Evidence on Value Creation in the Financial Services Industries

8/10/2019 Evidence on Value Creation in the Financial Services Industries

1/22


2/22

214 K. C. Gleason et al./The Financial Review 38 (2003) 213234

Keywords: joint ventures, strategic alliances, long-horizon performance

JEL Classifications: G21/G29/G14

1. Introduction

Financial services firms, like any corporation, expand through cooperative strate-

gies like joint ventures and strategic alliances, as well as more traditional, integrative

ones (McConnell and Nantell, 1985).1 What we currently know about these expan-

sion strategies is rather one-sided. On the one hand, evidence on the wealth effects of

financial mergers and acquisitions mimics that found in nonfinancial activity. Specif-

ically, the combined gains to bidder and target firms appear to be positive in severalstudies (Baradwaj, Fraser, and Furtado, 1990; Cornett and Tehranian, 1992; Zhang,

1995; Gupta, LeCompte, and Misra, 1997), while other researchers find evidence of

a negative price reaction to bidders (Baradwaj, Fraser, and Furtado, 1990; Madura

and Wiant, 1994; Gupta, LeCompte, and Misra, 1997) or insignificant changes to

acquiring firm value (Sushka and Bendeck, 1988; Baradwaj, Dubofsky, and Fraser,

1991). In fact, Palia (1994) shows that two-thirds of the studies on bank mergers

report negative abnormal returns for acquiring banks.

Another line of research argues, on the other hand, that cooperative activities

are effective alternatives to more competitive strategies such as acquisitions. Chan,

Kensinger, Keown, and Martin (1997) find evidence that strategic alliances are aneffective alternative to more integrated forms of corporate combinations, while Kogut

(1988) and McConnell and Nantell (1985) make similar arguments for the alternative

use of joint ventures. While these studies provide important insights into the broad

use of these strategies, in each case, they do so by examining heterogeneous samples

comprised of many different industries. The purpose of this paper is to examine the

role of these cooperative activities in the growth strategies of firms in financial ser-

vices industries.2 The financial services industries provide interesting opportunities

to better understand these inter-corporate transactions for several reasons.

First, joint ventures and strategic alliances are an increasingly important mecha-

nism for growth in the financial services industries, as evidenced by the substantially

increasing number of transactions since 1990. In addition to this increased frequency

of joint venture and strategic alliance activities, several well-known firms make large-

scale use of these devices. For example, American International Group, Citicorp, Bank

1 Joint ventures generally result in the creation of a separate legal entity similar to corporations, part-nerships, or limited partnerships. The other type of organizational form is a category of less formal ar-rangements, strategic alliances. Alliances represent considerably less structured contractual agreements,involving marketing, licensing, and research and development agreements. We refer to joint ventures andstrategic alliances collectively as cooperative strategies or cooperative activities.

2 Our analysis of financial services industries includes commercial banks, investment services firms (suchas brokerage houses), and insurance companies.


3/22

K. C. Gleason et al./The Financial Review 38 (2003) 213234 215

of America, and Chase Manhattan have participated in 55, 39, 32, and 29 events, re-

spectively, since 1984. This high level of activity by major financial services firmssuggests that these cooperative arrangements are playing a more prominent role in

the growth strategies of these firms.

Second, a substantial amount of attention has recently focused on consolidation

activities in the financial services industries (Berger, Demsetz, and Strahan, 1999), as

well as in banking (Rhoades, 1998; Milbourn, Boot, and Thakor, 1999; Hughes, Lang,

Mester, and Moon, 1999), life insurance (Cummins, Tennyson, and Weiss, 1999), and

credit unions (Fried, Lovell, and Yaisawarng, 1999). In fact, the lead article of a special

issue of the Journal of Banking and Finance reviews the existing literature and cites

over 250 articles related to consolidation activities in the financial services industries

(Berger, Demsetz, and Strahan, 1999).Despite this interest in consolidation activities,very little academic research has examined how financial service firms use joint

ventures and strategic alliances to obtain strategic assets, enter new product markets,

and expand geographically. One notable exception is a study by Jacobsen and Tshoegl

(1999) in which they examine the strategic alliance strategies of Norwegian banks.3

Third, ongoing changes in the regulatory environment, culminating in the de facto

repeal of the Glass-Steagall Act in 1999, suggest that financial services firms will

continue to strategically position themselves in this evolving competitive marketplace.

One important possibility arising from these regulatory changes is cross-industry

activity between larger banking organizations and other financial service providers,

such as brokerage firms or investment banks (Berger, Demsetz, and Strahan, 1999;Milbourn, Boot, and Thakor, 1999). In late November 1999, for example, the Spanish

bank Banco Bilbao Vizcaya was seeking a strategic alliance with Italys Unicredito

Italiano. This strategic alliance is perceived as an important first step towards eventual

merger of the two banks. An official for Unicredito stated, The final goal is to merge

the Spanish and Italian banks into a European bank in two or three years (Edmonson

and Robinson, 1999).

We analyze 638 joint ventures and strategic alliances involving financial services

firms from 1985 through 1998. We look at the motives for entering into strategic

alliances and joint ventures, and the wealth effects arising from these cooperative

activities. We find that joint ventures and strategic alliances are value-generating

transactions when they are used to enter both new product and geographic markets. For

example, all three types of financial service firms generate gains for their shareholders

when they participate in international ventures or alliances. Further, we also find that

3 This disproportionate treatment of integrative and cooperative strategies extends beyond traditional re-search activities as well. The 2001 Report on Consolidation in the Financial Sectorissued by the Bankfor International Settlements discusses joint ventures and strategic alliances anecdotally. In a nine-pagedefinitional overview discussion, joint ventures and strategic alliances are discussed in half a page. The

rest focuses on mergers and acquisitions (M&A). In the Appendix, joint venture activity is summarized ina single page (Table A.18), while the previous 17 tables take up 63 pages to provide the same statisticalcoverage on M & A activity around the world.


4/22


these cooperative strategies enable commercial banks and insurance companies to

enter unrelated lines of business that may be (or may have been) forbidden throughtraditional acquisition means due to regulation.

The remainder of the paper is organized as follows. We discuss the nature of these

cooperative strategies and the motivations for using them in Section 2. We describe

the data and research method in Section 3, while the empirical results are presented

in Section 4. Section 5 concludes the paper.

2. The nature of joint ventures and strategic alliances

2.1. Theoretical motivations

Harrigan (1985) describes integrative and cooperative growth strategies as a

range of alternatives that vary in the degree of equity ownership and contractual con-

trol. Mergers and acquisitions make up the end of this spectrum that is characterized

by full equity ownership. The acquiring firms in these transactions take ownership

of the asset and skill base in exchange for assuming all of the risk inherent in those

assets. Growth alternatives where no new entity is created and there is no ownership

position make up the other end of Harrigans classification structure. These strategies

are pure contractual agreements or strategic alliances. The middle range contains

alternative transactions that are characterized by both partial ownership and some de-

gree of contractual control (e.g., joint ventures). These latter two categories providethe firm with viable growth strategies while leaving the asset base and management

structure of the parent firms virtually intact. This paper focuses on them.

A substantial body of literature has evolved over time that seeks to explain the

theoretical motivations for broadly-defined cooperative strategies in general (see,

e.g., Contractor, 1990), and for joint ventures in particular (see, e.g., Harrigan, 1985;

Kogut, 1988). While they use different terminology for their general classifications,

there is considerable overlap in these three discussions. One common motivation is

based on the theory of transaction costs (Williamson, 1975, 1985), which suggests

that firms transact through the mode that minimizes production and transaction costs.

This view considers these strategies as an efficient solution to what Kogut (1988)calls the hazards of economic transactions.

A second motivation relates to internal uses for these strategies, which are con-

cerned with the sources of the firms internal or firm-specific advantages. Cooperative

efforts lead to the creation of competitive strengths through resource aggregation and

risk-sharing efforts. In some frameworks, this is known as organizational knowledge

and learning (Lorange and Roos, 1992). Firms partnering in these cooperative ar-

rangements use them as a means for learning and extending their capabilities, as a

means for retaining some knowledge base, and as a mechanism for informational

exchange.

The third motivation focuses on the strategic issues facing the firm. Under the

strategic motivation, firms choose the cooperative growth mode that holds the best


5/22


potential for maximizing profits by improving their competitive position. In other

words, their choice of potential joint venture partners is made within the contextof their competitive environment (e.g., rivals and consumers). Harrigans framework

classifies this choice behavior into two distinct categories: competitive uses that serve

to strengthen the firms existing strategic position, and strategic uses that hold the

potential for enhancing or adding to the firms current strategic position. Competitive

motivations include preempting competitors by looking for first-mover advantages,

and influencing the evolution of the firms industry by overcoming trade barriers/

restrictions, accessing global opportunities, or creating more efficient competitors.

The strategic uses focus on identifying and exploiting synergies, technology/skill

transfer, and opportunities to expand the scope of the firm through diversification.

While the motivations outlined in these studies are presented in a general sense,they have specific relevance for the cooperative activities undertaken by the financial

services firms we examine. We consider these issues in this study by looking at two

important dimensions of these transactions. First, we look at the relationship between

those transactions that are domestic (partners are from the same country) versus those

that are cross-border or international (partners are not from the same country). Second,

we look at the relationship between those transactions that involve the same product or

industry versus those that are cross-product or diversifying in scope. In fact, Bodnar,

Tang, and Weintrop (2001) suggest that failure to consider both product market and

geographic characteristics may bias any attempt to calculate the value arising from

diversification.

2.2. Domestic and international cooperative activities

Social and political changes in areas such as Eastern Europe, the Middle East, and

Latin America, combined with technological changes such as the dynamic evolution in

telecommunications, have set the stage for the globalization of business. Corporations,

in general, are seeking to become less provincial and more global in their activities, at

the minimum to remain competitive, and, ideally,to enhance competitiveness. This

corporate vision is especially important since many firms cannot meet their growth

objectives solely through internal means. Thus, international transactions providealternative strategies for accomplishing these growth objectives.

Most firms lack the capability to launch large-scale multinational campaigns

through more traditional integrated organizational forms. They may lack the experi-

ence, the human and/or financial capital, and the infrastructure to implement growth

through full equity ownership alternatives. Cooperative joint ventures and strategic

alliances offer opportunities for cross-border expansion in spite of these deficiencies.

The popularity of these strategies is rising, partly because the advantages of control

and ownership do not outweigh the risks, or the financial and managerial commit-

ments facing the firm (Newman and diCiccio, 1998). These strategies facilitate entry

into countries that were once inaccessible (e.g., China, the republics of the former

Soviet Union, and Eastern Europe) or were governed by regulations mandating a local


6/22


partner (e.g., Chile). Firms can also benefit because these relationships can provide

access to existing distribution systems (in lieu of building or buying), new customerbases, new technology, information or skills, and existing or low-cost manufacturing

capabilities. Expanding firms can establish associations and relationships with ex-

isting firms, products, or franchises. Kvint (1998) argues that both the domestic and

international partners can use these strategies to decrease the risks they face.4

It is apparent that these strategies offer firms benefits and a means for reducing

some of the risks inherent in using acquisition for international expansion. We incor-

porate these important considerations into our analysis by looking at both domestic

and international joint ventures and strategic alliances. We classify a transaction as a

domestic transaction if all participating partners are from the United States. We clas-

sify the transaction as international if at least one of the partners is from a non-U.S.country.

Appendix A provides a listing of examples of joint ventures and strategic al-

liances analyzed in this study. In Appendix A.1, joint ventures one and two are

domestic transactions because all partners are from the United States. Joint ventures

three and four are illustrative of international joint ventures because at least one of

the partners is non-U.S. In the second section of Appendix A, we see that one and

two are domestic alliances and three and four are international alliances.

2.3. Integrating and diversifying activitiesThe cooperative activities of joint ventures and strategic alliances are used as

a strategic means of diversifying or expanding the scale and/or scope of the firms

operations. The resulting pattern of diversification depends on the scope of operations

of the financial services parent firms and the cooperative activity. If both the parent

and the cooperative activity are in the same industry, then the resulting venture or

alliance is an example of a horizontal cooperative activity. The new entity, in this case,

operates in the same product or service market although it may do so in a different

geographic locale. The cooperative activity may arise from the need to expand or add

to the firms product line, or take better advantage of excess capacity. It could also

be used to preempt competitors from entering a particular market venue.5

Diversifying or cross-product strategies, on the other hand, arise when the finan-

cial services parent and the cooperative activity are not in the same industry. Firms

pursue these strategies in search of new skills, technology, or resources. These types

4 Kvint (1998) cites the following as examples. Product risk arises because of specific risk issues relatedto the inputs of production (i.e., materials and labor), credit, operating and regulatory environments, andliability concerns. Financial risk includes currency and exchange rate risk, as well as concerns stemmingfrom interest rates, liquidity, settlement, derivatives and hedging, and the possibility of fraud.

5 Chan, Kensinger, Keown, and Martin (1997) show that horizontal strategic alliances generate higherabnormal returns than their diversifying counterparts.


7/22


of strategies may also be beneficial when entering new geographic or product markets

where barriers to entry make partnering an attractive alternative.We classify the transactions in our sample in terms of horizontal or related

product (i.e., the same industry) versus diversifying or cross-product (i.e., different

industry) agreements. In the first section in Appendix A, we see that joint ventures

one and three are horizontal agreements since the financial services parent and the

joint venture are in the same industry. Joint ventures two and four are diversifying

activities as the parents and the venture are not in the same industry. For the strategic

alliances listed in Appendix A.2, we note that one and four are horizontal in nature,

while two and three are examples of diversifying alliances.

3. Research design

3.1. Sample selection

Data are obtained from the Securities Data Corporation (SDC) International Joint

Ventures database by screening on several criteria. In the data screening process, U.S.

commercial banks, investment services, and insurance companies involved in joint

venture or strategic alliance activity from 1985 to 1998 are selected. This screen

involves selecting SIC codes 6021 and 6022 for banks, 6200s for investment services,

and 6300s and 6400s for insurance companies. At least one of the financial services

parents must have financial data available for an announcement to be included inthe analysis. Financial data is often available for more than one financial services

partner for an announcement. Some firms, such as Chase Manhattan, engaged in

multiple cooperative activities during the sample period. Thus, 187 firms participated

728 times in 628 cooperative activities. Share price data for 513 announcements is

available from CRSP.

Our data fall into three distinct categories. First, we use accounting-based data

from Standard and Poors Research Insights for the sample period from 1985 to

1998. Our data from Research Insights encompasses the 728 participants in the 638

cooperative activities. The descriptive statistics in Tables 1 through 4 are based on

these 728 participants.Second, we use CRSP stock market returns data to identify short-term stock

market performance associated with the announcement of the cooperative strate-

gies for our sample firms. We had the 1997 CRSP files available at the time of the

analysis for this paper. Further, during the 1985 to 1997 period, stock returns were

not available for some sample firms. The analysis in this category includes 513 firms

and is reported in Tables 5 and 6.

Finally, for our long-term performance analysis, we use both CRSP and Research

Insights data. We omit firms with multiple announcements within the respective

six-, 12-, or 18-month holding period. If the relevant holding period contains more

than one announcement per partnering firm, the holding period returns may not

capture the impact of the announcement event under consideration. Further, due to


8/22


size considerations, we are unable to match certain firms in the sample (i.e., there was

no firm in the same SIC and size classification that had not engaged in a cooperativeactivity or merger during the same time horizon as the sample firm). Thus, the results

reported in Table 7 include 187 announcements.

3.2. Research methods

3.2.1. Event study

To test the markets reaction to the announcements of these joint ventures and

strategic alliances, we use event study methodology with the significance tests based

on the standardizedcross-sectional method (Boehmer, Musumeci,andPoulsen, 1991).

Returns are modeled using ordinary least squares. A 100-day estimation period was

chosen from t110 to t11 where t = 0 is the event date.6 Cumulative abnormal

returns (CARs) are reported for the (1,+1) and (1,0) event windows.

3.2.2. Long-horizon holding period returns

We further examine wealth effect issues related to these cooperative transac-

tions using long-horizon holding period returns. Previous studies have shown that

estimating long-horizon returns by accumulating short-term abnormal returns leads

to biased test statistics (see, e.g., Conrad and Kaul, 1993; Barber and Lyon, 1997).Barber and Lyon provide evidence that using benchmarks comprised of control firms

yield well-specified test statistics because this benchmarking approach mitigates the

biases discussed (e.g., new listing, rebalancing, and skewness biases).

We use a matching firm sample procedure for estimating the long-horizon aver-

age holding period abnormal returns (AHAR) for our sample firms. We first identify

matching firms for our sample of firms on the basis of four-digit SIC codes. From

this set we next match firms on the basis of size as measured by total assets. We then

calculate raw returns for both sample and matching firms from the month subsequent

to the announcement date for six-, 12-, and 18-month holding periods. The reported

AHARs are the average difference between the holding period returns for the sampleand matching firms. The numbers of positive and negative long-horizon holding pe-

riod abnormal returns are also reported along with the test statistic for the two-tailed

generalized sign test.

6 We varied the parameter estimation period from 100 to 150 days. Results similar to the ones reportedwere obtained. In some cases our total sample includes more than one announcing firm for the same date.Accordingly, our significance tests adjust for cross-sectional dependence in the regression residuals. Wealso adjust for serial dependence as well as for changes in event induced variance (see the manual for

Eventus, version 7.0). Although not reported here, the use of Scholes-Williams betas produced similarresults to the ones reported here. CARS are based on an equally weighted market portfolio. Similar resultswere obtained with the value weighted market portfolio.


9/22


Table 1

Distribution of joint ventures and strategic alliance announcementsThis table provides thedistribution of announcements by year of announcement over thesampleperiod from

1985 to 1998, by the mode of cooperative activity, i.e., whether it was a joint venture or a strategic alliance,

and by industry classification. Some of the announcements involve more than one banking, investment

services, or insurance firm. Thus, the total number of f irms in the sample is larger than the total number

of announcements.

Panel A: By year

Year of Number of

announcement announcements % of total

1998 78 12.23

1997 93 14.58

1996 67 10.50

1995 123 19.28

1994 74 11.60

1993 62 9.72

1992 35 5.49

1991 42 6.58

1990 35 5.49

1989 6 0.94

1988 8 1.25

1987 3 0.47

1986 11 1.72

1985 1 0.16Total announcements 638 100.00

Panel B: By mode classification

General classification Number of Number

of cooperative activity announcements of firms

Joint venture 385 420

Strategic alliance 253 308

Total 638 728

Panel C: By industry classification

Number of Number

Industry SIC codes announcements of firms

Banks 6000s 343 407

Investment services 6200s, 6400s 92 96

Insurance 6300s 203 225

Total announcing firms 638 728

4. Empirical results

4.1. Summary of sample characteristics

Table 1 provides information regarding the distribution of announcements. Panel

A of Table 1 shows the frequency of announcements from 1985 to 1998. The


10/22


Table 2

Summary of financial characteristics of the partner f irmsThis table provides summary financial characteristics of the 728 firms participating in the sample of

638 events for which both Research Insights data are available. The mean values for selected financial

characteristics for the partner firms are provided for the sample as a whole, as well as by industry type. All

of the financial data are reported for the year prior to the joint venture or strategic alliance announcement.

Some firms are represented more than once in the sample. Thus, the figures in this table represent value-

weightedaverages. Return on assets [equity] is measured as net income divided by the book value of total

assets [equity].

Variable Total sample Banks Investment services Insurance

Net income ($ 100,000) 888 977 593 902

Total assets ($ 100,000) 84,061 98,998 76,138 63,147

Market value of equity ($ 100,000) 13,325 13,903 6,630 15,745Return on assets (%) 1.49 1.00 2.23 2.04

Return on equity (%) 13.57 14.08 14.38 12.26

frequency of announcements by year indicates that only 29 of the 638 (4.54%) an-

nouncements occurred during the 1985 to 1989 period. After 1989, the pace of an-

nouncements per year accelerated. The number of annual announcements increased

from 35 in 1990 to 78 in 1998. While this may in part reflect the broader scope of

coverage by SDC in the 1990s, it may also be indicative of increased commercial

bank activities due to deregulation.

Panel B, Table 1, shows a breakdown of the sample by type of cooperative

activity. Three hundred eighty-five of the 638 announcements are for joint ventures

involving 420 partner firms. The remaining 253 transactions are strategic alliances

with 308 firms participating. It is interesting to note that while for joint ventures the

average number of firms per announcement is 1.09, it is 1.42 for strategic alliances.

This difference may be explained in terms of the nature of the collaborative activity

and the sharing of control in the venture. While joint ventures typically involve an

equity stake and ownership implications, alliances are less constricting in terms of

governance and more focused on general resource sharing.The breakdown of the sample by number of announcements and SIC classifica-

tion of firms is provided in Table 1, Panel C. There are 407 banks associated with 343

announcements, 96 investment services firms associated with 92 announcements, and

225 insurance firms associated with 203 announcements.

A summary of the financial characteristics of the sample firms is provided in

Table 2. Mean values of selected variables are shown for the total sample and for

each industry subgroup and measure the respective variables in the year of the firms

cooperative activity announcement. On a value-weighted basis,7 the sample is made

7 Some firms are represented more than once in the sample (e.g., a firm that appears in our total sampleten times would have all of its ten preannouncement total asset figures included in the computation).


11/22


up of large firms averaging $84.1 billion in total assets. The largest firms in terms of

total assets are in the banking industry, averaging almost $99 billion, while insurancefirms are the smallest on average ($63.1 billion in total assets). Market capitalization

averages $13.3 billion for the total sample, with banks, investment services firms,

and insurance firms averaging $13.9, $6.6, and $15.7 billion, respectively. Investment

services and insurance companies generated average return on assets of over 2% in

the year prior to the announcement of the joint venture or strategic alliance, while

banks lagged behind with an average ROA of slightly less than 1%. Return on equity

for all three types of financial services firms ranged from 12% to 15%.

Interesting details are revealed when the participation characteristics of the co-

operative activity firms are examined. Panel A of Table 3 shows the variation in the

number of partners for the sample of announcements. The number of partners in anyof these transactions ranges from the required minimum of two to 16. Just over 74%,

or 539 of these announcements, involve two partner firms. How often these partnering

firms engage in these cooperative activities is shown in Panel B, Table 3. These data

show the number of times (and how many) firms participated in these events. For

example, 83 firms were involved in a single cooperative activity during the sample

period, while another 29 firms participated in two transactions. The average firm en-

gaged in 3.41 cooperative activities during the sample period. Seventeen firms were

involved in more than ten joint ventures or strategic alliances. These include Equitable

Life and First Chicago (11 events each), as well as Bank of America (32), Citigroup

(39), and American International Group (55). Thus, we see that these cooperative ac-tivities are an important element of the growth strategies for these firms, particularly

the larger firms in their respective industries.

Earlier, we addressed two important strategic competitive considerations. First,

we noted that firms pursue cooperative strategies such as joint ventures and strategic

alliances in search of new skills, technology, or resources. The cooperative activi-

ties may also be beneficial when entering new geographic or product markets where

barriers to entry make partnering an attractive alternative. We incorporate these im-

portant considerations into our analysis by considering whether the transaction is:

(i) domestic or international, and (ii) horizontal or diversifying. Table 4 shows the

distribution of the sample in terms of geographic strategy (Panel A), product orbusiness relatedness (Panel B), and the various combinations of the two together

(Panel C).

Four hundred and forty-two, or almost 61%, of the firms are involved in domestic

transactions,while 278, or 38%, are associated with a cooperative activitythat involves

at least one foreign entity. Domestic activities for banks account for almost 65%, while

representing 71% for investment services firms. Only for the insurance firms is the

ratio of domestic to international activities nearly even.

Panel B, Table 4 shows that horizontal activities occur about 26% of the time.

The vast majority of transactions, about 74%, are of the diversifying type. These

results indicate that firms in the financial services industries are seeking to acquire

complementary competencies from nonrelated partners. Sample financial institutions


12/22


Table 3

Participation characteristics of the cooperative activity firmsThe table shows the frequency of joint venture and strategic alliance activity by all participants. The number

of financial services firms in the sample is 728. Panel A shows the number of announcements that involve

the specified number of partners. Panel B shows the number of firms involved in a particular number of

events over the sample period. The names (and frequency of participation) of the financial services firms

in the sample with more than 10 activities each during the sample period are also provided.

Panel A: Announcements by number of partner firms

Number of Number of

partners announcements

2 539

3 604 19

5 5

6 4

7 3

8 3

9 2

15 1

16 2

Total announcements 638

Panel B: Frequency of cooperative activity use by partner firms

Number of Number of events firms Firm name

1 83

2 29

3 to 5 43

6 to 8 15

11 to 12 4 Equitable Life Insurance, First Chicago (11 each)

JP Morgan, PNC (12 each)

14 to 16 6 Bank of Boston, First Union, Wells Fargo (14 each)

Cigna, State Street (15 each)

Aetna (16)

18 to 20 2 Mellon (18)Bank One (20)

23 1 Bankers Trust

29 1 Chase Manhattan

32 1 Bank of America

39 1 Citigroup

55 1 American International Group

engaged in joint ventures across a wide spectrum of complementary industries such

as data processing, information technology, and other financial services.

Panel C in Table 4 shows the interactive geographic and product strategies.

While most of the strategic combinations are domestic- and product-diversifying,


13/22


Table 4

Distribution of announcing firms by strategic/competitive motivesThis table provides the distribution of announcing firms by strategic and/or competitive motives. All firms

involved in a domestic cooperative activity are U.S. firms. International cooperative activities involved

at least one non-U.S. firm. A horizontal cooperative activity is one in which both the financial services

parent firm and the cooperative activity are in the same industry. A diversifying cooperative activity is one

in which the financial services parent firm and the cooperative activity are not in the same industry.

Category Banks Investment services Insurance Total

Panel A: Geographic relatedness

Domestic 264 68 110 442

International 137 28 113 278

Undisclosed 6 0 2 8Total 407 96 225 728

Panel B: Business relatedness

Horizontal 89 24 74 187

Diversifying 215 72 150 535

Undisclosed 5 0 1 6

Total 407 96 225 728

Panel C: Strategic combinations

Domestic-horizontal 49 15 34 98

Domestic-diversifying 210 53 77 340

International-horizontal 40 9 40 88International-diversifying 98 19 73 190

Undisclosed 10 0 2 12

Total 407 96 225 728

190 (26%) are both geographic- and business-diversifying. The results indicate that

financial services firms readily source strategic assets in foreign as well as domestic

markets.

4.2. Event study results

The event study results are presented in Table 5. We present the cumulative ab-

normal returns (CARs) for the (1,+1) and (1,0) windows for the sample of 728

financial services announcements related to 638 joint venture or strategic alliance

events. Returns data for event study methodology were available for 513 announce-

ments. Results are presented for the total sample (Panel A), for each industry clas-

sification in the sample (Panel B), and for each of the strategic competitive motives

(Panel C).

For the total sample, the (1,+1) and (1,0) event window abnormal returns

are 0.66% and 0.51%, respectively, both significant at the 1% level. For a firm with a

market value of about $13 billion, the 0.66% abnormal return represents an increase


14/22


Table 5

Cumulative abnormal returns by industry classification and strategic/competitive motivesThis table presents the announcement period cumulative abnormal returns (CARs) for the (1,+1), and

(1, 0) eventperiod windows forthe sampleof 728 financialservicesf irms participating in 638jointventure

or strategic alliance events. Abnormal returns are calculated using the market model estimated from 110

to 11 days prior to the event announcement. The CRSP equally-weighted market index is used. CARs

represent the cumulative market model-adjusted change over the relevant event window. The tstatistics

(given in parentheses) are based on the standardized cross-sectional method. The number of positive and

negative CARs for the (1,+1) window (+/) are reported in the last column, with the Z statistic for the

nonparametric generalized sign test reported in parenthesesunder+/. Domestic, international, horizontal,

and diversifying cooperative activities are defined in Table 4.

CAR event windows

No. (1, +1) (1, 0) +/

Panel A: Total sample

Total sample 513 0.66 (4.88) 0.51 (4.67) 297/216 (4.66)

Panel B: By industry classification

Commercial banks 278 0.69 (3.72) 0.45 (2.97) 157/121 (3.96)

Investment services 69 0.52 (1.41) 0.60 (1.96) 34/35 (0.35)

Insurance 166 0.65 (2.81) 0.58 (3.08) 106/60 (4.07)

Panel C: By strategic motive8

Domestic 311 0.59 (3.40) 0.45 (3.20) 176/135 (3.23)

International 197 0.72 (3.56) 0.60 (3.63) 116/81 (3.07)

Horizontal 134 0.84 (3.31) 0.61 (2.96) 80/54 (2.68)

Diversifying 376 0.57 (3.98) 0.47 (3.99) 214/162 (3.68)

Indicates significant difference from zero at the 0.01 level. Indicates significant difference from zero at the 0.05 level.

in market value of approximately $86 million. These results provide evidence that

financial services firms partnering in cooperative activities significantly increase

shareholder wealth.

The abnormal returns associated with these cooperative activities can be com-

pared with the abnormal returns reported in previous studies on mergers in the fi-

nancial services industry. A comprehensive survey article by Palia (1994) shows that

two-thirds of studies on bank mergers report negative abnormal returns for acquiring

banks. Our results, when compared with results from previous bank merger stud-

ies, suggest that joint ventures and strategic alliances may be more desirable than

8 There are five announcements that could not be classified according to geographic markets and three

according to product markets. Thus, the combined total sample size for domestic and international sub-samples is 508, and for diversifying and horizontal subsamples the total sample size is 510, rather than513 as given for the total sample.


15/22


Table 6

Cumulative abnormal returns by combined strategic/competitive motivesThis table presents the announcement period cumulative abnormal returns (CARs) for (1, +1) and

(1, 0) event windows for the sample of 728 financial services firms participating in 638 joint venture

or strategic alliance events. Abnormal returns are calculated using the market model estimated from

110 to 11 days prior to the event announcement. CARs represent the cumulative market model-adjusted

change over the relevant event window. The CRSP equally-weighted market index is used. The tstatistics

(given in parentheses) are based on the standardized cross-sectional method. The number of positive and

negative CARs for the (1,+1) window (+/) are reported in the last column, with the Z statistic for the

nonparametric generalized sign test reported in parenthesis under+/. Domestic, international, horizontal,

and diversifying cooperative activities are defined in Table 4.

CAR event windows

No. (1, 1) (1, 0) +/

Panel A: Total sample9

Domestic-horizontal 67 0.49 (1.23) 0.25 (0.78) 35/32 (0.08)

Domestic-diversifying 244 0.62 (3.40) 0.51 (3.49) 141/103 (3.28)

International-horizontal 66 1.18 (3.04) 1.00 (3.13) 44/22 (3.10)

International-diversifying 131 0.49 (2.22) 0.41 (2.25) 72/59 (1.48)

Panel B: Commercial banks10



International-horizontal 28 1.79 (2.83)

1.22 (2.13)

22/6 (3.01)


(continued)

mergers for financial services firms seeking to expand their scale and/or scope of

operations.

Commercial banks and insurance companies in our sample experience signifi-

cant positive wealth gains of a similar magnitude. The evidence on investment firms

is mixed. For these firms, while both the (1,+1) and (1,0) CARs are positive, only

the latter is statistically significant.Panel C of Table 5 reports the results for subsamples based on strategic/

competitive motives.11 The CARs for all four subsamples are significantly

9 There are five announcements that could not be classified by geographic or product markets. Thus, thecombined number of geographic and product events analyzed is 508 rather than 513 as stated for the totalsample in Table 5.

10 Two announcements for which CARs could be generated could not be classified by geographic orproduct markets and are not included in the analysis, accounting for the 276 bank announcements out of278 total bank announcements represented in Table 5.

11 The CARs for the subsamples for joint ventures and strategic alliances are not significantly differentfrom each other. Thus, we do not stratify our sample based on the type of cooperative activity.


16/22


Table 6 (continued)

Cumulative abnormal returns by combined strategic/competitive motives

CAR event windows

No. (1, 1) (1, 0) +/

Panel C: Investment services





Panel D: Insurance12





Indicates significant difference from zero at the 0.01 level. Indicates significant difference from zero at the 0.05 level. Indicates significant difference from zero at the 0.10 level.

positive. At 0.84% for the (1,+1) window, they are the highest for horizontal co-

operative activities, followed by international cooperative strategies.

The parametric results for the (1,+1) window are confirmed by the nonpara-

metric generalized sign tests for the event window. Thus, we see that these cooper-

ative strategies are value-generating activities for banks, insurance companies, and

to a lesser extent, other investment services firms. These significant results hold re-

gardless of the strategic/competitive motivations involved (i.e., these results hold for

transactions that are domestic, international, horizontal, or diversifying).

Next, we examine whether certain combinations of transaction motives provide

additional insights into how these firms in the financial services industries utilize

joint ventures and strategic alliances. To do this, we look at the following motive

classifications: domestic-horizontal, domestic-diversifying, international-horizontal,and international-diversifying. The results for this part of our analysis are presented

in Table 6 for the total sample (Panel A), for banks (Panel B), for investment services

(Panel C), and for insurance companies (Panel D).

For the total sample, those transactions that expand the domestic scope of the

firmsoperations (domestic-diversifying) add value to the sample partner firms. Ex-

panding the international scale of the firms (international-horizontal) as well as the

international scope of firms (international-diversifying) results in positive abnormal

12 Table 5 shows that the total number of announcements for insurance firms is 166. However, three couldnot be classified according to their product or geographic status. Therefore, the sum of the insurancesubsample observations in this table is 163.


17/22


returns of 1.18% and 0.49% for the (1, +1) event window, respectively. Both of

these CARs are significant at the 1% level. Domestic-horizontal transactions generatepositive returns over both event windows, but neither is statistically significant.

Panels B through D show the results for these strategic combinations by type of

financial services industry. What is striking about these results is that they indicate

how these gains are generated across the sample industries. Specifically, the bank-

ing industry drives the results for domestic-diversifying and international-horizontal

combinations, and the insurance industry results drive the international-diversifying

combinations. Furthermore, the results for banks show that diversifying domestically,

and expanding horizontally internationally, produce significant, positive abnormal

returns. For investment services firms, international-horizontal transactions gener-

ate abnormal returns averaging 1.87%, which are significant at the 10% level forthe (1, 0) window. However, it should be noted that this subsample includes only

seven announcements. Diversifying transactions abroad involving insurance compa-

nies generate significant positive returns at the 5% level for both of the near-term

windows (0.74% and 0.66%, respectively).

Finally, we examine CARs by time subsamples (19851986, 19871989, 1990

1993, and 19941998), representing changes in the regulatory environment related

to allowable bank activities.13 However, we find no significant differences in CARs

across time periods and do not report the results in Table 6.

The results in Table 6 provide insights into which strategic motives generate

value for which type of financial services firm. The most significant activity is forthe banking subsample. Activities involving domestic diversification and expansion

of the scale of operations internationally add value to bank shareholder wealth when

conducted through a cooperative mode. For the other two subgroups, international

cooperation exhibits positive wealth effects, although for different reasons. While

investment services firms benefit from international horizontal efforts, significant

gains for insurance companies come from ventures and alliances that diversify or

expand their scope of operations internationally.

4.3. Long-horizon holding period returnsIn Table 7, we examine the long-horizon average holding period abnormal re-

turns (AHAR) generated by these transactions for the 187 announcements without

13 Specifically, we are referring to pieces of legislation aimed at expanding products, services, and marketsin which U.S. commercial banks are activethe 1987 Competitive Equality Banking Act (CEBA), whichprevented creation of new nonbank banks by defining a bank as any firm that accepts deposits, changesin the early 1990s to regulatory tolerance for relaxation of Glass-Steagall restrictions (cases of holdingcompanies being allowed to issue and underwrite a variety of securities and taking limited equity stakesin nonbank firms), and the 1994 Riegle-Neal Interstate Banking and Branching Efficiency Act, which

changed the competitive environment by allowing U.S. and non-domestic banks to expand interstate byconsolidating out-of-state bank subsidiaries into a branch network and/or acquiring banks. Comparablechanges in regulatory regime began in Europe primarily in January of 1998.


18/22


Table 7

Longhorizon holding period abnormal returnsThis table reports the long-horizon average holding-period abnormal returns (AHAR) for our sample

firms. We compute long-horizon holding period raw returns for the sample firms (HPRFi) and for the

matched firms (HPRCi). AHARi = HPRFi HPRCi. The two-tailed significance tests are based on the

standardized cross-sectional method. HPRFi, HPRCi and AHARi are calculated starting with the month

after the announcement date for six-, 12-, and 18-month holding periods. Firms in the sample with multiple

transactions that confound the long-horizon analysis are omitted from the analysis. Further, due to size

considerations, we are unable to match certain firms in the sample. Thus, the sample sizes in these tests

are smaller than those in the near-term event study.

1-year HPR event windows

Max. no.14 (1, 6) (1, 12) (1, 18)

Panel A: AHAR by industry classification

Total sample 187 4.86 (1.70) 7.18 (2.62) 7.94 (2.00)

Commercial banks 86 5.46 (2.44) 7.24 (2.03) 7.45 (2.12)

Investment services 37 4.33 (1.13) 9.44 (1.86) 10.26 (1.98)

Insurance 64 4.35 (1.49) 6.29 (0.97) 7.40 (1.74)

Panel B: AHAR by strategic/competitive motive15

Domestic 134 5.13 (1.89) 6.90 (2.02) 9.40 (2.05)

International 52 4.15 (1.40) 7.90 (2.28) 4.05 (0.66)

Horizontal 54 7.48 (1.79) 7.76 (1.87) 14.79 (2.28)

Diversifying 131 3.78 (1.69)

6.94 (2.12)

5.10 (1.12) Indicates significant difference from zero at the 0.01 level. Indicates significant difference from zero at the 0.05 level. Indicates significant difference from zero at the 0.10 level.

confounding events for which a matched firm could be identified. The reported re-

turns are the difference between the returns of our sample firms less those of the

matched sample. We look at (+1,+6), (+1,+12), and (+1,+18) month long-horizon

event windows. Panel A reports the returns for the total sample and by industry type,

while Panel B reports those for the four strategic/competitive motives.For the sample as a whole, we observe significant positive AHARs across all

three event windows, ranging from 4.86% for six months to 7.94% for the 18-month

period (significant at least at the 5% level). Abnormal returns to bank shareholders

14 The sample size shows the maximum number of announcements for which data are available for anyof the three windows. For example, for the (1,6) window, 187 returns could be calculated, but the numberdeclines to 135 as the time horizon extends to 18 months.

15 Panel A shows that the total number of events for which we could estimate AHARs for the (1,6) month

event window is 187. However, one event could not be classified with regard to its geographic status. Thus,domestic and international activities combined resultin analysis of 186 events. Similarly, the diversificationstatus for two events could not be determined, resulting in analysis of 185 events.


19/22


are also significant across all three windows at the 5% level. Investment services

firm returns are significant for 12- and 18-month horizons only (9.44% and 10.26%,respectively). Only the 12-month window shows significant returns (at the 10% level)

for the insurance subsample.

Positive abnormal returns are observed for all four strategic competitive motives

for the (+1, +6) month window, and are statistically significant for all motives ex-

cept international. For the (+1, +12) month window, the returns for all four motive

categories are significant and range from 6.90% for all domestic events to 7.90%

for international transactions. Over the 18-month period, only the domestic and the

horizontal activities add significant value. Domestic cooperative activities generate

returns of 9.29%, while horizontal activities result in AHARs of 14.79%. Both motive

types are significant at the 5% level.

5. Conclusion

Several studies have looked at merger and acquisition strategies in the finan-

cial services industries. More recently, consolidation activity in these industries has

captured the attention of researchers as well. A very important and emerging com-

ponent of the growth strategy of these activities is cooperative activity made up of

joint ventures and strategic alliances, and these activities have received little attention,

relative to the volume of work that has been done on bank mergers and acquisitions.

In this study, we examine how firms in the banking, investment services, and insur-ance industries use these cooperative strategies and whether these strategies generate

wealth-increasing benefits for their shareholders.

This analysis contributes to our understanding of both the motives for these co-

operative strategies, as well as the impact of these activities on the short-term and

long-term performance of participating firms. Our analysis indicates that there is a

significant increase in the wealth of the shareholders of firms that announce their par-

ticipation in cooperative strategies. Banks and insurance companies are better able to

extract gains through cooperative activities compared to investment services firms.

However, this result may be more of an artifact related to the relatively smaller sam-

ple size of investment services firms. International activities and horizontal activitiesdominate domestic and diversifying activities, respectively. As expected, when coop-

erative strategies are both international and horizontal, the gains are larger than those

associated with the other three combination motives. Financial services firms that can

expand their scope of operations are able to capitalize on their core competencies,

and this strategic advantage is recognized by investors. This factor is particularly

significant for banks, which appear to possess the ability to successfully export their

expertise in their core area through joint ventures and strategic alliances.

The long-horizon holding period returns analysis indicates that the stock prices

of participating firms appreciate more than those of matching firms. Further, even

though the results for investment services firms are mixed in the short run, in the long

run these firms also gain from entering into cooperative activities.


20/22


The results of this paper hold significant implications for managers of finan-

cial services firms. In general, the use of joint ventures and strategic alliances byfinancial services firms holds the potential for increasing profits. Second, expanding

horizontally into international markets provides financial services firms with the op-

portunity to capitalize on their domestic expertise and reap handsome profits. In an

environment that is becoming increasingly competitive, cooperative activities provide

financial services firms with strategic and competitive advantages through which they

can improve financial performance.

Appendix A: Sample descriptions of cooperative activityannouncements

The following descriptive material provides examples of the type of cooperative

activities analyzed in this study. The information provided is taken directly from the

Securities Data Corporation International Joint Venturesdatabase.

A.1. Joint ventures

1. Banc One Corp entered into a joint venture with BankAmerica Corp and

Chemical Banking Corp to operate a nationwide system for microchip-

embedded smart cards. The venture was called SmartCash. The smart

card was to enable value from a cardholders deposit or credit account to

be loaded on it. The card could then be used in place of cash and coins for

purchases targeted at the $20 and under range. Other financial institutions

were expected to join the joint venture. Financial terms were not disclosed

(August 16, 1995) [D R].

2. Marsh & McLennan Company and J.P. Morgan & Co. formed ajoint venture

company to exploit the rising global interest in the catastrophe reinsurance

market. The market had suffered losses after hurricanes Andrew and Iniki

in 1992 (November 6, 1992) [D C].

3. Citibank NA and the Government of Saudi Arabia signed an agreement to

form Saudi American Bank, a joint ventureto provide banking services inSaudi Arabia (January 1, 1990) [I R].

4. State Street Boston Corp and an undisclosed European bank formed ajoint

venture to provide foreign currency trading services to European companies.

The joint venture was to be named Galleon Capital Corp (April 19, 1995)

[I C].

A.2. Strategic alliances

1. Ameritech and Citibank formed a strategic alliance to offer interactive bank-

ing services. The companies planned to combine telephone and banking

services delivered to the home of the consumer using screen telephones.

Consumers were to do their banking through these interactive services. The

telephones were equipped with digital display screens and a typewriter-like


21/22


keyboard. Ameritech and Citibank planned to market the telephones di-

rectly to consumers. Financial terms were not disclosed (November 9, 1993)[D R].

2. Smoky Mountain Technologies, a unit of UniComp Inc., and First Tennessee

National Corp entered into astrategic allianceto provide a combined credit

card, debit card, and frequent shopper program targeted at the quick-service

food market (May 30, 1997) [D C].

3. Lippo Ltd., a Philippines unit of Lippo Capital, entered into a strategic al-

liancewith First Union Corp. The purpose of the alliance was to combine

their marketing efforts into a trade finance network in Southeast Asia and

the Pacific Rim. Under the terms of the agreement, the network would in-

clude seven existing Lippo offices in Southeast Asia and the Pacific Rim. Inaddition, First Union planned to open a Los Angeles office at Lippo Bank, to

process letters of credit related services. Financial details were not disclosed

(February 15, 1995) [I C].

4. Banco Nacional de Mexico entered into a strategic alliance with Wells Fargo

Bank, a unit of Wells Fargo & Co., to develop a series of banking products

and services. Under the terms of the agreement, the two banks would offer

foreign trade financing and related products to firms operatingin each of their

respective markets. In addition, the banks would offer financial products to

firms and individuals with operations in Mexico and the United States, with

each attending clients of the other in its home market. The banks believedthe alliance would capitalize on the rapidly expanding interaction between

their respective customer bases, which was enhanced by the North American

Free Trade Agreement. Financial details were not disclosed (May 15, 1995)

[I R].

Legend: D = Domestic I = International R = Related C = Cross-Product

References

Bank for International Settlements, 2001. Group of Ten Report on Consolidation in the Financial Sector.

Baradwaj, B., D. Dubofsky, andD. Fraser, 1991.Bidder returns in interstate andintrastate bank acquisitions,

Journal of Financial Services Research5, 261273.

Baradwaj, B., D. Fraser, and E. Furtado, 1990. Hostile bank takeover offers: Analysis and implications,

Journal of Banking and Finance14, 12291242.

Barber, B. and J. Lyon, 1997. Detecting long-run abnormal returns: The empirical power and specification

of test statistics,Journal of Financial Economics 43, 341372.

Berger, A., R. Demsetz, and P. Strahan, 1999. The consolidation of the financial services industry: Causes,

consequences, and implications for the future, Journal of Banking and Finance23, 135194.

Bodnar, G., C. Tang, and J. Weintrop, 2001. Both sides of corporate diversification: The value impacts of

corporate geographic and industrial diversification.Working Paper, NBER WP #6224.

Boehmer, E., J. Musumeci, and A.B. Poulsen, 1991. Event-study methodology under conditions of event-induced variance,Journal of Financial Economics 30, 253272.

Chan, S., J. Kensinger, A. Keown, and J. Martin, 1997. Do strategic alliances create value? Journal of

Financial Economics46, 199221.


22/22


Conrad, J. and G. Kaul, 1993. Long-term market overreaction or biases in computed returns?Journal of

Finance48, 3963.

Contractor, F., 1990. Contractual and cooperative forms of international business: Towards a unified theory

of modal choice,Management International Review30, 3154.

Cornett, M. and H. Tehranian, 1992. Changes in corporate performance associated with bank acquisitions,

Journal of Financial Economics31, 211234.

Cummins, J., S. Tennyson, and M. Weiss, 1999. Consolidation and efficiency in the U.S. life insurance

industry,Journal of Banking and Finance23, 325357.

Edmonson, G., and A. Robinson, 1999. Bank invaders scales the Pyrenees, Business Week, November 29,

62.

Fried, H., K. Lovell, and S. Yaisawarng, 1999. The impact of mergers on credit union service provision,

Journal of Banking and Finance23, 367386.

Gupta, A., R. LeCompte, and L. Misra, 1997. Acquisitions of solvent thrifts: Wealth effects and managerial

motivations,Journal of Banking and Finance21, 14311450.Harrigan, K., 1985.Strategies for Joint Ventures (Lexington Books, Lexington, MA).

Hughes, J., W. Lang, L. Mester, and C. Moon, 1999. The dollars and sense of bank consolidation,Journal

of Banking and Finance23, 291324.

Jacobsen, S. and A. Tshoegl, 1999. The Norwegian banks in the Nordic consortia: A case of international

strategic alliances in banking,Industrial and Corporate Change8, 137145.

Kogut, B., 1988. Joint ventures: Theoretical and empirical perspectives, Strategic Management Journal9,

319332.

Kvint, V., 1998. Nature of international joint ventures and their role in global business, in International M

and A, Joint Ventures and Beyond, ed. D. BenDaniel and A. Rosenbloom (John Wiley and Sons, New

York).

Lorange, P. and J. Roos, 1992.Strategic Alliances: Formation, Implementation, and Evolution(Blackwell,

Oxford).Madura, J. and K. Wiant, 1994. Long-term valuation effects of bank acquisitions, Journal of Banking and

Finance18, 11351154.

McConnell, J.,and T. Nantell, 1985. Common stock returns and corporate combinations: The case for joint

ventures,Journal of Finance40, 519536.

Milbourn, T., A. Boot, and A. Thakor, 1999, Megamergers and expanded scope: Theories in bank size and

activity diversity,Journal of Banking and Finance23, 195214.

Newman, L. and G. diCiccio, 1998. Strategic Choices, inInternational M and A, Joint Ventures and Beyond,

ed. D. BenDaniel and A. Rosenbloom (John Wiley and Sons, New York).

Palia, D., 1994. Recent evidence on bank mergers, Financial Markets, Institutions, and Instruments 3,

3659.

Rhoades, S., 1998. The efficiency effects of bank mergers: An overview of case studies of nine mergers,

Journal of Banking and Finance22, 273291.Sushka, M., and Y. Bendeck, 1988. Bank acquisitions and shareholder wealth, Journal of Banking and

Finance12, 551562.

Williamson, O., 1975. Market and hierarchies: Analysis and antitrust implications (Basic Books, New

York).

Williamson, O., 1985.The Economic Institutions of Capitalism (Free Press, New York).

Zhang, H., 1995. Wealth effects of U.S. bank takeovers, Applied Financial Economics5, 329336.

Evidence on Value Creation in the Financial Services Industries

Documents

Transcript of Evidence on Value Creation in the Financial Services Industries