justice and popular contentment. And that flexibility could never be given by judge trial. The judge * * * must write out his opinion, declaring the law and the findings of fact. He cannot in this public record deviate one jot from those requirements. The jury, and the secrecy of the jury room, are the indispensable elements in popular justice."
 See Professor Handler's scholarly article "The Supreme Court and the Antitrust Laws: A Critic's Viewpoint", 1 Georgia L.Rev., 339, 1967.
 A canvas of Philadelphia Saving Fund Society's accounts revealed that 2.84% by number came from the excluded counties. The reason Philadelphia Saving Fund Society was used as a sampling device was that it was the only financial institution in the four-county area which broke down its accounts on a county basis. (Intervenor's brief p. 30).
 A case in point is finance companies which allegedly compete with commercial banks in the consumer loan field. However, the rates of interest at a finance company are substantially higher (N.T. 50) and the majority of a finance company's source of funds is derived from commercial banks. Thus, finance companies cannot be said to be in direct competition with commercial banks, since they derive most of their lendable funds from commercial banks and assume loan risks which commercial banks would be unwilling to take.
 Derek Bok, after examining the various structural approaches, felt that Courts or agencies, in assessing mergers, should look to the change that has taken place in the aggregate share possessed by the largest firms in the market. The number of firms selected for this purpose should be large enough to include the acquiring firm. The merger should then be disallowed if the aggregate share of the market following the acquisition substantially exceeds (by 7%-8% or more) the aggregate share of the market controlled by the same number of firms at any time during a period reasonably prior to the acquisition, e. g., (5-10 years). This rule should be subject to an exception where the share of the acquiring firm following the merger is no larger than it was in the base period. Bok, supra.
George Stigler, in his article "Mergers and Preventive Antitrust Policy", 104 U. of Pa.L.Rev. 176 (1955), also dealt with concentration ratios and the question of at what point should mergers be deemed violative of Section 7. The center of controversy, according to Stigler, arises somewhere in between what can be characterized as a competitive industry where the largest firm controls 10% or less of output and the non-competitive industries where 2-5 firms control 75% or more of output. Every merger which produces a firm controlling 20% of the industry's output should be presumed violative under the statute. In other mergers below this proscriptive percentage, several factors must be examined: (1) ease of entry into the market, (2) rate of industry growth, and (3) closeness of substitute products made by other industries.
 Congressman Patman stated in 112 Cong.Rec. 2334 (Feb. 8, 1966) that "It should be made very clear that there is no intent on the part of the (House) Committee to change the application of the antitrust laws as they apply to bank mergers. If the Committe had wanted to do this, it would simply have exempted bank mergers from the application of antitrust laws * * * This bill is intended to have the effect of making the bank supervisory agencies give substantially more emphasis to the antitrust standards in determining the competitive effects of a merger than they did under the 1960 law * * *."
The only effect this amended Act has upon traditional antitrust law is that for bank mergers, the anticompetitive standard is no longer to be the sole criterion for adjudging the merits of the merger. Courts must also consider convenience and needs. (Senator Robertson, p. 2538).
 United States v. Philadelphia National Bank, supra; United States v. First National Bank & Trust Co. of Lexington, 376 U.S. 665, 84 S. Ct. 1033, 12 L. Ed. 2d 1 (1964); United States v. Continental Can Co. et al., supra.
 Since World War II, only three new commercial banks have been established in the four-county area. Another new commercial bank will be opened this year.
 See Kerslake, N. T. 2591; Bradshaw, N. T. 2769; Myers, N. T. 3158-59; see also defendants' Exhibits 105, 106.
 See Weiler, N. T. 2871, 2882; McGinley, N. T. 2885; Wood, N. T. 2924; Moyer, N. T. 3080.
 Since the ratio of time to demand deposits is 40-60 in the four-county area, we took the resultant bank's share of 2.74% in the time and savings account category and added this figure to 60% of 11.7%, which is 7%, giving us a total of 9.74%.
 Competition from other financial institutions for mortgage loans was not computed since the ratio of these mortgages to total loans was not known. However, we do know that commercial banks held 12.52% of the total mortgage loans in the four-county area with the remainder being held by mutual savings banks and saving and loan associations about equally. The resultant bank would control .95% or less than 1% of the total mortgage loans in the four-county area. Thus, the proposed merger would not be anticompetitive in this specific market. (Defendants' Exhibit 20).
 This distinction was recognized by the Supreme Court in Brown Shoe Co., Inc. v. United States, supra, 370 U.S. at 320, 82 S.Ct. at 1521: "Taken as a whole, the legislative history (of Section 7) illuminates congressional concern with the protection of competition, not competitors, and its desire to restrain mergers only to the extent that such combinations may tend to lessen competition." (Emphasis supplied).
 The resultant bank would have the third largest lending limit among the four-county commercial banks. Presently, Provident has a lending limit of 8.4 million, while Central-Penn's is 3.2 million. First Pennsylvania Banking and Trust Company's legal lending limit is 12 million and Philadelphia National Bank's is approximately 15.8 million. (Defendants' Exhibit 7).
 Girard-Doylestown National Continental-Sonsitaly, Fidelity-Doylestown Trust, and this merger.
 See Via, Jr., "Antitrust and the Amended Bank Merger and Holding Company Acts: The Search for Standards", 53 Va. L.Rev., 1115 (1967).
 Congressman Reuss, in 112 Cong.Rec., 2337 (Feb. 8, 1966) put it this way: (The convenience and needs exception) "intentionally creates a heavy burden for the proponents of a merger, and I anticipate very few cases in which this burden can be sustained."
 On December 6, 1967, Frederic L. Ballard, Esquire, counsel for the banks, during the final argument of this case, characterized the banks' burden of proof in the following manner: "And I think that `clearly' is something that has to go within your Honor's judgment, and I think that it is probably of no more significance than the baseball rule that a tie favors the runner. I'm afraid that I would have to say here that a tie favors the Department of Justice." (N.T. 89).