Home > Industry Analysis > Research & Analysis > FDIC Banking Review




FDIC Banking Review


Banking Industry Consolidation: Financial Attributes of Merging Banks
by John P. O'Keefe


Congress eliminated the remaining federal legal restrictions on interstate banking and branching in September 1994, with the passage of the Riegle-Neal Interstate Banking and Branching Efficiency Act of 1994 (Riegle-Neal Act).1 While many banking organizations had effectively circumvented legal barriers to interstate banking prior to the Riegle-Neal Act, they had to do so through the formation of multibank holding companies. Multibank holding companies may own and establish banks across states, provided they obtain a separate bank charter in each state.2 Many industry observers expect, therefore, that the Riegle-Neal Act will facilitate the consolidation of multibank holding companies into multistate branch bank networks. Moreover, the reduced legal barriers to market entry might also encourage mergers between unaffiliated banking organizations. The potential for large structural change in the industry and the resultant reallocation of real and financial resources across markets is of interest to market participants and observers.

In order to analyze what form this structural change might take, this study draws upon merger and consolidation activity among depository institutions between January 1984 and June 1996. Merger and consolidation transactions for the period are used to develop financial profiles of participating banks. Next, those profiles are used to develop statistical models that predict the likelihood that a bank will become involved in a merger, either as the acquirer or target, in the near term. The Riegle-Neal Act represents an unprecedented change in the legal environment in which mergers occur. Empirical merger-prediction models draw upon known merger histories and cannot incorporate the effects that changes to the legal or regulatory environment might have upon mergers. Nevertheless, an understanding of the financial characteristics of many acquirers and target banks should be useful in predicting mergers and consolidations in the near term.

The merger forecasts presented here indicate that the current rapid pace of bank mergers and consolidations is likely to continue into the near future. In addition, the forecasts indicate a substantial divergence between the number of potential acquirers and target banks in several geographic regions. Consequently, the continued growth of interstate banking organizations is also likely. The first section of this study describes recent trends in mergers and consolidations, and discusses the impact of the thrift and banking crises during the 1980s upon industry restructuring. Section two reviews the incentives that bank owners and managers have to act upon the reduced legal restrictions on interstate banking. The potential for merger activity is examined in the third section. Specifically, the financial profiles of acquirers and targets in mergers and consolidations developed previously are used to form statistical models that relate the incidence of mergers and consolidations to important financial characteristics of banks. Those models are then used to predict the likelihood of an institution being a target bank or an acquirer in a merger or consolidation over a two-year horizon. The study concludes with a discussion of the merger forecasts, as well as the geographic areas where interstate merger activity appears most likely to occur.

Recent Trends in Mergers and Consolidations

The recent legalization of full interstate branch banking could alter banking industry structure in two ways. First, many larger banking organizations used the multibank holding company organizational structure to form interstate banks before the passage of the Riegle-Neal Act. Some of these multibank holding companies might consolidate operations into multistate branch-bank networks if such networks offer advantages over existing organizational structures. Second, the ability to enter markets across state lines via branching might be a lower-cost alternative to the chartering of a new bank, as was required before the new legislation. Consequently, if barriers to market entry are reduced, there might be shifts in merger activity as banks implement their strategic merger plans.

The removal of legal impediments to interstate banking does not necessarily mean that more interstate banking organizations will develop. Mergers involve changes in ownership and, more importantly, can result in the reallocation of real and financial resources across markets. Such reallocations are motivated by the long-term expected risks and returns on invested capital. The present and expected future profitability of the industry will play an important role in such capital reallocations. In order to gain some perspective on what structural changes might occur, this section first examines recent merger and consolidation trends among depositories.

Table 1 shows the number of mergers and consolidations among commercial banks, savings banks and savings associations between January 1984 and June 1996. Mergers and consolidations were partitioned into five groups: (1) the formation of FDIC bridge banks; (2) RTC conservatorships; (3) FDIC-assisted failed-bank acquisitions; (4) FSLIC/RTC-assisted failed-thrift acquisitions; and (5) unassisted acquisitions. When a bank or thrift fails, the deposit insurer in its role as receiver has two general options to resolve the failure. The first is to liquidate the institution and compensate its creditors, particularly insured depositors; this type of resolution is known as a payoff. The second option is to sell some or all of the failed institution's operations intact to a financially sound bank or thrift; this type of resolution is known as a purchase-and-assumption transaction. The main criterion for selecting a resolution method is to select the method that is least costly to the deposit insurance fund.3 Only those transactions that involve keeping some portion of the failed institution's franchise intact are included in Table 1.



                                                 Table 1
                        Mergers and Consolidations:  January 1984 - June 1996
                   (All Commercial Banks, Savings Banks and Savings Associations)
                                    Number and Percent of Regional Total

                 FDIC           RTC                           FSLIC/
                 Bridge       Conserva-          FDIC-         RTC-
Region           Banks        torships         Assisted     Assisted        Unassisted      Total

Northeast       1   0.1%     59    7.7%     119     15.4%   78   10.1%      514    66.7%      771
Mid-Atlantic    0   0.0      27    7.3       11      3.0    31    8.4       300    81.3       369
Southeast       1   0.1     132    7.7       75      4.4   155    9.1     1,344    78.7     1,707
Central         0   0.0      71    4.0       38      2.1   118    6.6     1,561    87.3     1,788
Midwest         0   0.0      59    4.0      190     12.9    75    5.1     1,152    78.0     1,476
Southwest      20   0.9     159    7.3      685     31.4   311   14.2     1,009    46.2     2,184
West            0   0.0      87    7.1      188     15.4   150   12.3       794    65.1     1,219
Islands         0   0.0       1    9.1        2     18.2     3   27.3         5    45.5        11
Total          22   0.2%    595    6.2%   1,308     13.7%  921    9.7%    6,679    70.1%    9,525


In some cases the government will take over ownership of the failed institution prior to arranging a merger with another bank or thrift (government-assisted mergers). This is included in Table 1 as bridge bank and conservatorship transactions. When the government eventually sells the bridge bank or conservatorship to another depository or sells the failed institution immediately upon closure to another depository, these are shown as either FDIC-assisted or FSLIC/ RTC-assisted mergers in Table 1. All other mergers and consolidations that do not involve the deposit insurers or the RTC are designated as unassisted mergers. Table 1 shows that unassisted mergers comprised 70.1 percent of all mergers and consolidations during the period. The remainder involved government-assisted acquisitions of failed banks and failed thrifts by healthy depositories.4



                                                Table 2
                                       Mergers and Consolidations
                     (All Commercial Banks, Savings Banks and Savings Associations)
                                    Number and Percent of Yearly Total
             FDIC            RTC                           FSLIC/
             Bridge        Conserva-          FDIC-         RTC-
Year         Banks         torships         Assisted      Assisted        Unassisted       Total

1984        0    0.0%     0     0.0%       75    15.6%    14     2.9%     393    81.5%      482
1985        0    0.0      0     0.0        94    18.4     33     6.4      385    75.2       512
1986        0    0.0      0     0.0       121    22.1     52     9.5      375    68.4       548
1987        0    0.0      0     0.0       176    21.0     54     6.5      607    72.5       837
1988        0    0.0      0     0.0       201    19.2    189    18.0      658    62.8     1,048
1989        0    0.0     34     4.1       197    23.9    138    16.7      455    55.2       824
1990        1    0.1    259    23.8       158    14.5    221    20.3      448    41.2     1,087
1991        0    0.0    166    17.6       118    12.5    145    15.3      516    54.6       945
1992        0    0.0     54     7.3       109    14.8     64     8.7      508    69.1       735
1993       20    2.9     23     3.4        37     5.4      8     1.2      594    87.1       682
1994        1    0.1     58     8.0        13     1.8      1     0.1      655    90.0       728
1995        0    0.0      1     0.1         6     0.8      2     0.3      723    98.8       732
June 1996   0    0.0      0     0.0         3     0.8      0     0.0      362    99.2       365
All        22    0.2%   595     6.2%    1,308    13.7%   921     9.7%   6,679    70.1%    9,525



Regional differences in merger activity during the period have been driven in large part by economic conditions, as well as by changes in state banking laws. For example, the regional recessions in the Southwest and Northeast in the late 1980s and early 1990s, respectively, contributed to the high proportions of government-assisted mergers in those regions. Table 2 also shows that both the number and proportion of unassisted mergers rose steadily after 1990 as the economy improved.

Another interesting aspect of industry restructuring is the proportion of industry consolidation that has involved affiliated versus unaffiliated banks. Table 3 identifies consolidations of multibank holding companies versus mergers between unaffiliated banks. Because historical information on thrift holding company affiliations was not available, savings associations are excluded from Table 3. The data show a marked increase in consolidation activity in 1995, driven partly by the recent relaxation of federal restrictions on interstate banking and branching.



                                  TABLE 3
                        Mergers and Consolidations
                     Commercial Banks and Savings Banks
                        January 1984 - December 1995

                             Banks in                         Banks in
         Consolidations      Multibank     Non-afflicted      One-Bank
        With Holding Co.     Holding*      Transactions       Holding      Total
Year   Number  (Yearly %)    Company     Number  (Yearly %)   Companies   Mergers

1984    107    (36.2%)       3,741        189    (63.9%)      11,032        296
1985    188    (44.4)        4,127        235    (55.6)       10,670        423
1986    165    (36.9)        4,510        282    (63.1)       10,159        447
1987    333    (46.4)        4,422        284    (53.6)        9,766        717
1988    420    (52.3)        4,226        283    (47.7)        9,388        803
1989    286    (46.7)        4,067        327    (53.3)        9,131        613
1990    286    (51.4)        3,925        271    (48.7)        8,892        557
1991    314    (55.3)        3,677        254    (44.7)        8,708        568
1992    264    (49.8)        3,474        266    (50.2)        8,522        530
1993    261    (47.9)        3,375        284    (52.1)        8,192        545
1994    298    (51.7)        3,287        278    (48.3)        7,790        576
1995    402    (65.2)        3,073        215    (34.9)        7,459        617

Total 3,324    (49.7%)                  3,368    (50.3%)                  6,692

*The number of banks in multibank holding companies, as well as in one-bank
companies, are as of the calenday year-ends.

It is difficult to know what portions of the national merger trends seen in Tables 2 and 3 were driven by industry and regional economic conditions versus changes in federal and state banking laws. A state-by-state comparison of trends with associated changes in business and regulatory conditions would be too lengthy to present here. To gain some perspective on the impact that local economic conditions and changes in bank regulation can have upon merger activity, Tables 2b and 3b present bank and thrift merger trends for Texas depositories. Texas banking markets provide a useful illustration because of the severe changes in the state's economy and banking laws during the 1980s. Texas underwent a severe business downturn in the late 1980s. In addition, intrastate bank branching prohibitions in Texas were relaxed in 1987, permitting branching in contiguous counties; subsequently, statewide branching was permitted. Both events contributed to consolidations in Texas banking markets in the late 1980s and early 1990s.

Table 2b shows that merger activity in Texas increased sharply, from 41 to 233 transactions, between 1986 and 1987. Although a portion of this activity was due to commercial bank failures, most of the transactions involved unassisted mergers and consolidations. Table 3b shows that consolidations rose from 9 to 121 between 1986 and 1987. This increase was largely a result of the relaxation of state branching restrictions and consisted of unassisted transactions. Interestingly, mergers of non-affiliated banks between 1986 and 1987 also rose, from 23 to 96. While 45 of the 96 mergers in 1987 involved failed banks, the remaining 51 non-affiliate mergers represented a substantial increase over 1986 mergers.



                                               Table 2b
                                  Texas Mergers and Consolidations
                  (All Commercial Banks, Savings Banks and Savings Associations)
                                  Number and Percent of Yearly Total


                FDIC            RTC                            FSLIC/
                Bridge        Conserva-        FDIC-            RTC-
Year            Banks         torships        Assisted        Assisted      Unassisted       Total

1984          0     0.0%    0       0.0%      6    40.0%     1     6.7%     8     53.3%       15
1985          0     0.0     0       0.0      10    47.6      0     0.0     11     52.4        21
1986          0     0.0     0       0.0      22    53.7      6    14.6     13     31.7        41
1987          0     0.0     0       0.0      45    19.3      5     2.1    183     78.5       233
1988          0     0.0     0       0.0     110    29.9     89    24.2    169     45.9       368
1989          0     0.0     8       3.4     131    55.3     39    16.5     59     24.9       237
1990          0     0.0    45      19.9      99    43.8     45    19.9     37     16.4       226
1991          0     0.0    28      27.2      30    29.1     10     9.7     35     34.0       103
1992          0     0.0     5       7.7      29    44.6      0     0.0     31     47.7        65
1993         20    23.8     0       0.0      10    11.9      0     0.0     54     64.3        84
1994          0     0.0     0       0.0       0     0.0      0     0.0     31    100.0        31
1995          0     0.0     0       0.0       0     0.0      0     0.0     51    100.0        51
June 1996     0     0.0     0       0.0       2     4.7      0     0.0     41     95.3        43
All          20     1.3%    86      5.7%    494    32.5%   195    12.8%   723     47.6%    1,518



Incentives for Mergers and Consolidations Merger Theory

There are two participants in all mergers, the acquiring firm and the target firm. Because of the high degree of regulatory oversight of bank mergers, nearly all bank mergers result from the joint decisions of the controlling directors and shareholders of both of the merging banks. A discussion of the decision on whether to merge should, therefore, consider both the acquiring and target bank's perspectives. This section reviews the potential motives underlying the merger decision, drawing upon the bank merger studies of Rose (1987), (1988).5



                                        TABLE 3B
                             Texas Mergers and Consolidations
            Commercial Banks and Savings Banks, January 1994 - December 1995

                             Banks in                         Banks in
         Consolidations      Multibanks    Non-afflicted      One-Bank
        Within Holding Co.   Holding       Transactions       Holding      Total
Year    Number  (Yearly %)   Companies*   Number   (Yearly%)  Companies   Mergers

1984      1     (16.7%)        757          5      (83.3%)     1,096         6
1985      7     (43.8)         819          9      (56.3)      1,117        16
1986      9     (28.1)         860         23      (71.9)      1,111        32
1987    121     (55.8)         675         96      (44.2)      1,091       217
1988    132     (47.5)         467        146      (52.5)      1,025       278
1989     50     (26.3)         323        140      (73.7)        990       190
1990     22     (16.4)         255        112      (83.6)        928       134
1991     19     (30.2)         223         44      (69.8)        898        63
1992      9     (16.1)         191         47      (83.9)        898        56
1993     24     (30.8)         172         54      (69.2)        839        78
1994      5     (16.1)         186         26      (83.9)        802        31
1995     15     (31.9)         193         32      (68.1)        755        47

Total   414     (36.1%)                   734      (63.9%)                1,148

*The number of banks in multibank holding companies, as well as in one-bank 
campanies, are as of the calendar year-ends.


Merger motives can be classified into two broad categories: shareholder wealth maximization and managerial"well being." The notion that mergers are motivated by shareholder wealth maximization is a fundamental assumption of most economic theories on firm investment decisions. Under the wealth-maximization motive,mergers are treated like any other investment decision. Target firms in mergers are priced by bidders based upon the present discounted value of the expected returns on the acquisition, where the discount rate and return expectations consider the assumed firm's performance within the acquirer's portfolio of assets. In mergers, acquirers can share a portion of expected gains from the transaction with target firm owners to help encourage the merger. Such gains can result from post-merger improvements in the efficiency and profitability of the target bank's franchise or when the merged entity is expected to perform better than both of the individual firms. In either situation, the target bank's owners can be offered more than the current market value of their shares, because the "going concern value" of the target bank will be less than its value when combined with the acquirer's franchise.6 If the merged firm has greater long-term market value than the simple sum of the parts, merger synergies are said to have occurred. Specific sources of merger synergies are risk diversification in revenues and costs, economies of scale and scope, and market power.

Investment theory shows that as one increases the number of assets in an investment portfolio whose returns are positively correlated, the total variance in the portfolio's return decreases and approaches the average covariance between individual asset returns.7 If bank mergers increase portfolio diversification, the risk-reduction will benefit bank owners. The potential for increased geographic loan exposure diversification is probably the most likely source of benefits from interstate banking and branching. The regional concentrations of bank failures during the 1980s and 1990s were fueled by many banks' geographic lending concentrations, particularly those in commercial real estate.

Economies of scale refer to the ability to spread fixed operating costs over larger output levels, thereby reducing average total production costs. For example, bank mergers can reduce average costs when overlapping branch offices are closed, or fixed information processing costs and advertising costs are spread over increased revenues. In addition, personnel costs can be reduced when tasks overlap. Acquirers can benefit from applying "fixed" managerial and technical expertise to a larger business operation. Economies of scope are similar in nature, except that the cost savings result from applying fixed resources to a broader range of services, as opposed to simply increasing the level of the current mix of services. In addition, economies of scale can be achieved in financing. The costs of issuing debt and equity include a substantial fixed component. Consequently, larger banking organizations can spread fixed financing costs over larger equity issues, reducing per share issuance costs.

Mergers can also enhance market shares for acquirers for both balance-sheet and off-balance-sheet activities. This can confer some pricing advantages and improve profitability; however, there are limits to the extent to which mergers can be used to garner market power. Federal antitrust laws and regulatory policies restrict merger transactions in banking and other industries and are intended to prevent undue concentrations of market power. The primary federal antitrust laws that restrict merger activity are the Clayton Act of 1914, the Sherman Act of 1980 and the Federal Trade Commission Act of 1914. The Bank Merger Act of 1960, which was amended in 1966, clarifies federal bank regulators' role regarding bank merger policy. The U.S. Department of Justice (DOJ) and the Federal Trade Commission (FTC) are responsible for ensuring that bank merger transactions do not violate federal antitrust laws. The DOJ and the FTC have developed and published horizontal merger guidelines that present their policies and interpretation of appropriate merger practices.8

If a bank's owners or equity shareholders are not well represented on the firm's board of directors, the merger decision can be driven by managers' interests rather than those of shareholders. For example, managers seeking to protect their employment positions might actively block takeover attempts by many means, such as making preemptive acquisitions to ensure the firm is "too big to be a target." Since the assumed firm's management is often placed at risk of job loss in a merger, there is the potential for this motive to cause a divergence between shareholders' interests and managers' interests among targeted firms. The fact that managerial compensation usually increases with the revenues and assets of the firm also gives acquiring firms' managers an empire-building motive. This motive might not translate into increased wealth for their shareholders.

Finally, third-party influences on the merger decision can result in mergers with little or no benefit to either acquirers or target banks. Third parties involved in facilitating the transaction, such as investment bankers and securities dealers and underwriters, can profit from a merger transaction even when it does not produce the expected benefits to the acquiring firm's shareholders.9 As Rose (1988) points out, with such a large and diverse array of possible motives for mergers it is not unexpected that empirical studies differ in explaining why mergers occur. One can expect that some combination of the previous factors have influenced bank mergers over the past decade. The empirical analysis of merger motives developed in this study draws upon the motive of shareholder wealth maximization.

Practical Considerations: Identifying Likely Targets and Acquirers

Banks that are actively seeking to expand operations through mergers can have unique characteristics that distinguish them from their peers. Businesses that are in an expansion mode should be perceived to be in sound financial condition and could be expected to be outperforming their peers. An adequate equity capital base and healthy profit rates are necessary to attract the additional capital often needed to finance mergers. Conversely, managements that are not successfully operating an organization could not be expected to do any better with expanded responsibilities and should not be engaged in mergers.

While these traits could be found among banks actively seeking mergers, such banks might not always be able to translate their abilities into action, that is, acquire other banks. One reason for inaction might be the lack of worthwhile merger candidates within a bank's geographic market or targeted new markets. State and federal restrictions on branching and interstate banking might also have limited the scope of merger candidates available to some banks. Prior to the Riegle-Neal Act, regional banking compacts limited banks' ability to acquire banks in states that did not have reciprocal agreements. Finally, a variety of factors, such as expectations of regional and national economic recessions, or constraints on existing managements' ability to assume new responsibilities, can delay merger activity. Thus, while acquirers could have common characteristics, these traits might also be present in banks not active in merger markets.

Similar generalizations might be possible for target banks in mergers. Target banks might be underperforming their peers and could benefit from mergers. Inefficient scale and scope of operations can, at times, only be overcome with difficulty when banks have limited access to capital markets. While target banks might be underperforming peers, one would not expect acquirers to seek out targets with substantial problems or weak franchises. Hence, targets are likely to have deficiencies that can be remedied without substantial cost to acquirers. Deficiencies need not always be present in target banks, however. One commonly cited example is that of owner-managers of closely held banks. These owners can choose takeovers as a means to cash out on their investment at retirement, particularly when leaving the business to family members is not a consideration. Finally, as with acquirers, to be a target bank implies that acquisition mode banks must exist within the target's geographic market or out-of-market acquirers must find the potential target's market attractive. While many target banks might have common traits, one can expect these traits also to be present among banks that have not yet become merger targets.

If one can identify potential acquirers and targets within markets, more might be said about the likelihood of future merger activity. One first needs to identify common traits of acquirers and target banks. This section looks at the financial characteristics of both groups in three ways. First, acquirers' and target banks' income statements and balance sheets are reviewed to learn whether certain attributes appear just before mergers occur, or whether they are longstanding. Second, comparisons of financial characteristics with peer groups are made to determine whether acquirers and targets differ from banks of similar size and location. Third, acquirers are compared with their targets to investigate possible motives for mergers such as portfolio diversification and improvements in operating efficiency.

The attractiveness of target banks' franchises to potential acquirers is influenced by market demographics, as well as current and expected future economic conditions in the local and regional markets. Demographic data and economic activity measures might aid in explaining merger activity. This study relied, however, upon the financial statements of banks, as well as bank examiners' assessments (CAMEL ratings) of banks' financial condition in analyzing merger activity. Both market demographics and business cycles affect financial statements; therefore, these factors are not entirely ignored when relying upon financial statements. Moreover, because the geographic scope of most banks' markets is not well known, relating merger activity to demographic and economic activity measures involves uncertainties. For example, high commercial property vacancy rates in a particular market might be expected to reduce the attractiveness of area target banks with substantial commercial real-estate loan exposures; however, banks do not report geographic loan exposures to federal bank regulators. Thus, the relevance of local vacancy rates to all potential target banks is uncertain. Banks do report nonperforming asset levels that directly show the effect of market conditions upon bank asset quality.

Peer Group Comparisons

In order to learn how acquirers and target banks differ from each other and their peers, a sample of 890 mergers occurring among commercial banks and savings banks between January 1984 and December 1995 was obtained. To help ensure that banks' financial profiles would not be distorted by the accounting changes that can appear with mergers and consolidations, acquirers were required to not have made acquisitions nor consolidations over the eight quarters prior to a merger.10 All financial trends were tracked over the eight quarters prior to mergers. Second, only mergers between unaffiliated organizations were considered; consolidations of banks within the same holding company were excluded. Finally, all government-assisted mergers were excluded because of the unique nature of such mergers.11 The characteristics of the sample of acquirers and their target banks are given in Tables 4, 5 and 6. As shown in these tables, the sample of mergers is fairly diverse across geographic regions and over time.

All comparisons of financial performance were made based upon mean values of income statement and balance-sheet variables, expressed as a percent of bank assets, or, for loan portfolio analysis, as a percent of gross loans and leases. Financial ratios were computed over each of the eight quarters prior to mergers. In order to determine whether the financial performance of acquirers and targets differed from that of their peers, samples of peer banks were selected for both groups. The peer groups consisted of banks of similar size, regional geographic location, and timing of financial data as the target banks (acquirers). Financial trends were computed using an abstract time measure, the number of quarters from a merger. As a result, each quarter consists of data "pooled" from several points in time for the 890 banks. For comparability, peer banks that had contemporaneous financial data with the target banks (acquirers) were selected. The mean values of several important financial ratios and their differences between groups of banks are available from the author upon request. The results of that analysis are summarized next.



                          Table 4
  Sample of 890 Mergers Commercial Banks and Savings Banks,
            1984 - 1995(Dollars in Billions)
 

               Acquirer  Location*  Acquirers'  Targets'
Region          Number   (Percent)   Assets      Assets

 Northeast          56       (6)      $72.6      $17.8
 Mid-Atlantic       36       (4)       47.8        6.6
 Southeast         138      (16)      115.9        7.3
 Central1           40      (16)       39.1        9.0
 Midwest           253      (28)       46.6        9.3
 Southwest         140      (16)       63.2       16.4
 West              127      (14)      226.0       75.2
 Total             890      $611.3   $141.6
 
* The regional locations of target banks were the same as
  those for their acquirers in all but one instance.


Table 5 Sample of 890 Mergers Commercial Banks and Savings Banks 1984 - 1995 Acquirers Target Banks Asset Size Number (Percent) Number (Percent) $5 Billion or More 20 (2) 3 (0) $5 Billion to $1 Billion 64 (7) 15 (2) $500 Million to $1Billion 57 (6) 10 (1) $100 Million to $500Million 273 (31) 99 (11) $50 Million to $100Million 185 (21) 127 (14) $25 Million to $50Million 175 (20) 192 (22) Under $25 Million 116 (13) 444 (50) Total 890 890

Income and Expenses

The sample of target banks had significantly higher average loan-loss provisions and total noninterest expenses than did their acquirers.12 These higher expenses were not offset, on average, by interest and noninterest income among target banks. The result was a significantly lower average return on assets (ROA) among target banks than for their acquirers during the eight quarters before mergers took place. Net interest margins did not differ significantly between the sample of acquirers and their targets, however. The target banks were also typically much smaller than their acquirers. As a result, profit rates on assets (ROA) might differ due to bank size and capitalization. Targets could offset a lower ROA by increasing leverage; however, this will also increase the volatility in profits. This was not the case, however, because the target banks' equity capitalization was somewhat higher, on average, than that of their acquirers. In order to control for the effect of bank size on performance, peer bank comparisons also were made and are discussed next.



                 Table 6
         Sample of 890 Mergers
    Commercial Banks and Savings Banks
               1984 - 1995


                               Median
                              Ratio of
                             Target-to-
                              Acquirer
    Year   Number  (Percent)   Assets*

    1986     77       (9)       22.3%
    1987     93      (10)       30.2
    1988     99      (11)       34.1
    1989     68       (8)       28.1
    1990     61       (7)       21.9
    1991     77       (9)       27.8
    1992     93      (10)       31.7
    1993    103      (12)       26.3
    1994    124      (14)       26.8
    1995     95      (11)       22.4

    Total   890

    *Asset values were measured as of the quarter-end prior to
     the merger.



Target banks had significantly higher rates of loan-loss provisioning and expenses on premises than did their peers. The result was that target banks' ROAs were significantly lower than those of their peers over most of the pre-acquisition period. Target banks were both less efficient than their peers and riskier based upon loan-loss provisioning. There were no significant differences between the net interest margins of target banks and their peers over the eight quarters prior to acquisition. Acquirers had significantly higher net interest margins than did their peers over the pre-acquisition period. Acquirers, however, did not have significantly different loan-loss provisions nor noninterest expenses than their peers. The result was that acquirers had marginally higher ROAs than did their peers during the pre-acquisition period, but the divergence in profit rates was not statistically significant.

Portfolio Composition: Assets

Target banks were, on average, significantly more liquid than their acquirers, with a greater proportion of their assets in cash balances, federal funds sold and resale agreements than their acquirers. Correspondingly, acquirers had significantly higher proportions of total assets comprised of gross loans and leases than did their targets. Acquirers might find target banks' liquid asset levels attractive because these assets can be turned into loans at low cost if lending opportunities exist. One negative aspect of targets was their significantly higher proportions of other real estate owned, which includes repossessed real estate, than those of acquirers.

Target banks had significantly higher proportions of their assets in cash balances, federal funds sold and resale agreements than did their peers; however, targets also had significantly lower levels of securities than did peer banks. Target banks also held significantly higher proportions of their assets in gross loans and leases. Moreover, as one might expect, target banks had higher levels of assets in bank premises and other fixed assets. This might explain the higher overhead expenses found for target banks as compared with their peers. Finally, target banks had higher asset concentrations in other real estate owned, which includes repossessed assets. This latter finding supports the prior statement that targets might be riskier, on average, than their peers in terms of the credit quality of assets.

Surprisingly, acquirers compared to their peers in much the same way that target banks compared to their peers. Acquirers had higher asset concentrations in cash balances, federal funds sold and resale agreements than did their peers. In addition, acquirers held lower proportions of assets in securities and more in gross loans and leases than did peer banks. Finally, acquirers had higher proportions of assets in premises and other fixed assets than their peers. However, there were no significant differences between acquirers and their peers with respect to levels of other real estate owned.

Portfolio Composition: Liabilities

There were several differences between acquirer and target banks' liability composition, as measured as a percent of total assets. Acquirers had slightly lower levels of deposit funding than did their target banks and relied more on federal funds purchased and repurchase agreements, other borrowed money, banks' liabilities on acceptances outstanding, and subordinated debt. Interestingly, acquirers had significantly higher levels of volatile liabilities than did target banks and, consequently, lower levels of core deposits. Acquirers might, therefore, seek target banks with a more stable core deposit base. Equity capitalization was slightly lower among acquirers than target banks, but the difference was not statistically significant in most instances. Because target banks were, on average, about one-quarter the asset size of their acquirers, one might have expected significantly higher capitalization for the small target banks than for the larger acquirers, based upon historical capitalization rates across bank size groups. This asset size difference can explain some of the divergence in liability composition difference between acquirers and targets.

Target banks had higher levels of total deposit funding than did their peers but no significant differences existed for other major liability items, including large time deposit accounts and brokered deposits. Target banks, however, did have significantly lower equity capitalization rates than their peers. Although acquirers' deposit funding levels were similar to those of their peers, acquirers relied significantly more on volatile liabilities. In addition, acquirers had higher levels of federal funds purchased and repurchase agreements than did peer banks. The higher reliance upon volatile liabilities among acquirers could be an important motivator in mergers. Finally, acquirers had significantly lower equity capitalization rates than their peers. Because one must judge capital adequacy in relation to a bank's entire operations, including important factors such as loan-loss reserves, nonperforming asset levels, and profitability, these lower capitalization rates do not necessarily imply acquirers had weaker capital positions than their peers.

Portfolio Composition: Loans

There were some significant differences in the loan concentrations of acquirers and their target banks. Comparisons of loan portfolio composition, as a percent of gross loans and leases, indicated that acquirers had a slightly different mix of loans than did target banks and acquirers appear to have somewhat riskier loan concentrations than target banks. Acquirers had significantly higher average loan-to-asset ratios than did target banks and higher concentrations of both short- and long-term commercial real- estate loans (construction and land development loans and loans secured by nonfarm nonresidential real estate) than did target banks. Acquirers also had significantly higher concentrations of commercial and industrial loans and municipal loans than did target banks. Conversely, acquirers had lower proportions of loans in 1-to-4 family residential mortgages, consumer loans and loans to officers, directors and principal shareholders (insider loans) than did their targets.

Target banks had significantly higher ratios of gross loans and leases to assets than their peers. Target banks differed from their peers primarily in terms of real-estate lending, with higher loan concentrations in all areas of real-estate loans, including commercial real estate, than that of peer banks. Acquirers had significantly higher ratios of gross loans and leases than their peers. Among major loan categories, acquirers were somewhat more heavily concentrated in real-estate loans, particularly loans secured by nonfarm nonresidential properties (long-term commercial real-estate loans).

Predicting Acquirers and Target Banks

The previous section indicated that the sample of acquirers and their target banks differed systematically from each other and their peers prior to mergers. If these differences in financial characteristics are common and persistent over time, as the prior analysis indicates, it might be possible to use this information to identify banks that will become acquirers or targets in mergers. This section presents the results of logit estimation of models predicting the likelihood of being an acquirer or target bank in a merger. Logit estimation is used to relate mergers, either from the acquirers' or target banks' perspective, to a number of the factors, both endogenous and exogenous to a bank that the previous analysis indicated can affect the incidence of mergers.13 Because acquirers differ from their target banks, and both groups differ from their peers, separate logit estimations were obtained for acquirers and target banks.

Acquirers and their target banks appear to differ from their peers in terms of many important financial characteristics. Therefore, logit models were formed relating the incidence of mergers to the major attributes of banks' financial condition: capital adequacy, asset quality, management, earnings, and liquidity (henceforth, CAMEL attributes). Broad measures of bank condition were used in order to obtain models that would be robust across time and geographic regions. Therefore, details on loan portfolio composition and other factors likely to be correlated with time or location were excluded from the logit analysis.

To obtain general measures of condition, bank assets were partitioned into broad groups based upon earnings, liquidity, risk, and asset quality. Total assets were first partitioned into risk and nonrisk assets. Nonrisk assets were defined as the sum of cash balances due, securities, and federal funds sold plus resale agreements. Risk assets were, therefore, defined as total assets minus nonrisk assets. Nonrisk assets were further partitioned into two groups, noninterest-bearing nonrisk assets (that is, noninterest-bearing cash balances due) and interest-bearing nonrisk assets (that is, the sum of interest-bearing cash balances due, securities, and federal funds sold plus resale agreements). Risk assets were partitioned into performing and nonperforming risk assets. Nonperforming risk assets were defined as the sum of loans and leases past due 90 days or more, nonaccrual loans and leases, other real estate owned, and goodwill.

Two additional aspects of banks' asset portfolios were included in the analysis: lending levels and loan portfolio concentration. The proportion of banks' total assets comprised of loans and investment securities with maturities of five years or more was included as a measure of asset liquidity. In addition, a summary measure of loan-portfolio concentration was devised and included in the analysis. Specifically, total bank loans were divided into 15 well-defined categories of loans, which comprised nearly all of total loans. Next, the loan portfolio shares were obtained for these loan categories and the sum of squared shares were computed to form a concentration index analogous to the Herfindahl-Hirschman Index (HHI).14

The peer group analysis also indicated that measures of operating expense and profitability would be useful in predicting merger activity. Three components of noninterest expense were considered: expenses on salaries and employee benefits, expenses on fixed assets and premises, and all other noninterest expense. Bank profitability was measured by the return on earning assets (ROEA), which was defined as the ratio of operating income to earning assets. Operating income was measured by income before taxes and extraordinary items, gross of loan-loss provisions. Earning assets were defined as the sum of interest-earning cash balances, securities, federal funds and repurchase agreements sold, net loans and leases, and assets held in trade accounts, minus nonperforming assets.

Profitability and financial health are ultimately reflected in banks' capital adequacy; therefore, bank equity capital and loan-loss reserves were included in the models. Further, a bank's deposit franchise appeared to be an important factor in merger decisions. The main deposit measure considered was core deposits, defined as total deposits minus volatile liabilities. Volatile liabilities were defined as the sum of time deposits of $100,000 or more, all foreign-office deposits, federal funds purchased and securities sold under repurchase agreements, demand notes issued to the U.S. Treasury, and other borrowed money. Previous studies have also shown core deposit growth rates, as well as growth rates in gross loans and leases, might be important terms in predicting target banks.

Bank performance also varies systematically with bank asset size. It was hypothesized that the influence of asset size upon performance and condition decreases as total assets increase; therefore, the logarithm of bank assets was included as a size measure. In addition, de novo or recently established banks often have unusual financial characteristics when compared to established banks. These banks can also be precluded from being targets for a period after establishment by their chartering authority. Consequently, a de novo bank dummy variable, set equal to one for all banks in existence for three years or less (as of the model estimation date) and zero for all other banks, was included.

Because bank mergers require regulatory approval before transactions can proceed, bank regulators' assessments of banks' financial condition were particularly relevant to merger prediction. Bank regulators rate five aspects of banks' condition during periodic safety-and-soundness examinations: capital adequacy, asset quality, management, earnings, and liquidity. Banks receive ratings in each of the five CAMEL component areas that vary in integer levels from "1" to "5." Generally speaking, ratings of "3," "4," or "5" are given to banks considered to have moderate to serious deficiencies, respectively, that need to be addressed by bank management. These deficiencies can present risks that increase the chances of failure. Banks rated "1" are considered to be performing well-above-average, while a rating of "2" is given to banks with adequate performance, as dictated by regulatory safety-and-soundness standards. To gauge the extent of regulatory concern regarding banks' condition, the five CAMEL component ratings were included in the analysis using dummy variables set equal to one for banks rated "3," "4," or "5" for the component area and zero otherwise.

Equation 1 presents the most general form of the predictive equation, henceforth referred to as model 1.15 Model 1 was used to predict the likelihood of being either an acquirer or a target bank. Model 1 was estimated separately for target banks and acquirers, yielding two different sets of coefficient estimates. Finally, to control further for systematic differences in condition measures across bank asset-size groups, all balance-sheet variables were measured as percents of bank assets, and income and expense items were measured as percents of average assets. A second model, which excluded the bank examination terms, was also tested and is presented below (model 2). Comparisons of models 1 and 2 allow one to see the additional information that bank examination ratings add to merger prediction.

A stepwise logit estimation procedure was used in all estimations. This procedure systematically identifies those terms that have a significant relationship with the likelihood of being an acquirer or target bank and excludes all other terms. This allows one to include several measures of the same attribute in the logit model, allowing the estimation procedure to isolate the most important factors in terms of predicting merger activity.

The samples of banks used in estimating the models consisted of all commercial banks and savings banks reporting financial data at year-ends between 1984 and 1995. Further, two different samples of "merger events" were used for estimating the models. In the first sample all unassisted mergers between unaffiliated banks, as well as consolidations of member banks of a multibank holding company (affiliates) were defined as merger events. Assisted mergers were not counted as merger events, but were, however, left in the population of all other nonmerging banks. The assisted mergers were excluded from the definition of merger events because identification of assisted target banks would yield bank-failure prediction models rather than the type of target-bank prediction models of interest to this study. As shown in Tables 1 and 2, the proportions of assisted transactions were quite high during part of the estimation period; therefore, the "failure-prediction" results could best be avoided by the exclusion of assisted transactions. In addition, while consolidations might be of interest to some groups, such consolidations do not alter the actual number of economic agents or banking firms. Therefore, a second sample of merger events was defined as all unassisted mergers between unaffiliated banks. Banks involved in consolidations and assisted mergers were, however, left in the population of nonmerging banks. By construction, these two samples of merger events will allow for predictions on both aspects of industry consolidation from the population of banks.


EQUATION

The models were estimated by relating year-end financial and related data, to the incidence of mergers and consolidations over the subsequent two years. Estimates were obtained separately for target banks and acquirers. Nine sets of estimates were obtained, starting with 1985 and ending at 1993. The stepwise logit estimations are presented in the Appendix in Tables A-1 through A-6. In interpreting the results, a positive (negative) coefficient estimate, k, implies that an increase in that variable will increase (decrease) the likelihood of being an acquirer (or target) bank.16

Tables A-1 through A-6 also give some indication of the accuracy of the prediction models. To do this, estimated probabilities of being an acquirer or target bank were obtained for each period. Next, a critical probability value was chosen and banks whose estimated transaction probabilities were above that critical value were labelled likely acquirers or targets. A critical probability value of five percent was used in all "in-sample" merger forecasts. The five percent value was chosen based upon a subjective judgment that this value yielded fair predictive power. Finally, another measure of logit models' explanatory power is the pseudo R2 statistic.17 The pseudo R2 statistic is equal to 1 when the model perfectly predicts bank mergers, and zero when the explanatory variables provide no predictive information. For values between zero and 1, the pseudo R2 statistic measures "the percent of uncertainty in the data explained by the empirical results."18

Results of Logit Estimations

The results of the logit estimations were generally consistent with the profit-maximization, cost-reduction motives often given for mergers and consolidations. For brevity, the discussion will focus on positive results and will not attempt to explain why certain factors tested did not affect merger activity. Generally speaking, estimates of models 1 and 2 for target banks suggest that target banks had lower earnings, higher expenses on fixed assets and more liquid asset portfolios than non-target banks (all other banks). Table A-1 shows that the likelihood of being a target bank in an unassisted merger or consolidation increased with the proportion of liquid assets. In addition, the likelihood of being a target rose with expenses on premises and fixed assets, as well as with all other noninterest expenses. Interestingly, Table A-1 shows that the probability of being a target declined as expenses on employee salaries and benefits rose. This latter result, however, did not appear when consolidations were excluded from the definition of merger events (see Tables A-2 and A-3). The employee expense result in Table A-1 seems at odds with the cost-reduction motive for consolidations and might be related to high "trimming costs" that result from paying accrued benefits and severance pay when personnel is cut back. While such costs can occur with all mergers, in consolidations the willingness to incur such costs might play a larger role in decisions.

The coefficient estimates for the examination indexes for the five CAMEL component areas also suggest some additional characteristics of target banks. Comparisons of Tables A-1 and A-2 indicate that the likelihood of being a target in a consolidation decreases among banks with poorly rated management (Table A-1), but the management rating was not a factor in mergers of unaffiliated banks (Table A-2). The management rating might be expected to be related to acquirers' desire to keep some staff and management of assumed banks. The "human capital" value of the acquired banks' staff can come from knowledge of the local market or general experience and expertise. The result that target banks' management quality is not a determinant of mergers of unaffiliated banks could indicate acquirers' willingness to replace key staff.

Because the management of targets in consolidations is from the same holding company organization as the acquirers, the role of the management rating in Table A-1 might actually be driven by the overall organization involved in the consolidation. If this is the case, then the interpretation of the management rating among targets could have more to do with the acquirer side of the transaction. Results for the predictions of acquiring banks (Tables A-4 and A-5) show that the likelihood of being an acquirer decreases among banks with poor management ratings. This result is logical, given the importance of acquiring banks' management in obtaining regulatory approval of mergers, as well as shareholders' support. Tables A-1 and A-2 also show that the likelihood of being a target bank deceases the poorer the bank's earnings performance, perhaps indicating weaker franchises. In addition, comparisons of Tables A-1 and A-2 show that smaller banks are more likely to become targets in mergers of unaffiliated banks, but size is not a significant factor in consolidations. The results of model 2 in Table A-3 support the previous results and indicate the importance of some factors, such as earnings strength, to target-bank prediction can be captured by returns on earning assets or by examiners' ratings of earnings strength.

Tables A-4 through A-6 present similar tests of models relating banks' financial condition to the likelihood of being an acquirer in a merger or consolidation. Those results generally support the premise that acquirers are larger banks with well-rated management. Banks with more-liquid portfolios and lower loan concentration (loan HHI index) were also more likely to be acquirers. While these results agree with prior expectations regarding the characteristics of acquiring banks, the most striking result is how few factors were useful in predicting acquirers. Stated differently, the results clearly show how difficult it is to forecast merger activity. Of the 22 factors used to explain the characteristics of acquirers, only four appeared to be consistent in significance: asset size, liquidity, loan concentration, and the quality of management.

These results are consistent with a very common portrayal of bank mergers, whereby a large, well-run bank acquires a much smaller, less efficient bank. The target bank offers favorable attributes, such as liquidity. The acquirer is also able to correct any deficiencies of the target bank, thereby maintaining efficiencies for the combined organization. While this study does not investigate the post-merger performance of acquirers, the pre-merger profiles of targets and acquirers fit the scenario.

Forecasts of Acquirers and Targets

An important test of the usefulness of a forecasting model's predictive power is how well it predicts events that occur outside of the period used in estimating the model. For example, estimates obtained by logit estimations that related 1990 financial data to 1991 and 1992 merger events can be applied to 1992 financial data to forecast likely targets (acquirers) in 1993 and 1994. A series of such "out-of-sample" forecasts were made for the period 1987 to 1997 and are presented in this section. Forecasts on the number and asset size of the group of potential target banks and acquirers can be done in two ways. First, estimates of the probability of being a target can be obtained for an out-of-sample period for all banks. Second, all banks whose measured probabilities of being a target are above some critical probability value can be designated potential targets. Comparisons of actual targets with predicted targets will give an indication of model accuracy, much like the in-sample forecasts presented in Tables 1 through 3. The same approach can also be used to predict acquirers. One difficulty with this approach is the lack of any guide for selecting a critical probability. In practice, the lower the critical probability, the more likely one will correctly forecast actual targets or acquirers. However, this will also result in a large number of nonmerging banks being identified as targets or acquirers.

Because the models assign all banks some probability of being a target bank, one can simply take the sum of the predicted probabilities of being a target as the number of predicted targets during the two-year forecast period. While this alternative approach cannot be used to identify individual banks likely to be engaged in mergers, it can give an indication of the total level of merger activity predicted by the model. Whether these banks are large or small in asset size is also of interest. To measure the asset size of the potential target-bank group, one can multiply each bank's predicted probability of being a target by its total assets. Such statistical forecasts of the size of the target and acquiring bank populations were made for the period 1987 to 1997, based upon the logit estimations of unassisted mergers and consolidations (Tables A-1 and A-4). Those forecasts are shown in Figures 1 and 2, along with the size of the actual target and acquiring bank populations. More specifically, Figure 1 shows the asset size of all target banks as of the start of the two-year forecast period, as well as that of the forecasted targets. Figure 2 shows similar forecasts for acquirers.

TARGET BANKS ASSETS

Figures 1 and 2 show that the asset sizes of groups of potential targets and acquirers have been rising since 1990. The same is true of groups of actual merging banks. There are two interesting aspects of these forecasts. First, the large overestimation of forecasted target banks for the 1989 to 1990 period (Figure 1). This result might be due, in large part, to the higher proportion of assisted mergers in those two years. Assisted mergers accounted for about half of all transactions in 1989 and 1990 (Table 2). The forecasts shown in Figures 1 and 2 are for unassisted transactions. Many banks that fit the profile of target banks might not have been acquired during 1989 and 1990 because acquirers made government-assisted acquisitions instead. Acquirers might prefer an assisted acquisition over an unassisted acquisition because of the risk-reducing assurances the government typically grants to failed-bank acquirers.19 A second feature of the forecasts is the dramatic increase in the size of both the target and acquiring bank populations for the period 1996 to 1997. While the actual extent of mergers and consolidations will fluctuate with economic and regulatory events, Figures 1 and 2 indicate a substantial potential for continued industry consolidation in the near future.

ACQUIRERS'ASSETS Changes in States' Banking Markets

The Riegle-Neal Act will undoubtedly contribute to the continued increase in interstate banking organizations, particularly through the advent of interstate bank branching. Many industry observers are, therefore, curious about which geographic regions will undergo the greatest change. Such information would be particularly helpful to state and federal bank regulators who wish to know the future demands upon their organizations. The previous sections showed the difficulties in profiling acquirers and target banks. Forecasting the geographic location of merger activity adds additional unknowns to the forecasting question by requiring one to know the likely pairs of targets and acquirers. Predicting pairs of merging banks is extremely difficult because those decisions are driven by the individual characteristics of both sides of the transaction. Statistical forecasts, however, are driven by the average attributes of groups. Consequently, this section looks at the geographic distribution of future merger activity indirectly, through forecasts of potential targets and acquirers at the state level.

PROJECTION OF MERGER ACTIVITY

Figure 3 and Table 7 compare forecasts of the relative size of groups of potential targets and acquirers within each state for the period 1996 to 1997. Those forecasts were done using the same methodology outlined previously for national forecasts of unassisted bank mergers and consolidations. Target banks in mergers and consolidations are typically much smaller in asset size than acquirers. Table 6 showed that for a sample of 890 mergers between 1984 and 1995, the ratios of target-bank assets-to-acquirer assets averaged about 26 percent. To gauge the extent of potential targets and acquirers within each state, Figure 3 shows the ratio of forecasted target banks' assets to acquirers' assets within each state. There might be greater potential for interstate activity in regions where there is a high relative surplus of target banks, as indicated by a ratio of target-bank assets to acquirer assets much higher than the average rate of 26 percent.

Figure 1 shows the Midwest region has a high concentration of states where potential target banks' assets exceed 70 percent of potential acquirers' assets. Conversely, in regions where potential target banks are relatively few compared to potential acquirers, acquirers might be encouraged to look outside their geographic markets. Figure 3 shows this is the case in California and Texas, as well as in several large East Coast markets where potential target banks' assets averaged 30 percent or less than acquirers' assets. Following this line of reasoning, one might infer from Figure 3 forecasts of increased mergers between coastal banks and Midwestern banks in the near future. There are, however, some important caveats to such predictions. First, acquirers might not seek out small banks that fit the profiles of targets if market demographics are not attractive and offer limited growth opportunities. Second, the models this study develops can, at best, only give an indication of banks that have the attributes of acquirers and targets in mergers. Forecasting which banks will pair up in mergers requires more information than these statistical models can provide.


                           

                                                                   Table 7
                                               Projected Acquirers and Target Banks: 1996 - 1997
                                                      Unassisted Mergers & Consolidations
                                                             Dollars in Thousands
                    Number of  State Banking     Projected        Projected Targets'  Projected   Projected             Projected      Projected
       State          Banks       Assets       Target Assets        State Assets       Acquirer  Acquirer State           Number         Number
                                                                       Ratio            Assets   Assets Ratio           of Acquirers   of Targets
         AK              9       5,603,888        493,232               8.8          1,555,104       27.8                  1.4            0.7
         AL            186      56,322,295      6,149,388              10.9         27,535,026       48.9                 12.8           15.5
         AR            243      29,157,505      3,328,037              11.4          3,947,776       13.5                 16.7           24.9
         AZ             34      51,393,634      7,866,666              15.3         23,173,264       45.1                  4.6            3.3
         CA            383     368,222,678     42,646,860              11.6        246,549,406       67.0                 29.8           36.4
         CO            231      37,351,782      5,083,485              13.6         11,621,795       31.1                 13.1           19.0
         CT             86      66,818,979      9,746,381              14.6         23,148,034       34.6                  7.1            8.9
         DC             13       9,836,783      1,374,280              14.0          2,231,971       22.7                  1.3            1.6
         DE             42     111,177,750      3,539,376               3.2          6,672,123        6.0                  2.2            2.8
         FL            333     150,762,005     21,325,205              14.1         60,004,077       39.8                 30.8           33.1
         GA            383     132,670,009     25,315,300              19.1         69,383,306       52.3                 27.4           34.2
         HI             14      21,722,636      2,547,194              11.7         11,278,133       51.9                  2.1            1.1
         IA            491      42,018,384      4,567,017              10.9          8,265,976       19.7                 25.2           46.4
         ID             18      12,890,701      2,330,384              18.1          5,201,836       40.4                  2.6            2.2
         IL            917     244,479,856     26,336,931              10.8         80,644,262       33.0                 56.4           86.7
         IN            221      70,041,613      9,802,370              14.0         23,341,953       33.3                 21.1           22.7
         KS            433      31,428,922      3,578,873              11.4          4,961,034       15.8                 18.0           37.8
         KY            276      49,638,651      6,351,407              12.8         13,750,515       27.7                 19.8           28.3
         LA            190      46,221,952      4,849,155              10.5         13,249,696       28.7                 12.8           17.3
         MA            235     163,138,452     15,034,191               9.2         57,059,667       35.0                 13.3           17.7
         MD             93      69,033,564      8,662,148              12.5         34,804,499       50.4                  9.3            8.5
         ME             37      15,212,125      2,047,830              13.5          4,177,485       27.5                  3.5            3.0
         MI            180     118,900,391     15,425,932              13.0         65,769,103       55.3                 16.2           16.1
         MN            525      70,201,176      8,555,405              12.2         26,564,292       37.8                 27.1           45.4
         MO            459      80,481,252     14,834,989              18.4         31,330,691       38.9                 31.0           44.8
         MS            112      27,183,016      2,558,715               9.4          7,543,576       27.8                  9.2            9.6
         MT            104       8,193,543        963,920              11.8          1,854,750       22.6                  5.1            8.4
         NC            104     185,083,269     34,073,825              18.4        137,493,851       74.3                 10.8            8.8
         ND            127       8,036,833        785,129               9.8          1,096,467       13.6                  5.8           10.8
         NE            336      26,688,333      3,269,710              12.3          4,815,469       18.0                 15.2           30.7
         NH             42      17,763,748      3,996,050              22.5          4,154,242       23.4                  3.2            4.6
         NJ            110     121,721,651     15,039,922              12.4         52,692,790       43.3                 11.3           10.5
         NM             68      14,773,677      1,726,965              11.7          2,844,086       19.3                  5.3            6.4
         NV             25      26,334,174      2,106,497               8.0          3,954,798       15.0                  2.3            2.0
         NY            214     987,473,090     78,001,125               7.9        518,616,102       52.5                 21.6           16.1
         OH            287     162,398,816     22,490,673              13.8         70,605,006       43.5                 26.5           28.6
         OK            342      34,526,994      3,880,763              11.2          6,424,699       18.6                 17.2           28.0
         OR             45      31,570,733      4,249,998              13.5         15,703,178       49.7                  5.5            3.3
         PA            271     202,354,452     24,778,430              12.2        105,860,073       52.3                 25.0           26.0
         RI             11      22,708,039      2,644,297              11.6         10,814,279       47.6                  1.9            0.8
         SC             71      24,588,928      3,061,755              12.5          9,148,011       37.2                  6.2            5.7
         SD            116      28,614,803        633,473               2.2            513,370        1.8                  4.5            9.0
         TN            240      66,595,870      6,865,736              10.3         22,540,387       33.8                 18.8           22.5
         TX            948     210,076,213     22,519,635              10.7         82,562,199       39.3                 53.8           84.8
         UT             45      19,912,754      2,468,610              12.4          6,939,982       34.9                  3.9            4.3
         VA            158      77,818,910      7,401,301               9.5         32,086,378       41.2                 13.5           15.0
         VT             25       7,659,631      954,146                12.5          1,300,318       17.0                  2.5            2.8
         WA             99      74,214,908     12,243,496              16.5         44,599,653       60.1                 10.4            8.1
         WI            411      69,320,911      8,476,630              12.2         16,290,528       23.5                 30.5           41.7
         WV            118      21,255,773      2,177,964              10.2          3,874,201       18.2                  7.6           10.8
         WY             53       8,348,581        874,505              10.5          1,288,045       15.4                  3.1            5.2
         U.S. Total 10,514   4,539,944,633    520,035,306              11.5      2,021,837,462       44.5                725.9          962.9
          
         *Note that all asset values were based upon December 1995 Call reports.

		 

TABLE A-1 Model 1. - Predicting Targets in Mergers & Consolidations Stepwise Logit Estimation of the Relationships Between Banks' Year-End Financial Condition and the Incidence of Mergers and Consolidations Over the Succeeding Two Years. (FDIC-Assisted Mergers not classified as Mergers) Estimated Coefficients (Standard Error) Explanatory Variables 1985 1986 1987 1988 1989 1990 1991 1992 1993 Intercept -14.6565 -8.3238 -1.2914 0.8515 -2.5419 -3.4490 -3.5004 -4.4129 -1.4954 (2.0932) (1.8343) (0.1912) (0.5576) (0.1197) (0.2711) (0.2986) (0.4977) (0.1592) Interest-Bearing 0.1016 0.0734 - -0.0284 - - - - - Nonrisk Assets (0.0211) (0.0182) (0.0052) Noninterest-Bearing 0.1586 0.1169 0.0425 - 0.0651 0.0583 0.0671 0.0621 0.0288 Nonrisk Assets (0.0234) (0.0201) (0.0117) (0.0116) (0.0118) (0.0106) (0.0116) (0.0119) Performing 0.1292 0.1011 0.0260 - - - - - - Risk Assets (0.0215) (0.0181) (0.0051) Loan Portfolio - -0.0001 -0.0001 -0.0001 - - - - -0.0001 Concentration (HHI) (0.00004) (0.00003) (0.00004) (0.00003) Total Loans plus Securities -0.0187 -0.0343 -0.0277 -0.0211 - - - - - with >= 5 yrs maturity (0.0074) (0.0058) (0.0052) (0.0054) Expense on Salaries -0.5489 -0.7900 -1.1888 -0.7964 -0.6928 -0.6459 -0.5842 -0.6311 -0.7580 and Benefits (0.0745) (0.0851) (0.0949) (0.0841) (0.0809) (0.0884) (0.0848) (0.0835) (0.0797) Expense on Premises 0.7369 0.5554 0.6779 - - 0.6687 0.4559 0.5536 0.3629 and Fixed Assets (0.1106) (0.1205) (0.1342) (0.1438) (0.1538) (0.1582) (0.1686) All Other Noninterest 0.3942 0.4095 0.3247 0.2832 0.4356 0.3417 0.1873 0.1771 0.3090 Expense (0.0507) (0.0553) (0.0538) (0.0338) (0.0446) (0.0443) (0.0351) (0.0391) (0.0423) Return on Earning - -0.1838 -0.1024 -0.1465 - - - - -0.0743 Assets (0.0322) (0.0311) (0.0262) (0.0310) Equity Capital - - - - - - - - - Loan-Loss Allowance - 0.1612 0.1616 - - 0.2356 0.2334 0.2480 0.2382 (0.0552) (0.0493) (0.0501) (0.0480) (0.0519) (0.0507) Core Deposits - -0.0134 - - - 0.0065 0.0092 0.0142 - (0.0025) (0.0032) (0.0035) (0.0036) Core Deposit Growth - - - - - - - - 0.0010 (0.0004) Growth in Gross Loans - - - - - - - - - De Novo Bank Dummy -0.4653 -0.5728 - - - - - - - (0.1843) (0.1745) Supervisory Concern - - - -0.4513 - -0.2678 - - - for Capital Dummy (0.1317) (0.1197) Supervisory Concern - - - - - - - - - for Asset Quality Dummy Supervisory Concern - -0.3865 -0.4438 -0.3366 - -0.3290 -0.2460 -0.2149 -0.4336 for Management Quality Du (0.0923) (0.0951) (0.1130) (0.1041) (0.0923) (0.0888) (0.0980) Supervisory Concern - - 0.1901 0.4982 - 0.3831 0.4713 0.4263 0.2438 for Earnings Dummy (0.0838) (0.0919) (0.0934) (0.0885) (0.0859) (0.0945) Supervisory Concern - 0.2757 - - - - - - - for Liquidity Dummy (0.1061) Log of Total Assets 0.0947 0.0710 - - - - - 0.0697 - (0.0297) (0.0297) (0.0274) 2x Log of Likelihood 5875 7082 6646 5647 5705 5864 6085 6678 7213 Number of Observations 13,822 14,037 13,642 13,081 12,715 12,395 12,024 11,659 11,272 Pseudo R2 0.040 0.065 0.047 0.040 0.027 0.037 0.032 0.032 0.031 In-Sample of 10% Critical Probability Correct Predictions (%) Targets 18.7 45.5 36.1 18.2 14.2 24.9 28.7 41.1 63.1 Nontargets 92.7 83.1 86.5 93.3 95.6 91.1 87.0 77.2 58.3 Incorrect Predictions (%) False Failures or Type I 7.3 16.9 13.5 6.7 4.4 8.9 13.0 22.8 41.7 Missed Failures or Type I 81.3 54.5 63.9 81.8 85.8 75.1 71.3 58.9 36.9 Total Correct Predictions (%) 88.5 80.2 82.9 88.8 90.6 86.6 82.8 74.1 58.8 Number of Targets 801 1,072 965 775 783 830 878 1,016 1,154 Number of Nontargets 13,021 12,965 12,677 12,306 11,932 11,565 11,146 10,643 10,119
Table A-2 Model 1 - Predicting Targets in Non-affiliate Mergers Stepwise Logit Estimation of the Relationships Between Banks' Year-end Financial Condition and the Incidence of Mergers Over the Succeeding Two Years. (Consolidations and FDIC-Assisted Mergers Not Classified as Mergers) Estimated Coefficients (Standard Errors) Explanatory Variables 1985 1986 1987 1988 1989 1990 1991 1992 1993 Intercept -5.6309 -3.2043 -1.0598 -4.2086 -0.7494 -0.7118 -1.6850 -1.8376 -3.0233 (0.4245) (0.2347) (0.7064) (1.0652) (0.7197) (0.6216) (0.5133) (0.4645) (0.2228) Interest-Bearing -- -- -0.0264 -- -- -- -- -- -- Nonrisk Assets (0.0072) Noninterest-Bearing 0.0531 0.0493 -- 0.0389 0.0446 -- 0.0380 0.0384 -- Nonrisk Assets (0.0132) (0.0116) (0.0196) (0.0189) (0.0126) (0.0146) Performing -- 0.0240 -- -- -- -- -- -- -- Risk Assets (0.0086) Loan Portfolio -- -- -- -- -- -- -- -- 0.0001 Concentration (HHI) (0.00003) Total Loans Plus Securities -- -0.0318 -0.0344 -- -- -- -0.0094 -0.0111 -0.0138 with >= 5 yrs Maturity (0.0088) (0.0077) (0.0040) (0.0036) (0.0033) Expense on Salaries -- -- -- -- -- -- -- -- -- and Benefits Expense on Premises 0.5498 -- -- -- -- -- 0.2409 -- -- and Fixed Assets (0.0999) (0.1079) All Other Noninterest -- -- -- 0.0966 0.1093 0.1501 -- -- -- Expense (0.0390) (0.0513) (0.0387) Return on Earning -- -0.1727 -- -- -- -- -- -- -- Assets (0.0306) Equity Capital -- -- -- -- -- -- -- -- -- Loan-Loss Allowance -- -- -- -- -- -- 0.2510 0.1628 0.2067 (0.0653) (0.0704) (0.0631) Core Deposits 0.0158 -- -- 0.0217 -- -- -- -- -- (0.0050) (0.0076) Core Deposit Growth -- -- -- -- -- -- -- -- -- Growth in Gross Loans -- -- -- -- -- -- -- -- -- De Novo Bank Dummy -- -- -- -- -- -- -- -- -- Supervisory Concern for -- -- -- -- -- -- -- -- 0.3752 Capital Dummy (0.1516) Supervisory Concern for -- -- -- -- -- 0.4802 0.4651 0.3618 -- Asset Quality Dummy (0.1471) (0.1359) (0.1234) Supervisory Concern for 0.3625 -- -- -- -- -- -- -- -- Management Quality Dummy (0.1155) Supervisory Concern for -- -- 0.5912 0.5999 0.5115 0.6305 0.4400 0.7165 0.7940 Earnings Dummy (0.1206) (0.1382) (0.1430) (0.1485) (0.1329) (0.1156) (0.1173) Supervisory Concern for -- 0.3835 -- -- -- -- -- 0.3154 -- Liquidity Dummy (0.1440) (0.1469) Log of Total Assets -- -- -- -0.2014 -0.3684 -0.3480 -0.1975 -0.1351 -- (0.0668) (0.0661) (0.0576) (0.0463) (0.0407)
-2x Log of Likelihood 3206 3355 2810 2223 2091 2537 3168 3772 3768 Number of Observations 13,822 14,037 13,345 13,081 12,715 12,115 12,024 11,659 11,272 Pseudo R2 0.020 0.028 0.018 0.030 0.034 0.052 0.045 0.046 0.036
In-Sample at 4% Critical Probability Correct Predictions (%) Targets 15.4 22.2 5.4 7.9 13.6 38.2 42.7 56.6 49.8 Nontargets 93.0 91.1 96.6 97.4 96.9 86.4 78.6 69.1 72.5 Incorrect Predictions (%) False Targets or Type II Error 7.0 8.9 3.4 2.6 3.1 13.6 21.4 30.9 27.5 Missed Targets or Type I Error 84.6 77.8 94.6 92.1 86.4 61.8 57.3 43.4 50.2 Total Correct Predictions (%) 91.1 89.3 94.6 95.8 95.5 85.3 77.5 68.6 71.6 Number of Targets 351 374 297 227 213 280 372 472 470 Number of Nontargets 13,471 3,663 13,345 12,854 12,502 12,115 11,652 11,187 10,803
Table A-3 Model 2 - Predicting Targets in Non-affiliate Mergers Stepwise Logit Estimation of the Relationships Between Banks' Year-end Financial Condition and the Incidence of Mergers Over the Succeeding Two Years. (Consolidations and FDIC-Assisted Mergers Not Classified as Mergers) Estimated Coefficients (Standard Errors) Explanatory Variables 1985 1986 1987 1988 1989 1990 1991 1992 1993 Intercept -5.6860 -3.2075 -2.6312 -0.8507 -0.2565 -0.4328 -1.8917 -1.8731 -2.1432 (0.4264) (0.2337) (0.5478) (0.6630) (0.7052) (0.6148) (0.5002) (0.4600) (0.4522) Interest-Bearing -- -- -- -- -- -- -- -- -- Nonrisk Assets Noninterest-Bearing 0.0533 0.0494 0.0547 0.0400 0.0512 -- 0.0504 0.0411 -- Nonrisk Assets (0.0130) (0.0116) (0.0164) (0.0187) (0.0182) (0.0124) (0.0149) Performing -- 0.0228 -- -- -- -- -- -- -- Risk Assets (0.0086) Loan Portfolio -- -- -- -- -- -- -- -- 0.0001 Concentration (HHI) (0.00003) Total Loans Plus Securities -- -0.0293 -- -- -- -- -- -0.0074 -0.0109 with >= 5 yrs Maturity (0.0087) (0.0035) (0.0035) Expense on Salaries -- -- -- -- -- -- -- -- -- and Benefits Expense on Premises 0.5631 -- -- -- -- 0.2709 0.3398 0.4089 0.4173 and Fixed Assets (0.0974) (0.1138) (0.0993) (0.1296) (0.1375) All Other Noninterest -- -- -- -- 0.1371 0.1273 -- -- -- Expense (0.0446) (0.0461) Return on Earning -- -0.1827 -0.1227 -0.1313 -- -0.0761 -- -0.0925 -0.1251 Assets (0.0292) (0.0311) (0.0332) (0.0228) (0.0304) (0.0379) Equity Capital -- -- -- -- -- -- -- -- -- Loan-Loss Allowance -- -- -- -- -- 0.2507 0.3878 0.3628 0.3226 (0.0721) (0.0564) 0.0605) (0.0607) Core Deposits 0.0178 -- -- -- -- -- -- -- -- (0.0050) Core Deposit Growth -- -- -- -- -- -- -- -- -- Growth in Gross Loans -- -- -- -- -- -- -- -- -- De Novo Bank Dummy -- -- -- -1.2575 -- -- -- -- -- (0.5865) Log of Total Assets -- -- -0.1236 -0.3002 -0.4028 -0.3561 -0.2205 -0.1348 -0.0834 (0.0505) (0.0623) (0.0655) (0.0576) (0.0452) (0.0408) (0.0405) -2x Log of Likelihood 3240 3391 2829 2253 2110 2571 3218 3847 817 Number of Observations 13,928 14,110 13,695 13,116 12,740 12,412 12,035 11,660 11,272 Pseudo R2 0.017 0.024 0.012 0.024 0.029 0.040 0.030 0.027 0.023 In-Sample at 4% Critical Probability Correct Predictions (%) Targets 12.7 19.1 5.1 7.0 8.9 24.3 32.3 56.4 60.4 Nontargets 94.8 92.5 97.7 98.6 98.0 93.5 85.1 64.8 60.0 Incorrect Predictions (%) False Targets or Type II Error 5.2 7.5 2.3 1.4 2.0 6.5 14.9 35.2 40.0 Missed Targets or Type I Error 87.3 80.9 94.9 93.0 91.1 75.7 67.7 43.6 39.6 Total Correct Predictions (%) 92.7 90.5 95.7 97.0 96.5 91.9 83.5 64.5 60.0 Number of Targets 354 377 297 229 214 280 372 472 470 Number of Nontargets 13,574 13,733 13,398 12,887 12,526 12,132 11,663 11,188 10,803
Table A-4 Model 1 - Predicting Acquirers in Mergers and Consolidations Stepwise Logit Estimation of the Relationships Between Banks' Year-end Financial Condition and the Incidence of Mergers and Consolidations Over the Succeeding Two Years. (FDIC-Assisted Mergers Not Classified as Mergers) Estimated Coefficients (Standard Errors) Explanatory Variables 1985 1986 1987 1988 1989 1990 1991 1992 1993 Intercept -10.4927 -8.6842 -8.4512 -8.2614 -8.7272 -8.9289 -9.1472 -9.3531 -8.4877 (0.4298) (0.5009) (0.4077) (0.4733) (0.4273) (0.3883) (0.3567) (0.3425) (0.3916) Interest-Bearing -- -- -- -0.0080 -- -- -- -- -0.0119 Nonrisk Assets (0.0035) (0.0031) Noninterest-Bearing 0.0515 0.0546 0.0977 0.0715 0.0838 0.0776 0.0485 0.0338 -- Nonrisk Assets (0.0131) (0.0110) (0.0135) (0.0138) (0.0150) (0.0137) (0.0128) (0.0153) Performing 0.0156 -- 0.0081 -- 0.0119 -- -- -- -- Risk Assets (0.0040) (0.0035) (0.0038) Loan Portfolio -0.0002 -0.0002 -0.0003 -0.0002 -0.0002 -0.0003 0.0003 -0.0002 -0.0003 Concentration (HHI) (0.00006)(0.00006)(0.00005)(0.00006)(0.00005)(0.00006)(0.00005)(0.00004)(0.00005) Total Loans Plus Securities -- -- -- -- -- 0.0076 -- -- -- with >= 5 yrs Maturity (0.0038) Expense on Salaries -- -- -0.4633 -- -0.3692 -- -- -- -- and Benefits (0.1226) (0.1100) Expense on Premises -- 0.4342 0.5809 -- -- -- -- -- -- and Fixed Assets (0.1372) (0.1961) All Other Noninterest -- -- -- -- 0.1761 0.1117 0.1114 0.1048 -- Expense (0.0538) (0.0447) (0.0370) (0.0373) Return on Earning -- -- -- 0.0798 -- -- -- -- -- Assets (0.0365) Equity Capital -- -8.5427 -- -- -- -- -- -- -- (3.0338) Loan-Loss Allowance -- 0.1604 -- -- -- -- -- -- -- (0.0761) Core Deposits -- -- -- -- -- -- -- -- -- Core Deposit Growth -- -- -- -- -- -- -- -- 0.0016 (0.0005) Growth in Gross Loans -- -- -- -- -- -- -- -- -- De Novo Bank Dummy -- -- -- -- -- -- -- 0.6053 -- (0.2901) Supervisory Concern for -0.4026 -0.5235 -- -- -- -0.4078 -- -- -- Capital Dummy (0.1451) (0.1438) (0.1356) Supervisory Concern for -- -- -- -- -- -- -- -- -- Asset Quality Dummy Supervisory Concern for -- -- -0.4088 -0.6075 -0.5297 -- -0.3587 -0.5921 -0.6023 Management Quality Dummy (0.1218) (0.1414) (0.1335) (0.1153) (0.1155) (0.1426) Supervisory Concern for -- -- -- -- -- -- -- -- -- Earnings Dummy Supervisory Concern for -- 0.3247 -- -- -- -- -0.4286 -- -- Liquidity Dummy (0.1436) (0.1658) Log of Total Assets 0.5751 0.5028 0.4810 0.4853 0.4674 0.4877 0.5723 0.5900 0.6134 (0.0311) (0.0375) (0.0302) (0.0309) (0.0320)(0.0299) (0.0275) (0.0269) (0.0276) -2x Log of Likelihood 3623 4161 4070 3694 3677 3862 4001 4205 4220 Number of Observations 13,822 14,037 13,642 13,081 12,715 12,395 12,024 11,659 11,272 Pseudo R2 0.121 0.122 0.110 0.110 0.110 0.099 0.112 0.116 0.137 In-Sample at 10% Critical Probability Correct Predictions (%) Acquirers 26.8 32.0 26.7 25.8 26.9 26.6 32.5 36.2 41.2 Nonacquirers 96.1 94.6 95.0 95.6 95.3 95.3 94.0 92.8 90.7 Incorrect Predictions (%) False Acquirers or Type II Error 3.9 5.4 5.0 4.4 4.7 4.7 6.0 7.2 9.3 Missed Acquirers or Type I Error 73.2 68.0 73.3 74.2 73.1 73.4 67.5 63.8 58.8 Total Correct Predictions (%) 93.7 92.1 92.3 93.0 92.7 92.4 91.1 89.8 87.9 Number of Acquirers 473 565 544 485 487 515 557 605 636 Number of Nonacquirers 13,349 13,472 13,098 12,596 12,228 11,880 11,467 11,054 10,637
Table A-5 Model 1 - Predicting Acquirers in Non-affiliate Mergers Stepwise Logit Estimation of the Relationships Between Banks' Year-end Financial Condition and the Incidence of Mergers Over the Succeeding Two Years. Consolidations and FDIC-Assisted Mergers Not Classified as Mergers) Estimated Coefficients (Standard Errors) Explanatory Variables 1985 1986 1987 1988 1989 1990 1991 1992 1993 Intercept -10.0735 -10.2933 -9.2250 -11.3179 -8.8894 -7.2415 -8.6121 -9.0727 -9.8373 (0.5422) (0.4662) (0.4878) (1.1054) (0.5168) (0.5461) (0.4658) (0.4149) (0.4302) Interest-Bearing -- -- -- -- -- -- -- -- -- Nonrisk Assets Noninterest-Bearing -- 0.0383 0.0518 -- 0.0550 0.0533 0.0427 -- -- Nonrisk Assets (0.0158) (0.0201) (0.0223) (0.0174) (0.0154) Performing -- -- -- -- -- -- -- -- -- Risk Assets Loan Portfolio -0.0002 -- -- -0.0002 -- -0.0003 -0.0002 -0.0001 -0.0002 Concentration (HHI) (0.00009) (0.00009) (0.00008)(0.00007)(0.00005)(0.00005) Total Loans plus Securities -- -- -- -- -- -- -- -- -- with >= 5 yrs Maturity Expense on Salaries -- -- -- -- -- -- -- -- 0.1650 and Benefits (0.0551) Expense on Premises 0.4572 0.5837 -- -- -- -- -- -- -- and Fixed Assets (0.1785) (0.1576) All Other Noninterest -- -- -- -- -- -- -- -- -- Expense Return on Earning -- -- 0.0705 0.1588 -- -- -- -- -- Assets (0.0308) (0.0357) Equity Capital -- -- -- -- -- -- -- -- -- Loan-Loss Allowance -- -- -- -- -- -- -- -- -- Core Deposits -- -- -- 0.0173 -- -- -- -- -- (0.0067) Core Deposit Growth -- -- -- -- -- -- 0.0014 (0.0005) Growth in Gross Loans -- -- -- -- -- -- -- -- -- De Novo Bank Dummy -- -- -- -- -- -- 1.0700 0.8955 -- (0.3013) (0.3235) Supervisory Concern for -- -- -- -- -- -- -- -- -- Capital Dummy Supervisory Concern for -- -0.4223 -- -0.5878 -- -- -- -- -- Asset Quality Dummy (0.1578) (0.2320) Supervisory Concern for -0.6379 -- -0.5009 -- -- -- -0.6182 -0.6504 -0.4867 Management Quality Dummy (0.2069) (0.1991) (0.1540) (0.1549) (0.1855) Supervisory Concern for -- -- -- -- -- -- -- -- -- Earnings Dummy Supervisory Concern for -- -- -- -- -0.7987 -- -- -- -- Liquidity Dummy (0.3140) Log of Total Assets 0.5637 0.5141 0.4249 0.5439 0.3890 0.3194 0.4764 0.5200 0.5730 (0.0385) (0.0382) (0.0407) (0.0563) (0.0442) (0.0420) (0.0355) (0.0325) (0.0324) -2x Log of Likelihood 2077 2237 2060 1777 1756 2094 2485 2798 2666 Number of Observations 13,822 14,037 13,642 13,081 12,715 12,395 12,024 11,659 11,272 Pseudo R2 0.093 0.079 0.058 0.074 0.045 0.030 0.076 0.081 0.106 In-Sample at 4% Critical Probability Correct Predictions (%) Acquirers 32.1 26.7 17.5 19.8 12.6 14.0 35.9 44.6 48.2 Nonacquirers 95.2 94.4 95.9 96.1 97.2 96.5 89.6 84.5 85.8 Incorrect Predictions (%) False Acquirers or Type II Error 4.8 5.6 4.1 3.9 2.8 3.5 10.4 15.5 14.2 Missed Acquirers or Type I Error 67.9 73.3 82.5 80.2 87.4 86.0 64.1 55.4 51.8 Total Correct Predictions (%) 94.1 93.3 94.7 95.1 96.1 95.1 88.3 83.3 84.7 Number of Acquirers 224 240 212 182 174 215 284 336 330 Number of Nonacquirers 13,598 13,797 13,430 12,899 12,541 12,180 11,740 11,323 10,943
TABLE A-6 Model 2. - Predicting Acquirers in Nonaffiliate Mergers Stepwise logit estimation of the relationships between banks' year-end financial condition and the incidence of mergers over the succeeding two years. (Consolidations & FDIC-Assisted Mergers not classified as Mergers) Estimated Coefficients (Standard Errors) Explanatory Variables 1985 1986 1987 1988 1989 1990 1991 1992 1993 Intercept -10.6688 -10.3840 -9.5221 -11.7955 -8.7693 -7.2196 -8.7029 -9.2381 -9.9427 (0.4507) (0.4599) (0.4739) (1.0940) (0.5154) (0.5446) (0.4657) (0.4155) (0.4316) Interest-Bearing - - - - - - - - - Nonrisk Assets Noninterest-Bearing - 0.0352 0.0511 - 0.0588 0.0531 0.0407 - - Nonrisk Assets (0.0158) (0.0202) (0.0213) (0.0174) (0.0152) Performing - - - - - - - - - Risk Assets Loan Portfolio - - - -0.0002 - -0.0003 -0.0003 -0.0001 -0.0002 Concentration (HHI) (0.00009) (0.00008) (0.00007) (0.00005) (0.00006) Total Loans plus Securities - - - - - - - - - with >= 5 yrs maturity Expense on Salaries - - - - - - - - - and Benefits Expense on Premises - 0.5308 - - - - - - 0.1566 and Fixed Assets (0.1576) (0.0571) All Other Noninterest - - - - - - - - - Expense Return on Earning - - 0.0803 0.1705 - - - - - Assets (0.0294) (0.0342) Equity Capital - - - - - - - - - Loan-Loss Allowance - - - - - - -0.2316 0.2168 - (0.1053) (0.1016) Core Deposits - - - 0.0189 - - - - - (0.0067) Core Deposit Growth - - - - - - - - 0.0014 (0.0005) Growth in Gross Loans - - - - - - - - - De Novo Bank Dummy - - - - - - 1.0112 0.8820 - (0.3014) (0.3232) Log of Total Assets 0.5868 0.5173 0.4412 0.5659 0.3713 0.3177 0.4942 0.5446 0.5787 (0.0377) (0.0378) (0.0399) (0.0556) (0.0440) (0.0419) (0.0373) (0.0345) (0.0323) 2x Log of Likelihood 2097 2263 2068 1785 1785 2095 2498 2813 2674 Number of Observations 13,928 14,110 13,695 13,116 12,740 12,412 12,035 11,660 10,958 Pseudo R2 0.086 0.076 0.055 0.070 0.039 0.035 0.071 0.076 0.103 In-Sample at 4% Critical Probability Correct Predictions (%) Acquirers 29.9 28.1 16.5 19.2 11.4 14.0 35.2 42.0 45.8 Nonacquirers 95.6 94.7 96.1 96.3 97.3 96.6 90.4 85.3 86.1 Incorrect Predictions (%) False Acquirers or Type II Error 4.4 5.3 3.9 3.7 2.7 3.4 9.6 14.7 13.9 Missed Acquirers or Type I Error 70.1 71.9 83.5 80.8 88.6 86.0 64.8 58.0 54.2 Total Correct Predictions (%) 94.5 93.6 94.8 95.2 96.1 95.2 89.1 84.1 84.9 Number of Acquirers 224 242 212 182 176 215 284 336 330 Number of Nonacquirers 13,704 13,868 13,213 12,934 12,564 12,197 11,751 11,324 10,943
References

Billett, Matthew T., Jane F. Coburn and John P. O'Keefe. "Acquirer Gains in FDIC-Assisted Bank Mergers: The Influence of Bidder Competition and FDIC Resolution Policies." Federal Reserve Bank of Chicago, Proceedings of the 31st Annual Conference on Bank Structure and Competition, Chicago: n.p., 1995.

Hannan, Timothy H. and Stephen A. Rhoades. "Acquisition Targets and Motives: The Case of the Banking Industry." The Review of Economics and Statistics, Vol. LXIX, No. 1 (February 1987): 67-74.

Houston, Joel F. and Michael D. Ryngaert. "The Overall Gains From Large Bank Mergers." Journal of Banking and Finance, Vol. 18 (1994): 1155-1176.

Hunter, William C. and Larry D. Wall. "Bank Merger Motivations: A Review of the Evidence and Examination of Key Target Bank Characteristics." Federal Reserve Bank of Atlanta Economic Review, (September/October 1989): 2-19.

Judge, G. G., W. E. Griffiths, R. C. Hill, H. Lutkepohl, and T. Lee. The Theory and Practice of Econometrics. Second Edition. New York: John Wiley and Sons, 1985.

Maddala G. S. Limited-Dependent and Qualitative Variables in Econometrics. Cambridge, England: Cambridge University Press, 1983.

McLaughlin, Susan. "The Impact of Interstate Banking and Branching Reform: Evidence From the States." Federal Reserve Bank of New York Current Issues In Economics and Finance, Vol. 1, No. 2 (May 1995): 1-5.

Nolle, Daniel E. "Banking Industry Consolidation: Past Changes and Implications for the Future." Office of the Comptroller of the Currency Working Paper, 95-1, (April 1995).

O'Keefe, John P. "Bank-Failure Resolutions: Implications for Banking Industry Structure, Conduct and Performance." FDIC Banking Review, Vol. 5, No. 1 (Spring/Summer 1992): 17-35.

Rose, Peter S. "Characteristics of Merging Banks in the United States: Theory, Empirical Results, and Implications for Public Policy." Review of Business and Economic Research, Vol. XXIV, No. 1 (Fall 1988): 1-19.

Rose, Peter S. "The Impact of Mergers in Banking: Evidence From a Nationwide Sample of Federally Chartered Banks." Journal of Economics and Business, (November 1987) 39: 289-312.

Scherer, F. M. Industrial Market Structure and Economic Performance. Second Edition. Boston, Massachusetts: Houghton Mifflin Company, 1980.

Last Updated 8/2/1999 Questions, Suggestions & Requests