The Federal Reserve Board eagle logo links to home page

Skip to: [Printable Version (PDF)] [Bibliography] [Footnotes]
Finance and Economics Discussion Series: 2012-51 Screen Reader version

Consolidation and Merger Activity in the United States Banking Industry from 2000 through 2010

Robert M. Adams1
August 8, 2012

Keywords: Banking, mergers, antitrust

Abstract:

This study investigates trends in consolidation and merger activity in the United States banking industry from 2000 through 2010. Over this period, the U.S. banking industry has consistently experienced over 150 mergers annually, with the largest banking organizations holding an increasing share of banking assets. While the industry has undergone considerable consolidation at the national level, local banking markets have not experienced significant increases in concentration. The dynamics of consolidation raise concerns about competition, output, efficiency, and financial stability. This study uses a comprehensive proprietary data set to examine mergers and acquisitions involving banks and thrifts. The methodology in this paper expands the definition of mergers to include more types of transactions than previous studies on bank mergers.
JEL Classification: G2, L89


1 Introduction

Merger activity and overall consolidation are of particular interest in the U.S. banking industry. Since 1980, the structure of the U.S. banking industry has changed considerably, with over 10,000 mergers involving more than $7 trillion in acquired assets taking place. Furthermore, the number of institutions has declined dramatically over this period, and the concentration of assets held by the largest institutions has increased. There were 19,069 banks and thrifts operating in the U.S. in 1980 and 7,011 in 2010, a decline of over 60 percent. In 1980, the 10 largest banking organizations held only 13.5 percent of banking assets, increasing to 36 percent by 2000.2 By 2010, the 10 largest organizations held approximately 50 percent of banking assets. This paper updates previous work on bank merger trends in the U.S. and considers bank merger activity from 2000 through 2010.

As consolidation in the banking industry continues, banking antitrust policy plays a considerable role in shaping how the industry changes. Banking differs from most other industries because mergers and acquisitions must be approved by the relevant bank regulatory authority.3 The Bank Holding Company Act of 1956 and the Bank Merger Act of 1966 (and amendments) both define what types of bank transactions require regulatory approval. Along with the financial institution regulators, the Department of Justice (DOJ) also analyzes bank merger transactions for potential antitrust concerns. Unlike other regulators, however, the DOJ cannot deny a merger application; rather, they have to file an injunction to block or undo a merger.

The banking industry has undergone significant regulatory changes in the past 15 years. These regulatory changes have had significant effects on competition and structure, with some changes acting as the impetus for recent merger waves. For example, the Riegle-Neal Interstate Banking and Branching Efficiency Act of 1994 allowed branch banking beyond one state and throughout the United States, and the Gramm-Leach-Bliley Act of 1999 (Financial Services Modernization Act) allowed banks to enter other financial markets and provide additional financial services. Both of these laws are potential causes for the increase in bank mergers. With such regulatory changes and the overall changes in the bank industry structure, banking has moved from a fragmented industry with banks operating only in individual states to a more unified industry, dominated by banks operating in large regions of the country.

Several previous studies discuss bank merger activity over the past 50 years. Three studies by Rhoades (1985, 1996, 2000) consider bank merger activity from 1960 through 1998, and Pilloff (2004) considers merger activity from 1994 through 2003.4 Wheelock (2011) analyzes merger activity during the financial crisis from 2007 through 2010. These papers differ in their sources for bank merger data and how they count bank mergers. The next section will discuss these differences in greater detail.

Research on the motivations for and causes and effects of bank mergers is vast and covers numerous facets of the topic. The data sources employed in this literature vary tremendously as well. Most studies of bank mergers in the United States use regulatory documents, stock price data, and National Information Center (NIC) data, in addition to other sources. These data sources vary in their coverage of bank mergers.5 This study will not evaluate the quality of the different data sets or evaluate the literature on bank mergers; instead, it will describe a single comprehensive data set of bank merger transactions.

Several survey articles provide overviews of the research on bank mergers: They include Berger, Demsetz, and Strahan (1999); Amel, Barnes, Panetta, and Salleo (2004); and DeYoung, Evanoff, and Molyneux (2009).

This paper is organized in the following manner. In Section I, the data construction is discussed, and the methodology is compared to that of others. In Section II, overall merger trends and consolidation are described on a national level. In Section III, local market trends are described. Section IV concludes.

2 Merger Data

This study only covers transactions that involve commercial banks, savings banks, savings and loan associations, bank holding companies, thrift holding companies, or foreign bank organizations. Bank holding companies, thrift holding companies, and foreign banking organizations are firms that own banks or thrifts along with other subsidiaries.6 Bank holding companies are regulated by the Federal Reserve, while thrift holding companies were regulated by the Office of Thrift Supervision (OTS) during the time period under study. Thrift holding companies are now regulated by the Federal Reserve.7

Previous studies have measured and defined mergers differently. Rhoades (1985, 1996, 2000) considered only transactions where one banking organization purchases at least a 25 percent ownership share of the target. In his data, the acquirer is an active operating entity (for at least one year) rather than a de novo or non-operating bank. Both parties are bank holding companies or commercial banks, and both are either U.S. domestic banks or are owned by one. The target is not failing or likely to fail as judged by the regulator. Mergers are recorded in the year they are approved by the regulator. When multiple firms are acquiring a target entity, Rhoades treats the largest institution as the acquiring firm. Finally, when a multibank holding company is acquired, each acquired commercial bank in that holding company is treated as a separate deal.

Rhoades obtained his data from the merger filings with the various bank regulators. He uses documents from the Office of the Comptroller of the Currency (OCC), the Federal Deposit Insurance Corporation (FDIC), the OTS, and the Federal Reserve Board (FRB) as a basis to find merger transactions. His data only measure transactions filed with the OCC, the FDIC, the OTS, or the FRB and do not include transactions that did not require federal bank regulatory approval.8

Pilloff (2004) measures mergers in a slightly different manner from Rhoades. He uses a private proprietary data source, SNL Financial (www.snl.com), to identify mergers. SNL Financial records bank mergers from several sources including regulator publications and the popular press. Pilloff (2004) supplements and verifies the SNL financial data using regulatory data from the NIC. Pilloff's analysis defined a bank merger as a transaction where both the acquirer and target institutions (or one of their banking subsidiaries) is a commercial bank, savings bank, savings and loan, or industrial bank chartered in the United States. Acquisitions of banks by private investors or nonbanking firms and newly formed bank holding companies with no active bank subsidiaries are not counted as mergers. Additionally, acquisitions of multibank holding companies are counted as a single merger rather than as multiple mergers. Unlike Rhoades, Pilloff uses the actual date of consummation as noted by SNL as the date of the merger rather than the date of approval. Pilloff also requires that majority ownership be acquired for a merger. In other words, a high holding company must own at least 50 percent of any subsidiaries. His data also do not include failed or failing institution transactions.

Unlike Pilloff (though somewhat like Rhoades), Wheelock (2011) focuses on bank mergers where the acquiring institution absorbs and rebrands the target institution. Acquisitions by bank holding companies where the target institution becomes a subsidiary and retains its own name are not counted as mergers in his analysis. For some mergers, this methodology does not result in any difference, but for a merger involving a bank holding company, the results could substantially differ. For example in Wheelock (2011), of the largest transactions that took place during the financial crisis, two transactions involved institutions owned by the same bank holding company:9 the Wachovia Bank, N.A., merger with Wachovia Mortgage, FSB, and the Fifth Third Bank (Cincinnati) merger with Fifth Third Bank (Grand Rapids), which occurred about eight years after they become subsidiaries of the same bank holding company.10 Like Rhoades' methodology, Wheelock's methodology could either potentially result in multiple transactions when only a single bank holding company is acquiring another bank holding company or potentially have an inappropriate transaction date.

We do not use Wheelock's definition for three reasons. First, antitrust analysis scrutiny from either banking regulators or competition authorities occurs with the change in control of an institution. Institutions with the same high holder are unlikely to be competing. A change in control, however, changes incentives, so economic analysis is warranted at that point. Second, the timing of a transaction almost always differs for the bank holding company acquisition than for the subsequent merger. The economic motivations for both transactions may differ substantially because of the difference in timing and structure of the deal. Sometimes these transactions could be years apart. Finally, as in Rhoades's methodology, transactions between bank holding companies could count as multiple smaller transactions. These types of transactions would overstate the number of transactions that took place and understate the size of certain transactions. For example, in Wheelock's paper, the Bank of America merger with two LaSalle Bank subsidiaries is counted as two transactions that occurred on the same date.

While this study takes a slightly different approach than Pilloff and Rhoades, our methodology is most similar to Pilloff's method. Like Pilloff, we use data from SNL Financial to identify bank mergers and use NIC data to verify transactions. Additionally, we consider all transactions involving commercial banks, savings banks, savings and loan associations, industrial banks, thrift holding companies, and bank holding companies as the acquiring and the target institutions. We also use the date of consummation rather than the date of approval as the date the merger occurs. Unlike Pilloff's methodology, which requires a firm to obtain a 50 percent controlling interest, this study includes all transactions where the acquiring institution obtains a controlling interest of at least 25 percent.11 This percentage was chosen because the Bank Holding Company Act of 1956, which is the basis for most antitrust banking authority, defines control at 25 percent ownership or above, and these types of transactions have to be approved by bank regulators.12 This criterion is used for bank-to-bank transactions and determines the high holding institution(s) of the acquirer and the target. Some institutions can have multiple high holders, and all high holders are checked to determine if the transaction involves the same high holder for acquirer and target. A transaction is not counted as a merger if the high holder for the acquirer is also a high holder for the target.

Unlike previous studies, failed or failing institutions are included in the data.13 It is important to include these transactions because they undergo the same application and approval process as any other transaction (albeit with some modifications if necessary).14 While the reasons for merging are potentially different, the application is evaluated and could result in a denial. Rhoades's and Pilloff's methods of counting mergers would result in far fewer mergers from 2008 to 2010 because during these years a large number of failure-based mergers occurred.15

The data in this study also only identify the high holder of an institution as the acquiring firm, without any information about intermediate holding companies. The high holder of an institution is defined as the top-most firm that owns a controlling interest of the depository and all mid-tier holding companies. The majority of institutions are either independent with no high holder or an institution with a single high holder.16 While this data construction does not reveal details about transactions, such as which subsidiary is purchasing the target institution, it does identify the merging parties. The main drawback is that we cannot directly identify the regulator that approved the transaction, as subsidiary firms may have different regulators.17

Once acquirer and target institutions are identified, financial information on all merger participants is collected from the Consolidated Reports of Condition and Income (Call Report), Thrift Financial Reports, and the Summary of Deposits and Branch Office Survey.18 We use the December Call Report and Thrift Financial Report for bank-level financial data before the transaction occurred. Market-level data are derived from the preceding Summary of Deposits data (the previous June 30). While this method does not give the most up-to-date data on the institutions, it allows for a balanced comparison of all merger participants in a given year. This method also mitigates some of the dating issues with merger transactions.

For many of the transactions, most of the differences in criteria for defining mergers and acquisitions do not matter, as most transactions involve a merger of two institutions or the acquisition of a single institution by a holding company. However, a handful of transactions are more complicated and the differences in methodology become more apparent with such transactions. The biggest difference occurs with the inclusion of failed or failing institutions in the years 2008 to 2010. Not including these observations in the data has a significant effect on the observed number of transactions. However, it is important to note that these types of transactions would not have had a major effect on the Pilloff (2004) results because not many banks failed during his reference period.19

4 Local Market and State Trends

In banking antitrust, most concerns arise from competition in local banking markets. Retail banking and small business banking product markets (broadly defined) are still geographically local in nature. The next Tables describe how local banking markets have been affected by bank merger activity and what type of banking markets have experienced significant banking mergers. For the purposes of this analysis, we assume that banking markets are metropolitan statistical areas (MSA), micropolitan areas, and rural counties. We use 2004 MSA definitions. According to these market definitions, there are 369 MSA markets, 578 micropolitan markets, and 1,362 rural county markets. The last part of this section considers how banks might use mergers or acquisition to expand into other markets by considering the extent to which bank mergers cross state lines.

The Effect of Bank Mergers on Local Banking Markets

Table 7 describes mergers by type of banking market. During our sample period, all MSA markets experience a merger at some point, and 92 percent of micropolitan markets experience a merger, while 63 percent of rural markets experience a merger. The average number of mergers in a market over the sample period is 11 in MSA markets, 3 in micropolitan markets, and 1 in rural county markets. Median merger values do not differ significantly from the means, with 7 in MSA markets, 2 in micropolitan markets, and 1 in rural county markets.

Share per market measures the share of target deposits or of target offices involved in mergers. Over 14 percent of market deposits and offices are acquired annually, on average, with median values closer to 10 percent. Both the average and median percent of market deposits and offices acquired increase as market population decreases (from MSA to rural counties), even though the number of mergers is much greater in MSA markets than in rural markets. Rural markets typically have fewer institutions with greater market shares, so when a merger occurs in a rural market, it usually involves a larger share of market deposits.

Remedies to alleviate the competitive effects of merger applications are sometimes required. The main remedy for merger applications is the divestiture of branches in local markets where the competitive effects cannot be mitigated by other factors. Of the 2,399 mergers that occurred, approximately 101 transactions required divestitures in 159 markets before the Federal Reserve approved the transaction.30

Next, it is important to see how merger activity affected local market concentration. Table 8 describes average Herfindahl-Hirschman Indices (HHI), average number of organizations, and average number of offices for MSA markets, micropolitan markets, and rural county markets. For the purposes of this exercise, thrift deposits are weighted at 50 percent.31 During the past decade, average local market concentration decreased in all types of local markets: MSA average concentration fell from 1,641 to 1,622, micropolitan concentration fell from 2,393 to 2,311, and rural concentration fell from 4,245 to 4,148. The mean number of institutions increased in all types of markets: The mean number of institutions increased from 48 to 54 in MSAs, from 9 to 10 in micropoiltan, and from 5 to 6 in rural markets. Finally, the mean number of branches increased in MSA and micropolitan markets and remained constant in rural markets: In MSA markets, the mean number of branches increased from 493 to 568 branches; in micropolitan, from 24 to 25 branches; and in rural, it remained constant at 10 branches. Chart 3 describes the number of markets where target institutions have complete, partial, or no branch network overlap with the acquiring institution. Almost 18 percent of bank mergers involved institutions with 100 percent of overlap between acquirer and target institutions. Almost 70 percent of these mergers involved single market institutions. About 46 percent of mergers had some but not complete overlap and almost 36 percent of mergers had no overlap between acquiring and target institutions.

Table 9 describes the percent of population affected by bank mergers on an annual basis. It also describes the percent of population where at least 10 percent of deposits were acquired in a market on an annual basis. During the past decade, 24 to 40 percent of the population annually lived in a market where a bank merger occurred, an average of about 33 percent. MSA markets followed the same trend with 34 percent of the population, on average, with a range from about 25 to 41 percent. Micropolitan and rural population percentages affected by bank mergers were much lower, averaging about 11 percent and ranging from 5 to 21 percent annually.

Markets where at least 10 percent of the deposits were acquired affect less than 5 percent of the population in most years except for the years 2001, 2004, and 2008. As discussed earlier, a number of the largest mergers occurred in these years.

The Prevalence of Bank Mergers within or across State Lines

In addition to overlap in local banking markets, we also consider in-state and out-of-state acquisitions as a potential motivation for mergers and acquisitions. Hypothetically, banks could enter other states through mergers or acquisitions. State regulations of banks still vary and some barriers to entry still exist across state lines.32 One might posit that bank mergers may be motivated by a desire to expand across state lines. Banks could acquire banks in states where they do not already have a branch presence in order to gain a foothold in the state for further expansion.33 Table 10 describes how often acquisitions involve in-state or out-of-state acquirers on an annual basis. The data clearly show that mergers are not used primarily to expand into additional states. In every year of our sample, over 70 percent of acquisitions involve institutions that both have a presence in the same states.34 Table 11 weights the in-state and out-of-state acquisitions by deposits or offices.35 The outcome is the same looking at either deposits or offices: 96 percent of deposit-weighted transactions and 94 percent of office-weighted transactions involve an acquirer and target with a presence in the same state.

5 Conclusion

Bank merger activity over the past decade has continued at a fairly steady pace. The recent financial crisis resulted in a decrease in the number of mergers and a shift in the types of mergers from traditional mergers or acquisitions to mainly acquisitions of failed or distressed institutions. The decade saw a dramatic increase in concentration of banking assets at the national level by the largest institutions. Even with that increase, around 7,000 institutions remain.

Merger activity should continue at a steady pace. During the past decade, most mergers and acquisitions occurred between small and medium-sized institutions with less than $30 billion in assets. With around 7,000 institutions, of which 99 percent are small to medium sized, this trend should continue into the future. However, the new Dodd-Frank regulations will increase the regulatory hurdles for large institutions to complete a transaction, regardless of the size of the target.



References

Amel, Dean, Colleen Barnes, Fabio Panetta, and Carmelo Salleo (2004). "Consolidation, Efficiency and Competition in the Financial Sector: A Review of the International Evidence," in Michele Bagella, L. Bechettik, Iftekhar Hasan, and W.C. Hunter, eds., Research in Banking Finance, vol. 4: Monetary Integration, Markets and Regulation. Oxford: Elsevier Ltd., pp. 209-30.

Berger, Allen N., Rebecca S. Demsetz, and Philip E. Strahan (1999). "The Consolidation of the Financial Services Industry: Causes, Consequences, and Implications for the Future," Journal of Banking and Finance, vol. 23, pp. 135-94.

DeYoung, Robert, Douglas D. Evanoff, and Philip Molyneux (2009). "Mergers and Acquisitions of Financial Institutions: A Review of the Post-2000 Literature," Journal of Financial Services Research, vol. 36, pp. 87-110.

Pilloff , Steven J. (2004). Bank Merger Activity in the United States, 1994-2003, Staff Study 176. Washington: Board of Governors of the Federal Reserve System, May.

Rice, Tara, and Christian A. Johnson (2008). "Assessing a Decade of Interstate Bank Branching," Washington and Lee Law Review, vol. 65, pp. 73-127.

Rhoades, Stephen A. (1985). Mergers and Acquisitions by Commercial Banks, 1960-83, Staff Study 142. Washington: Board of Governors of the Federal Reserve System, January.

Rhoades, Stephen A. (1996). Bank Mergers and Industrywide Structure, 1980-94, Staff Study 169. Washington: Board of Governors of the Federal Reserve System, January.

Rhoades, Stephen A. (2000). Bank Mergers and Banking Structure in the United States, 1980-98, Staff Study 174. Washington: Board of Governors of the Federal Reserve System, August.

Wheelock, David C. (2011). "Banking Industry Consolidation and Market Structure: Impact of the Financial Crisis and Recession," Federal Reserve Bank of St. Louis REVIEW, vol. 93 (November/December), pp. 419-38.


Table 1: Acquired Assets, Deposits, and Offices 2000 to 2010
Year Number of Mergers Assets Mean Assets Median Assets Amount Assets Industry % Deposits Mean Deposits Median Deposits Amount Deposits Industry % Offices Mean Offices Median Offices Amount Offices Industry %
ALL 2403 1,705,992 134,713 4,099,498,050 - 1,077,202 111,820 2,588,517,311 - 17 3 41,615 -
2000 257 703,853 121,363 180,890,307 2.7 382,664 103,228 98,344,720 2.3 11 4 2,837 3.3
2001 235 1,422,742 133,910 334,344,324 4.7 935,266 107,669 219,787,415 4.8 20 4 4,649 5.4
2002 204 733,747 110,640 149,684,286 1.9 427,982 97,353 87,308,313 1.8 9 3 1,747 2.0
2003 203 450,153 108,540 91,381,062 1.1 334,370 88,311 67,877,014 1.3 9 3 1,776 2.0
2004 259 3,158,784 155,809 818,124,982 8.9 1,886,919 134,596 488,711,932 8.6 33 4 8,496 9.5
2005 207 566,092 130,743 117,181,125 1.2 400,936 108,475 82,993,846 1.3 10 3 1,972 2.1
2006 256 1,230,822 132,041 315,090,507 3.0 734,649 110,206 188,070,266 2.8 11 3 2,857 3.0
2007 249 1,442,120 140,215 359,087,886 3.1 914,140 120,049 227,620,907 3.3 13 4 3,319 3.4
2008 195 7,028,747 109,740 1,370,605,745 11.1 438,738 92,305 855,540,691 11.4 50 3 9,670 9.8
2009 158 1,629,688 195,617 257,490,686 2.2 1,179,248 166,917 186,321,236 2.4 18 4 2,831 2.8
2010 180 586,762 170,898 105,617,140 0.9 477,450 149,056 85,940,971 1.1 8 3 1,461 1.5


Table 2.1: Acquirer and Target Comparison by Assets
Year Mean Acquirer Mean Target Mean Ratio Median Acquirer Median Target Median Ratio
ALL 14,864,079  1,705,992 39.7% 1,034,541  134,713 14.2%
2000 15,917,224 703,853 25.3% 1,029,038 121,363 13.2%
2001 10,221,220 1,422,742 36.5% 902,498 133,910 14.3%
2002 10,977,785 733,747 22.8% 778,875 110,640 13.9%
2003 9,102,179 450,153 23.2% 946,753 108,540 12.6%
2004 14,250,810 3,158,784 31.7% 1,347,064 155,809 15.4%
2005 9,124,718 566,092 30.6% 1,120,961 130,743 12.6%
2006 16,661,477 1,230,822 24.8% 1,263,804 132,041 13.3%
2007 22,755,425 1,442,120 34.3% 1,035,305 140,215 14.0%
2008 29,035,911 7,028,747 33.3% 599,481 109,740 20.0%
2009 13,924,181 1,629,688 114.1% 913,657 195,617 24.0%
2010 9,806,776  586,762 94.7% 1,556,374  170,898 11.7%


Table 2.2: Acquirer and Target Comparison by Deposits
Year Mean Acquirer Mean Target Mean Ratio Median Acquirer Median Target Median Ratio
ALL 9,216,422  1,077,202 43.1% 817,989  111,820 15.4%
2000 10,030,609 382,664 34.4% 826,490 103,228 14.0%
2001 6,112,398 935,266 38.0% 715,130 107,669 14.3%
2002 6,470,681 427,982 24.0% 628,923 97,353 15.6%
2003 5,883,324 334,370 24.4% 783,395 88,311 14.6%
2004 8,902,116 1,886,919 32.9% 954,246 134,596 16.1%
2005 5,358,886 400,936 28.5% 872,840 108,475 13.2%
2006 11,105,217 734,649 25.5% 982,227 110,206 13.7%
2007 13,041,376 914,140 41.6% 817,989 120,049 15.5%
2008 17,185,413 4,387,388 35.6% 488,041 92,305 19.4%
2009 8,914,595 1,179,248 124.9% 697,212 166,917 25.5%
2010 7,520,062  477,450 99.9% 1,170,228  149,056 12.9%


Table 2.3: Acquirer and Target Comparison by Offices
Year Mean Acquirer Mean Target Mean Ratio Median Acquirer Median Target Median Ratio
ALL 144 17 63.0% 21 3 17.6%
2000 210 11 35.5% 22 4 19.5%
2001 121 20 85.8% 23 4 16.7%
2002 121 9 30.3% 19 3 17.6%
2003 111 9 33.1% 21 3 14.3%
2004 134 33 37.0% 24 4 20.0%
2005 79 10 132.8% 22 3 15.8%
2006 155 11 35.1% 21 3 14.7%
2007 169 13 37.8% 21 4 20.0%
2008 228 50 83.5% 14 3 25.0%
2009 148 18 142.8% 17 4 25.0%
2010 81 8 82.4% 23 3 14.8%


Table 3: Concentration of Assets and Deposits 2000 to 2010

Year Top 10 % of Total Deposits Top 50 % of Total Deposits Top 100 % of Total Deposits Top 10 % of Total Assets Top 50 % of Total Assets Top 100 % of Total Assets
2000 29.8 54.5 64.3 36.0 61.7 70.7
2001 33.5 56.6 65.4 38.5 62.8 71.2
2002 34.2 57.3 65.7 39.4 63.3 71.6
2003 34.3 57.4 66.0 39.7 63.3 71.8
2004 39.1 60.4 68.1 43.7 66.1 73.8
2005 38.8 61 68.6 43.2 66.5 74.2
2006 39.6 61.8 69.6 45.1 68.0 75.3
2007 40.5 63.3 70.7 46.8 70.2 76.8
2008 43.4 65.7 72.1 50.9 72.0 77.8
2009 44.8 66.1 72.1 49.8 71.3 76.9
2010 45.6 66.6 72.9 50.1 71.6 77.5


Table 4: Pooled Acquired Assets, Deposits, and Offices by Acquirer Type 2000 to 2010
Acquirer type Target type Number of Mergers Assets Mean Assets Median Assets Amount Assets % of total Deposits Mean Deposits Median Deposits Amount Deposits % of total Offices Mean Offices Median Offices Amount Offices % of total
ALL ALL 2403 1,705,992 134,713 4,099,498,050 100 1,077,202 111,820 2,588,517,311 100 17 3 41,615 100
BANK BANK 1874 1,618,683 124,943 3,033,411,277 74 1,046,762 106,720 1,961,631,224 76 17 3 31,698 76
BANK THRIFT 245 3,378,906 227,827 827,831,897 20 1,934,930 166,925 474,057,867 18 29 5 7,066 17
THRIFT BANK 112 580,772 152,388 65,046,489 2 450,812 131,904 50,490,938 2 9 4 1,008 2
THRIFT THRIFT 172 1,007,026 174,629 173,208,387 4 594,984 124,335 102,337,282 4 11 4 1,843 4


Table 5: Bank Mergers by Type Broken Out by Year
Year Total Mergers Bank-Bank Bank-Thrift Thrift-Bank Thrift-Thrift
ALL 2403 1874 245 112 172
2000 257 189 32 8 28
2001 235 183 24 9 19
2002 204 150 23 8 23
2003 203 148 22 13 20
2004 259 195 25 16 23
2005 207 157 22 13 15
2006 256 214 24 8 10
2007 249 200 20 14 15
2008 195 158 19 9 9
2009 158 131 18 7 2
2010 180 149 16 7 8


Table 6: Top 30 Deals by Target Assets 2000 to 2010 ($000s)
Buyer Target Rank Year Target assets Target deposits Target offices
WELLS FARGO & COMPANY WACHOVIA CORPORATION 1 2008 706,137,276 430,029,942 3367
J.P. MORGAN CHASE & CO. BANK ONE CORPORATION 2 2004 375,146,701 170,160,145 1880
J.P. MORGAN CHASE & CO. WASHINGTON MUTUAL, INC. 3 2008 329,974,652 182,409,135 2239
BANK OF AMERICA CORPORATION FLEETBOSTON FINANCIAL CORPORATION 4 2004 211,690,609 149,911,597 1534
PNC FINANCIAL SERVICES GROUP, INC. NATIONAL CITY CORPORATION 5 2008 139,835,067 84,633,240 1568
WACHOVIA CORPORATION GOLDEN WEST FINANCIAL CORPORATION 6 2006 134,537,638 65,131,102 287
BANK OF AMERICA CORPORATION LASALLE BANK CORPORATION 7 2007 127,146,381 60,616,782 406
FIRSTAR CORPORATION U.S. BANCORP 8 2001 106,107,923 65,185,303 1090
FIRST UNION CORPORATION WACHOVIA CORPORATION 9 2001 85,358,742 49,757,706 691
BANCO SANTANDER, S.A. SOVEREIGN BANCORP, INC. 10 2009 78,293,203 48,685,250 748
CITIGROUP, INC. GOLDEN STATE BANCORP INC. 11 2002 68,550,720 27,626,307 353
CHASE MANHATTAN CORPORATION (THE) J.P. MORGAN & COMPANY, INCORPORATED 12 2000 66,643,568 6,598,477 4
WACHOVIA CORPORATION SOUTHTRUST CORPORATION 13 2004 59,892,984 37,607,549 727
REGIONS FINANCIAL CORPORATION AMSOUTH BANCORPORATION 14 2006 56,860,727 37,861,341 689
MITSUBISHI UFJ FINANCIAL GROUP, INC. UNIONBANCAL CORPORATION 15 2008 54,794,172 41,214,629 342
TORONTO-DOMINION BANK COMMERCE BANCORP, INC. 16 2008 52,717,231 48,361,916 476
ROYAL BANK OF SCOTLAND GROUP PLC CHARTER ONE FINANCIAL, INC. 17 2004 49,289,202 31,856,960 684
FLEETBOSTON FINANCIAL CORPORATION SUMMIT BANCORP 18 2001 49,158,520 33,733,502 511
BANK OF NEW YORK COMPANY, INC. MELLON FINANCIAL CORPORATION 19 2007 40,753,621 23,254,118 46
REGIONS FINANCIAL CORPORATION UNION PLANTERS CORPORATION 20 2004 36,305,225 26,773,649 711
BANCO BILBAO VIZCAYA ARGENTARIA, S.A. COMPASS BANCSHARES, INC. 21 2007 35,952,113 24,403,298 421
WASHINGTON MUTUAL, INC. DIME BANCORP, INCORPORATED 22 2002 33,903,731 16,248,518 125
TD BANK FINANCIAL GROUP BANKNORTH GROUP, INC. 23 2005 31,993,620 21,767,675 412
WELLS FARGO & COMPANY FIRST SECURITY CORPORATION 24 2000 29,464,513 16,003,988 339
SUNTRUST BANKS, INC. NATIONAL COMMERCE FINANCIAL CORPORATION 25 2004 27,485,095 19,052,209 504
NORTH FORK BANCORPORATION, INC. GREENPOINT FINANCIAL CORP. 26 2004 26,523,003 14,709,237 93
BB&T CORPORATION COLONIAL BANK 27 2009 26,059,276 18,866,063 356
WASHINGTON MUTUAL, INC. BANK UNITED CORPORATION 28 2001 22,959,211 9,885,113 156
SOVEREIGN BANCORP, INC. INDEPENDENCE COMMUNITY BANK CORP. 29 2006 20,561,837 12,000,266 126
NATIONAL CITY CORPORATION PROVIDENT FINANCIAL GROUP, INC. 30 2004 19,601,090 12,288,917 66


Table 7: Market Level Summary Stats from 2000 through 2010
  All Market type MSA Market type Micro Market type Rural Market type
Number of markets Total 2309 369 578 1362
Number of markets With mergers 1691 369 522 800
% of markets w/ mergers 73.2 100.0 90.3 58.7
Number of mergers per market Mean 3.0 11.0 2.7 1.0
Number of mergers per market Median 1.0 7.0 2.0 1.0
Number of mergers per market Max 118.0 118.0 12.0 8.0
Share per market (%) Mean deposits 14.1 8.5 13.5 23.7
Share per market (%) Median deposits 9.2 4.9 9.6 18.0
Share per market (%) Mean offices 13.9 8.2 12.8 23.9
Share per market (%) Median offices 10.0 5.9 10.0 19.2



Table 8: Summary Market Structure by Year
Year HHI MSA Number of MSA orgs. Number of MSA offices HHI Micro Number of Micro orgs. Number of Micro offices HHI Rural Number of Rural orgs. Number of Rural offices
ALL 1611 52 493 2348 9 24 4183 5 10
2000 1641 48 419 2393 9 23 4245 5 10
2001 1629 48 417 2384 9 23 4212 5 10
2002 1605 49 427 2385 9 23 4200 5 10
2003 1601 50 442 2378 9 23 4216 5 10
2004 1611 51 463 2356 9 23 4220 5 10
2005 1630 52 496 2353 9 24 4197 5 10
2006 1597 56 518 2322 9 24 4147 5 10
2007 1580 57 544 2318 10 24 4143 6 10
2008 1599 57 560 2320 10 25 4146 6 10
2009 1608 55 564 2306 10 25 4136 6 10
2010 1622 54 568 2311 10 25 4148 6 10



Table 9: Population Affected by Mergers
Year All with at least one target (% of U.S. population) MSA with at least one target (% of U.S. population) Micro with at least one target (% of U.S. population) Rural with at least one target (% of U.S. population) All acquired with at least 10% deposits (% of U.S. population) MSA acquired with at least 10% deposits (% of U.S. population) Micro acquired with at least 10% deposits (% of U.S. population) Rural acquired with at least 10% (% of U.S. population)deposits
ALL 32.7 33.6 11.1 9.9 5.7 5.7 5.2 6.6
2000 24.0 24.9 8.6 7.0 0.8 0.6 4.9 4.6
2001 32.6 33.4 14.4 13.0 6.1 6.0 9.1 9.5
2002 33.1 34.1 5.4 6.6 0.5 0.4 2.0 4.4
2003 32.9 33.9 7.7 9.0 1.1 1.0 1.9 5.6
2004 40.2 41.0 20.9 18.7 18.9 19.2 10.6 12.5
2005 33.8 34.4 14.1 11.8 0.5 0.3 4.3 7.1
2006 29.0 29.8 7.6 6.2 1.8 1.7 3.6 3.7
2007 29.1 30.0 9.1 8.3 6.6 6.7 4.1 5.5
2008 34.0 35.2 14.1 11.0 24.9 25.8 8.6 7.8
2009 40.2 41.4 7.5 6.2 1.0 0.9 2.7 4.1
2010 31.0 31.4 13.1 11.5 0.3 0.1 5.3 8.0


Table 10: Percent In-State and Out-of-State Acquisitions
Year Mergers In-state acquirer Mergers In-state acquirer % of total Out-of-state acquirer Mergers Out-of-state acquirer % of total
ALL 2403 1833 76.3 570 23.7
2000 257 199 77.4 58 22.6
2001 235 187 79.6 48 20.4
2002 204 160 78.4 44 21.6
2003 203 159 78.3 44 21.7
2004 259 185 71.4 74 28.6
2005 207 145 70.0 62 30.0
2006 256 195 76.2 61 23.8
2007 249 194 77.9 55 22.1
2008 195 158 81.0 37 19.0
2009 158 117 74.1 41 25.9
2010 180 134 74.4 46 25.6


Table 11: Percent In-State and Out-of-State Acquisitions Weighted by Deposits and Offices
Year Total Acquired Deposits In-state acquirer Deposits In-state acquirer % of total Out-of-state acquirer Deposits Out-of-state acquirer % of total Total Acquired Offices In-state acquirer Offices In-state acquirer % of total Out-of-state acquirer Offices Out-of-state acquirer % of total
ALL 2,588,517,311  2,471,853,956 95.5 116,663,355 4.5 41615 39,044 93.8 2,571 6.2
2000 98,344,720  91,997,404 93.5 6,347,316 6.5 2837 2,595 91.5 242 8.5
2001 219,787,415 207,173,219 94.3 12,614,196 5.7 4649 4,211 90.6 438 9.4
2002 87,308,313 68,757,197 78.8 18,551,116 21.2 1747 1,473 84.3 274 15.7
2003 67,877,014 62,460,785 92.0 5,416,229 8.0 1776 1,650 92.9 126 7.1
2004 488,711,932 479,605,591 98.1 9,106,341 1.9 8496 8,226 96.8 270 3.2
2005 82,993,846 74,607,139 89.9 8,386,707 10.1 1972 1,754 88.9 218 11.1
2006 188,070,266 181,282,706 96.4 6,787,560 3.6 2857 2,684 93.9 173 6.1
2007 227,620,907 218,540,393 96.0 9,080,514 4.0 3319 3,078 92.7 241 7.3
2008 855,540,691 849,591,083 99.3 5,949,608 0.7 9670 9,562 98.9 108 1.1
2009 186,321,236 165,793,419 89.0 20,527,817 11.0 2831 2,601 91.9 230 8.1
2010 85,940,971  72,045,020 83.8 13,895,951 16.2 1461 1,210 82.8 251 17.2

Chart 1: Banking Organization Total Asset Distribution

Chart 1: Banking Organization Total Asset Distribution. See link below for figure data. Chart 1 Data

Chart 2: Deals by Assets Size Category

Chart 2: Deals by Assets Size Category. See link below for figure data. Chart 2 Data

Chart 3: Market Overlap

Chart 3: Market Overlap. See link below for figure data. Chart 3 Data


Footnotes

1. Division of Research and Statistics, Board of Governors of the Federal Reserve System, Washington, DC 20551. The views expressed here are those of the author and do not necessarily reflect those of the Board of Governors or its staff. I would like to thank Courtney Carter, Ryan Fackler, Jason Scott, and Onka Tenkean for their excellent research assistance. Any remaining errors are the responsibility of the author. Return to Text
2. From 1980 to 2010, the number of thrifts declined by 75 percent, while the number of banks declined by 55 percent. In our sample from 2000 through 2010, the number of thrifts declined by 28 percent and the number of banks declined by 22 percent. Return to Text
3. In fact, the United States differs from most other countries because the bank regulators have explicit antitrust authority. In most other countries, only the competition authorities deal with antitrust concerns in banking. Return to Text
4. Rhoades (1985) considers mergers from 1960 through 1983, Rhoades (1996) considers mergers from 1980 through 1994, and Rhoades (2000) considers mergers from 1980 through 1998. Return to Text
5. For example, stock data only include banks that are publically traded, and regulator data sources do not necessarily cover transactions between state-chartered banks. Return to Text
6. By law, institutions must apply with the Federal Reserve to become a bank holding company. These institutions can have multiple tiers; for example, a bank holding company can own another bank holding company, which owns a bank. A high holder institution is the institution in the highest tier. Return to Text
7. This change occurred in 2011, one year after the passage of the Dodd-Frank Act. Return to Text
8. These types of transactions would involve non-FDIC-insured state banks. Return to Text
9. See Wheelock (2011), table 10 on page 432. Return to Text
10. Wachovia Mortgage, FSB, was previously World Savings Bank, FSB. Fifth Third Bank (Grand Rapids) was previously Old Kent Bank. Return to Text
11. The result of this methodology is that some transactions in the data used for this paper involving incremental acquisitions of shares are considered to be reorganizations rather than mergers in Pilloff's data. Return to Text
12. In fact, an investment of 5 percent or more has to be approved by bank regulators, but acquisitions of 5 to 25 percent are generally considered to be noncontrolling. Return to Text
13. Pilloff (2004) did not include failed or failing institution transactions. Rhoades (1985, 1996, 2000) included them only when a regulator made a decision on the transaction. Return to Text
14. A conflict could occur between the requirements of the Federal Deposit Insurance Corporation Improvement Act (FDICIA) and antitrust laws governing a failing bank transaction. FDICIA requires that the FDIC pursue a least-cost resolution (to the insurance fund) of failing bank transactions. Return to Text
15. The previous major period of bank failures was the savings and loan crisis in the 1980s. Return to Text
16. Some organizations have very complex holding tiers. Pilloff (2004) uses a work case scenario of 13 tiers when checking the SNL data against the NIC data. Return to Text
17. Sometimes bank holding companies will apply for a bank-to-bank acquisition. If a bank holding company applies for approval, the transaction is evaluated by the Federal Reserve. In a bank-to-bank transaction, the regulator of the applying depository considers the transaction (regardless if the bank is part of a bank holding company). Return to Text
18. The Call Report and Thrift Financial Report are collected every quarter, whereas the Summary of Deposits is collected every year as of June 30. Return to Text
19. If we compare the overlap years in this study's sample and in Pilloff's, the number of transactions do not differ substantially: 2 in 2000, 4 in 2001, 1 in 2002, and 17 in 2003. Return to Text
20. See Pilloff (2004), table 1. Merger count data available from the FDIC website include reorganizations. Return to Text
21. If we use the same methodology as Pilloff (2004), we would see a much greater drop in merger activity because he does not include transactions of failed or failing institutions in his data. Return to Text
22. While many S&P 500 indices have returned to 2006 levels, regional bank stock indices are still well below 2006 levels. Return to Text
23. Percent change in deposits due to merger activity is defined as (acquired deposits by top 10 firms over decade) divided by (total end of decade change in deposits). Asset share due to merger activity is defined similarly. Consolidation can also occur as a result of natural growth of deposits or assets and through a composition effect. Composition effects occur as the basket of top institutions changes. The composition of the top institutions changes as institutions merge together or as new institutions become banking organizations. For example, a new institution was included into the top 10 institutions after Wells Fargo acquired Wachovia (both in top 10). Another example is when Goldman Sachs became a bank holding company and ranked in the top 10. Return to Text
24. Thrifts represent about 16 percent of total institutions. Return to Text
25. Using December 31, 2010 data, large institutions with greater than $30 billion in assets would include approximately the top 40 institutions in the United States. Return to Text
26. Dollar values are all in 2009 dollars. Return to Text
27. The JPMorgan Chase acquisition of Washington Mutual was the only transaction that involved FDIC support. Return to Text
28. Acquisitions of banks with headquarters in the same state as the acquiring institution are exempt. Also, thrift acquisitions and failing banks are exempt. The Dodd-Frank Act removed the thrift exemption. Return to Text
29. For more information on the 10 percent liability cap, see the Study and Recommendations Regarding Concentration Limits on Large Financial Companies provided to the Financial Stability Oversight Council. Return to Text
30. The SNL Financial data do not note if a transaction involved any divestitures. The number of divestitures was obtained from data on Board Orders from 2000 through 2010. These data do not include any divestitures required by other bank regulators or by the Department of Justice. Return to Text
31. In the FRB's standard structural analysis, thrifts are only given 50 percent weight because they do not provide all banking services. Typically, thrifts do not engage in business lending and services. In merger applications to the Federal Reserve, thrifts that provide all banking services are given 100 percent weight. Return to Text
32. See Rice and Johnson (2008). Return to Text
33. Some states require that a bank already have a branch presence in the state before it is able to build more branches in the state. Return to Text
34. In fact, entry follows the same pattern. Bank entry into new markets very rarely crosses state lines. Return to Text
35. Transactions cannot be weighted by assets for any geographic level less than national because assets are reported only at the institution level. Return to Text

This version is optimized for use by screen readers. Descriptions for all mathematical expressions are provided in LaTex format. A printable pdf version is available. Return to Text