Honorable Alan Greenspan
Dear Mr. Chairman:
This letter is in response to the letter dated December 17, 1991, from the Federal Reserve Bank of San Francisco, enclosing for our information a copy of an application filed with the Board of Governors, pursuant to Sections 3 and 4 of the Bank Holding Company Act and Sections 25 and 25(a) of the Federal Reserve Act, by BankAmerica Corporation, of San Francisco, California ("BAC"), for prior approval to acquire Security Pacific Corporation, of Los Angeles, California ("SPC").
The Department has reviewed the application, as supplemented by BAC through March 9, 1992, and has conducted an extensive investigation to determine the likely competitive effects of the proposed transaction. During the course of that investigation, the Department interviewed officers and employees of the parties, other banks and thrift institutions in the affected geographic areas and elsewhere, business customers of banking services, and non-bank suppliers of products and services that might be claimed to compete with products and services supplied by the merging banks, among others. We also reviewed a substantial quantity of documents and business information provided to us by the parties and others, financial data of the firms in the various markets, and other materials. We have considered the application in light of BAC's commitment to divest branches, deposits, loans and other assets to competitively suitable purchasers, and to permit the selective solicitiation for hire of branch and loan personnel.
Based on the evidence available to date from our extensive investigation, the Department has concluded, after a thorough investigation and analysis of all of the relevant facts, that, subject to the divestiture of the branches, deposits, loans and other assets identified in BAC's submission of March 9, 1992, as described below, the approval of the merger is not likely to have a significantly adverse effect on competition in the markets affected by this transaction. That divestiture commitment followed negotiations with this Department and with the Attorneys General of the affected states, and revised the divestiture commitments previously made by BAC on January 29, 1992.
Due to the very large number of markets involved and in light of our conclusion, we will describe our analysis generally, rather than particularly for each of the more than 100 local areas in which deposit concentration raised concerns under the thresholds in the Department's Merger Guidelines.
Having collected the evidence, we began our analysis of the merger by examining the particular products and services offered by the merging parties. This investigation and prior investigations have indicated that there are enough differences in the competitive dynamics of business and retail banking to conclude that it is useful to evaluate these product areas separately. The business banking products we examined include transaction accounts, business lines of credit, working capital term loans, other types of term loans (generally for equipment finance), and cash management services (particularly lockbox, controlled disbursement, wire transfer and account reconciliation). The retail banking products we examined include transaction accounts, savings accounts, home mortgages, home equity loans and personal loans.
We focused our analysis of business banking on two customer groups: small businesses (generally those with annual revenues of less than $10 million) and middle market businesses (those businesses with annual revenues of approximately $10 million to $100 million, and perhaps more). Each of these groups of customers have different banking and borrowing needs, and different access to potential suppliers, than do larger corporations. While the largest business customers may be able to obtain operating finance from distant institutions or from the public debt securities markets, neither small businesses nor middle market businesses generally have these options. Small and middle market businesses are therefore more likely to be adversely affected by an in-market bank merger than are larger businesses.
In this investigation and in prior investigations, we have learned that small business customers rely on commercial banks for operating or working capital credit, to meet short-term or seasonal funding needs. Based on the evidence we reviewed in this investigation, we found that the purpose and characteristics of these loans make other credit products, including loans to finance equipment purchases and loans secured by residential real estate, poor substitutes.
There are sound economic reasons that lead small businesses to tend to purchase banking products locally, and for banks to tend to provide these products (particularly credit products) only to small businesses located in their area. Because they tend to be locally limited, in many areas small businesses can practicably obtain banking and lending services only from a relatively small number of suppliers.
We also found in this investigation that most middle market customers use bank lines of credit to finance their operating cash flow needs. These lines typically are for $1-10 million. Many middle market businesses, like smaller businesses, generally seek to obtain operating credit and other bank products (such as transaction accounts and cash management services) locally or regionally, for reasons of convenience and to reduce transaction costs. Many also find that there are good reasons to obtain operating credit, transaction accounts and cash management services from the same bank.
In each affected area, we sought to identify the actual product overlaps between the merging parties and the competition that would be eliminated by the merger, and we have analyzed the effects of the merger on that competition. This approach has led us to focus our attention on different types of banking activities in different states and in different areas within these states. Our investigation indicated the following:
Having identified in each state the product areas in which competition was most likely to be affected by the merger, we then examined the universe of potential suppliers of those products. In our investigations in First Hawaiian (1990), Fleet/Norstar (1991), and Society Corp. (1992), and here, the evidence indicated that small businesses had few, if any, alternative suppliers or close substitute products for their banking and lending needs. Small businesses would be unlikely to find other firms to be sources of close substitute products to banks in the event of a small but significant and nontransitory price rise for commercial bank products and services.
Middle market businesses generally have working capital credit needs that exceed the lending capacity of many smaller banks, and many of these cash management services are not available from smaller banks. However, these businesses generally do not have access to the commercial paper markets for short-term credit needs. Other lenders (e.g., out-of-market banks or brokerage firms) were found to have an insignificant presence in these markets, and little likelihood of expanding. Therefore, these firms typically have limited actual or potential effect on pricing of loan products.
Since middle market businesses often cannot be served by smaller banks, they generally are limited to the largest banks in their areas. These businesses may have specialized needs that further limit their alternatives. As a result, middle market businesses often have fewer alternatives than do smaller businesses. Moreover, based on the available evidence in this investigation, regional and local banks tend to be the only firms that can offer these products at prices that constrain local bank pricing.
Evidence gathered in the investigation indicated that, while thrifts compete in retail banking, thrifts in the western states (including federal and state savings banks) generally are not active competitors for business banking services. Almost without exception, thrifts are not now competitors, in any of the states affected by the proposed merger, for business loans not secured by real estate. Our interviews and other evidence indicates that it is unlikely that many thrifts would enter business banking or lending in the future, even in the event of a small but significant, nontransitory price increase for business banking services. The thrifts that had attempted to enter business banking in the past decade have withdrawn from those markets, and managers of other thrifts have learned from the thrift industry's lack of success in this area.
Our investigation also evaluated lending to business customers by non-bank institutions, including finance companies, brokerage firms, factoring firms and credit unions. Based on interviews and other information, we determined that, in the markets we analyzed in the western states, these institutions do not appear to be able to offer competitive substitutes to most customers for business operating credit at a price or quality sufficient to restrain a small but significant, nontransitory rise in the price of bank credit.
Under the Merger Guidelines, the Department determines the relevant geographic market for each particular product market at issue. In this case, given the number of geographic areas involved, we generally based our geographic market analysis on the work done in connection with this application by the Federal Reserve Bank of San Francisco. We nonetheless considered whether, with regard to the specific products we examined, the markets defined by the Reserve Bank were too large or too small. For example, we found that, for small business and middle market customers, some areas in California are probably more integrated than the Reserve Bank's market definitions suggest.
In examining market share and concentration for business banking and lending products, we included as suppliers all banks and thrifts that currently offer business operating credit. (Based on the evidence gathered in our investigation and in prior investigations, we found that firms not now engaged in business banking would probably need to incur significant sunk costs to enter that business. Therefore, we believed that the capacity of such firms should not be included in the market.) In examining market share and concentration for retail banking products, we included as suppliers all banks and thrifts. In some markets we also included credit unions, which offer financial products and services to individual consumers. We measured concentration based on deposits, number of competitors, number of offices and, in some markets, estimated loan volumes, using the Herfindahl-Hirschman Index and the HHI thresholds described in the Department's Merger Guidelines. We also examined, in the various markets, the number and characteristics of banks and thrifts in the markets, the number of branches, and other measures of competitive influence. In each instance, our conclusions were based on evidence in addition to deposit-based concentration measures.
Our examination of all of the market share and concentration evidence that we assembled indicated that there would be significant increases in business banking and lending concentration in several already concentrated markets as a result of the merger (without divestiture)> including Seattle and other markets in Washington, and in Reno and Las Vegas, Nevada. Concentration in retail banking would increase significantly in Phoenix and Tucson, Arizona, among other areas.
However, examination of concentration alone does not exhaust the issues for analysis of competitive impact. In conducting our analysis, we carefully reviewed evidence of potential competitive effects, such as conditions that would make coordinated or unilateral price increases likely or unlikely to occur and succeed. The evidence we have gathered to date (including interviews, documents, and other information) raised concerns that--in the absence of adequate divestiture--the transaction could substantially lessen competition through supracompetitive pricing by leading banks, particularly in markets in Washington and Nevada.
Our investigation also indicated that, primarily because of the number and characteristics of competitors, the information gathered through interviews of banks and customers, and the divestitures that had been proposed by BAC through January 29, 1992, significant adverse competitive effects were unlikely in most of the markets in California and Oregon in which HHI calculations had indicated threshold level concerns.
BAC had only recently entered Arizona, through the acquisition of insolvent thrifts, and had not developed a significant business banking or lending presence. Based on this history, on evidence collected through interviews in this investigation and from financial data, and on evidence from prior investigations, we concluded that there was not likely to be a significant adverse effect on competition in Arizona business banking or lending. In considering the effect of the merger on retail banking competition, the Department carefully evaluated the most current deposit data for major Arizona markets to determine the extent to which deposits had run off since June 1990. Our analysis showed that HHI calculations based on 1990 data significantly overstated the concentration in these retail markets.
Since retail deposit banking was our principal remaining area of concern in Arizona, we also examined concentration levels based on total deposits of banks, thrifts and credit unions. By that measure post-merger, post-divestiture concentration in Phoenix and Tucson does not exceed the Merger Guidelines' 1800 HHI point threshold even without adjusting for deposit runoff. For these reasons, and based on the number and identity of competitors in these markets, other information and close scrutiny of the specific divestitures proposed by BAC, the Department concluded that significant adverse competitive effects were unlikely in major markets in Arizona.
We also considered the prospects for timely and sufficient entry or expansion in Seattle, Washington, the market we believed would most likely suffer significant adverse competitive effects. We tentatively concluded that, whereas entry and expansion might occur, such entry and expansion was unlikely to be of sufficient magnitude to prevent or defeat a coordinated anticompetitive price rise. Although we did not undertake to analyze entry fully in all of the markets at issue, we concluded that such an analysis would not have led us to seek additional divestitures.
Department staff advised BAC and SPC of its tentative concerns and, in response, BAC agreed to make additional divestitures intended to satisfy those concerns. These additional divestiture commitments are detailed in BAC's supplemental letter of March 9, 1992. As more fully described below, BAC has agreed to divest branches, vault and operations facilities, deposits and earning assets (in particular commercial loans) of the type and volume necessary to provide reason to believe that the Washington and Nevada business banking and lending markets, in particular, will not become significantly less competitive as a result of this transaction. BAC has also agreed to divest branches in certain local markets in Arizona, Oregon and Washington, in which the Department had concerns for small business or retail banking.
The Department believes that the appropriate aim of divestitures is to introduce new and viable competitors whose presence will mean that the market is substantially as competitive as it was before the merger. The divestitures agreed to by BAC should address these goals. In Seattle, Washington, and in Las Vegas and Reno, Nevada, the divestitures should create a major competitor with an extensive branch network and a significant competitive presence in middle market, small business and retail banking and lending. The divestitures agreed to will create the fifth largest bank in Arizona statewide, in Phoenix and in Tucson, replacing the retail banking competitor lost by the merger. The divestiture will also ensure that there is no substantial lessening of competition in local markets in all five states.
By letter to Molly S. Wassom, of the Board's staff, and Kenneth R. Binning, of the San Francisco Reserve Bank, dated March 9, 1992, BAC has made the following commitments:
In total, these commitments require BAC to divest 211 branches with approximately $8.8 billion in deposits, and to offer to divest approximately $2.7 billion in loans.
On the basis of our current information, we conclude that, subject to the foregoing commitments to divest the above listed branches, loans and other assets to competitively suitable purchasers, all as more fully set forth above, and pursuant to an enforceable order of the Board incorporating these conditions, the proposed transaction will not have a significantly adverse effect upon competition.
We have also considered whether the purchasers identified in BAC's letter are competitively suitable purchasers. Based on our current information, we conclude that Union Bank is a competitively suitable purchaser for the branches it is acquiring in California, and that U.S. Bancorp is a competitively suitable purchaser for the branches it is acquiring in Nevada. The Department is in the process of considering whether U.S. Bancorp is also a competitively suitable purchaser for all branches it is acquiring in northern California. We understand that U.S. Bancorp has agreed to divest branches in the Red Bluff, Williams and Yreka markets, if the Board so directs.