In September 1999, the Federal Reserve Board approved the merger of Fleet Financial Group, Inc. and BankBoston Corp.—two of the largest bank holding companies in New England. The two organizations had total deposits of $69.7 billion and $48.5 billion and were direct competitors in 18 local markets. The nation's antitrust laws prohibit mergers that substantially lessen competition in any banking market, so how could a merger of this magnitude have been approved? To maintain healthy competition, regulators will often require merging banks to sell some offices in markets where they compete directly. This is known as divestiture. In this case, Fleet agreed to the largest divestiture in U.S. banking history (see table).

Divestiture has become an important antitrust remedy in banking over the past 20 years.1 One development that promoted divestiture was the Fed's decision in the 1980s to adopt the merger guidelines developed by the Department of Justice (DOJ).2 This reduced uncertainty and gave banks a clear indication of what was acceptable under the antitrust laws. Banks discovered that, in many cases, divestiture of a relatively small number of offices could bring a questionable merger into compliance with the law. Divestiture was also promoted by the relaxation of bank branching restrictions in the 1990s which increased the size and geographic scope of bank mergers. Divestiture is typically not practical in a merger of two small banks, but for mega-mergers—like those that characterized the past decade—it is often the only means to gain regulatory approval.

Diagnosing the Problem

U.S. banking law requires an antitrust review of proposed bank mergers and prohibits mergers that have substantial anti-competitive effects.3 To understand how divestiture can remedy these effects, it's first necessary to understand how such effects are measured in the markets where firms compete.

From an antitrust viewpoint, a market has two components: the product market and the geographic market. Banking industry definitions of these markets have been shaped by U.S. Supreme Court decisions. The product market refers to the actual products supplied by the merging firms. Commercial banks supply a variety of products, many of which—for example, deposit accounts and personal loans—are also provided by non-bank firms. The Supreme Court has ruled that the appropriate banking product market is the "cluster" of products and services provided by commercial banks, since the convenience to the consumer of having the products clustered together raises the cluster's value above the sum of its parts. Total deposits are typically used as a proxy to measure a bank's capacity to provide the product cluster.

The geographic market is the specific area where the products are sold. The Court has determined that convenience issues restrict many bank customers to their own local communities, making most banking markets local in nature. As a result, the primary focus of the antitrust review of bank mergers—even those involving giant, multi-state banks—is on the competitive effects in the local markets where the banks do business.

To apply these principles to a particular case, it's necessary to specify the geographic boundaries of local market areas and calculate the concentration of banking services in each market where the merging banks compete. Most analysts prefer the Herfindahl-Hirschman Index (HHI) over other market concentration measures because it accounts for both the number and relative size of firms in the market. Calculating the HHI is simple. First, calculate the percentage of total market deposits held by each bank. This is the bank's market share. Second, square the market shares. Finally, add the squared market shares. The greater the resulting HHI, the greater the market concentration and the less the presumed competition in the market. According to the DOJ guidelines, any market with an HHI above 1800 is highly concentrated. A merger that results in an HHI of less than 1800, or that increases the HHI by less than 200 in a highly concentrated market, is not considered anti-competitive and is generally approved.

Mergers in which the HHI increases by more than 200 points in a highly concentrated market violate the DOJ guidelines.4 While these mergers are not summarily denied, they are closely examined for any factors that might mitigate the potential reduction in competition resulting from the merger.5 If no such factors are discovered, the merger will likely be denied. This is where divestiture can be useful. Although competitive problems often arise in only a few of the markets where merging banks do business, just one problem market is enough to deny the entire merger. By divesting enough offices in problem markets, the changes in the HHI can be brought within ranges to satisfy the antitrust laws, and the merger can proceed.

Divest Two Offices and Call Me in the Morning

Bank mergers need approval from the appropriate bank regulator and the DOJ.6 In practice, the Federal Reserve and the DOJ supervise almost all mergers that require divestiture. Both agree that divestiture is an important remedy, but they differ about how it should be prescribed.7 The Fed does not usually specify particular offices to be sold or stipulate the details of divestiture. It generally takes the position that any full-service office can be a viable competitor and provide a full range of banking services to local customers. The Fed typically prefers that offices be divested to banks not already in the local market so that new competition is introduced; however, divestitures may also be to competitors already in the local market, as long as the resulting change in the HHI meets DOJ guidelines.

The DOJ and the Fed generally approach divestiture differently because each has its own definition of a bank's product market. The Fed continues to use the traditional commercial banking "cluster" product market—a definition that the DOJ abandoned in the 1980s, citing a new competitive environment with increased competition from non-bank firms. The DOJ instead began to focus on specific bank products, usually loans to small- and medium-sized businesses, since these firms have fewer non-bank credit alternatives than other bank customers.8 Because the DOJ believes that some bank offices are better positioned than others to compete in this narrow product market, it often dictates many divestiture details to ensure that the divested offices can become viable competitors.

The merger of Fleet and BankBoston is a good example of DOJ intervention in divestiture agreements.9 The DOJ required that 278 of the 306 divested offices be sold in one lot to a single buyer. Fleet was also required to divest the majority of its small- and medium-sized business loans, although no bad loans could be divested. The DOJ also insisted that divested offices retain employees and managers, and Fleet agreed not to hire the divested workers for at least three years. The intent of all these DOJ requirements was that the divested offices would have the resources and expertise to retain old customers and compete vigorously for new commercial loans.

The banking industry has experienced unprecedented merger activity in the past 20 years. Some of these proposed mergers held serious antitrust implications in local banking markets. Divestiture of offices in those markets remedied many of the problems, permitting the mergers to proceed. In most cases, only small divestitures were necessary, but, for some of the largest mergers, hundreds of offices and billions of deposit dollars had to be divested. If bank consolidation continues in the future, divestiture is likely to remain a key prescription for maintaining healthy competition.

Endnotes

1. See Burke (1998). [back to text]
2. The DOJ guidelines are applicable to mergers in all industries, not just banking. [back to text]
3. Holder (1993) contains a detailed description of the antitrust review of bank mergers. [back to text]
4. The guidelines actually state than an increases of more than 50 points is aunacceptable in a highly concentrated market. However, banks are heald to a less-restrictive standard in light of the substantial competition they face from non-depository financial institutions. [back to text]
5. A wide variety of factors may mitigate anti-competitive effects. Factors that measure competition from thrifts and non-bank firms, public benefits and the attractiveness of the a market for entry by new banks are typically considered. [back to text]
6. The Federal Reserve regulates bank holding companies and state member banks. The Comptroller of the Currency oversees national banks, while the Federal Deposit Insurance Corp. regulates state banks that are not Fed members. The DOJ reviews all three regulators' decisions and can sue to block any mergers it finds anti-competitive. [back to text]
7. See Burke (1998). [back to text]
8. See Baker (1992). The courts ruled against this narrow product market in the only case ever litigated on this issue. The DOJ has continued to focus on commercial loans, however. One side effect of divestiture has been the almost total elimination of anti-trust litigation in banking over the past 20 years. Without such litigation, the courts have not clarified or updated their earlier product market definition. [back to text]
9. See Keenan (1999). [back to text]

References

Baker, Donald I. "Searching for an Antitrust Beacon in the Bank Merger Fog," The Antitrust Bulletin (Fall 1992), pp. 651-66.

Burke, Jim. "Divestiture as an Antitrust Remedy in Bank Mergers," Finance and Economics Discussion Series Working Paper 1998-14, Board of Governors of the Federal Reserve System (February 1998).

Holder, Christopher L. "Competitive Considerations in Bank Mergers and Acquisitions: Economic Theory, Legal Foundations, and the Fed," Economic Review, Federal Reserve Bank of Atlanta (January/February 1993), pp. 23-36.

Keenan, Charles, "Justice Dept. Aims to Make Deal Spinoffs Competitive," American Banker, December 15, 1999, p.1.