Domestic and Cross-Border Merger and Acquisition Factor
Disclaimer: This dissertation has been submitted by a student. This is not an example of the work written by our professional dissertation writers. You can view samples of our professional work here.
Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of UK Essays.
Determinants Of Domestic And Cross-Border Merger And Acquisition Activities In Banking Sector
Global announced mergers and acquisitions (M&As) in banking sector rose to a new high record in 2007. Although now M&A activity has been hit notably because of the global financial crisis, it will have different performance in the time of post-crisis. Both China and UK will be the most active M&A areas in the future, especially in the financial service industry. To secure the success in future M&A wave, identifying the right potential targets is crucial.
The purpose of the study is to identify the determinants of bank's domestic and cross-border acquisitions in China and UK over the two specific periods which are before global financial crisis (2005-2007) and during crisis (2008-August 2009). This study will help bank managers to identify the right targets for future acquisitions and also help policy makers to understand which factors can increase the likelihood of bank acquisitions. Three aspects are examined by employing Logit analysis for the likelihood of domestic bank acquisitions which include bank's characteristics, market features and regulatory factors. Additional geographic factors will also be considered into cross-border bank acquisitions by using the same model.
The result of this study reveals the determinants of likelihood of being acquired for UK and Chinese banking industry respectively. It also shows the difference of determinants between domestic and cross-border acquisitions, and between the two specific time periods as well for two banking industries respectively. Detailed analysis is also provided for comparing the difference of determinants between China and UK, these two significantly different banking markets.
ABN AMRO, which created as of result of the 1991 merger between Amsterdam Rotterdam (AMRO) and ABN whose history dated back to 1824 and was one of the largest banks in Europe, was acquired by RFS Holdings B.V., a company jointly owned by RBS, Fortis N.V. and Banco Santander S.A. for a consequent amount of 71 billion Euros in October 2007 (Financial Times, 2007). This was the biggest banking takeover in history and together with other deals made global mergers and acquisitions (M&As) rose to a new high of US $ 4.8 trillion in 2007 (Francis and Hasan, 2008). However, the following global credit crunch has created a new international financing and banking network where M&As may be mainly involved as a survival solution for many entities.
This study will examine the determinants of bank's domestic and cross-border acquisitions both in UK and China, the most active and representative country in developed financial market and emerging market respectively.
Therefore, this chapter will give the overview of UK and Chinese banking industry, then focus on M&A activities, and illustrate objectives and the value of this study.
Global Banking Industry Overview
The global banking industry experienced strong growth before the end of 2007. Assets of the largest 1,000 banks in the world grew 16.3% during 2007 to a record $74.2 trillion (Figure 1). This growth rate is 5.4% higher than the previous year.
Worldwide assets of banking industry
The Banker database)
EU banks held the largest share, 53% in 2006/2007, up from 43% in 1999/2000. And among this, the share of UK banks is always on the top of other EU countries. US banks' share in term of assets remained relatively stable at the level around 14%. The left shares are from other Asian and European countries (IFSL, 2008a).
However, the global credit crunch which originated in the US sub-prime market impacted the whole financial system world widely. And it came close to collapse in the autumn of 2008, following the failure of Lehman Brothers and the ‘breaking the buck' of a large US money market mutual fund (MMMF). The subsequent panic was then across global financial markets especially the western world. Although in recent months, market conditions have picked up which improves the outlook for banking systems, there has been a dramatic shift based on market capitalization in global banking industries (Bank of England, 2009). In 2007, the average level of world top 20 financial institutions' market capitalization was peaked around $125bn, and banks from US and UK dominated the main financial market. However, in 2009, the market capitalization decreased by more than 50% for all financial institutions, and the top three largest ones are all Chinese banks. HSBS, which based on emerging markets at heart, is UK's sole representative in the list (Figure 2) (Financial Times, 2009a). Banking industry in both UK and China experienced great change through the global financial crisis.
UK Banking Industry
The UK banking industry is a vital and essential part of the UK economy. It has experienced a deep level of restructure since the 1980s which includes regulatory change and banking system reform etc. Four major changes are associated with this restructure: increasing level of competition, both in market and out-of-market; deregulation; increased diversification and acquisitions (Drake, 2001). In 2007, UK banking industry reached a new high record for its assets and contribution percentage to UK economy comparing to other sectors. Assets of UK banking sector reached £6,964bn at the end of 2007, up 11% on 2007. And there were 331 banks authorized to conduct business in UK. The 254 foreign banks physically located in the UK which is more than in any other country (IFSL, 2008a). Banking industry accounted for 7.6% of UK GDP in 2007, increased from 5.7% in 1997 (IFSL, 2008b).
However, the following global financial crisis impacted the UK banking industry materially. During 2008, the return on the major UK banks' total assets turned negative and the impact of this on return on equity was amplified by banks' high level of gearing (Figure 3). Meanwhile, the mark-to-market losses of major UK banks' book assets roughly doubled between October 2008 and January 2009, exceeding fresh capital raising over that period which partly reflected increase in expected losses (Figure 4) (Bank of England, 2009). But in recent months, market conditions have been improved, and it can be expected that UK banking industry which experienced restructuring and changed policies on bank regulations during the crisis will seek substantial domestic and cross-border business opportunities afterwards to strength its position in global financial market which gives great potential for M&A activities (Economics Outlook, 2009).
Bank Of England, Bloomberg)
Chinese Banking Industry
China's economy has been growing about 8% per year in real terms over the last decade. Even during the financial crisis, the Chinese economy continued to grow in a steady way (People's Bank of China, 2008). This rapid growth may be largely linked to the globalization of trade, but China has yet to ‘globalize' its banking sector.
The Chinese banking industry was dominated by four large state-owned banks with about 80% of industry assets, and very few foreign banks before WTO entry in December 2001. After 2001, the existing regulations and laws such as the Central Bank Law and Commercial Bank Law were revised to be aligned with the WTO agreement. There will be more liberalization of interest rates, more fair treatment of tax rates, less restrictions on ownership of takeovers and M&As, and great freedom of operational and geographical scope in the Chinese banking industry (Berger et al., 2009). In 2003, China Banking Regulatory Commission (CBRC) updated guidelines to encourage foreign share purchases. Foreigners then can own up to 25% of any domestic bank, with the ownership from any one investor allowed between 5% and 20%. This led to a historical high level of cross-border M&A activities in Chinese banking industry later on (CBRC, 2005). At the end of 2007, total assets of banking institutions increased by 19.68% to RMB 52.6 trillion than 2006. And the reform and opening up of Chinese banking industry continued to advance, the assets share of large state-owned commercial banks decreased from around 80% in 2001 into 53.25% while others includes foreign banks increased into 15.98% (Figure 5). This indicated that Chinese banking industry encouraged foreign investors more than before which gave big potential for further cross-border M&As in China.
People's Bank Of China)
Even in 2008's special context, Chinese banking industry followed the reform strategy and promoted the reform and opening up policy of banking industry, and its total bank assets amounted to RMB 62.4 trillion, 18.7% higher than in 2007 while most banks in western world decreased their bank assets by more than 50%. The Chinese banking industry took the financial crisis as an opportunity for further reforms, development and finally achieving globalization in banking sector (CBRC, 2009). This growth trend also provides great potential for both domestic and cross-border M&As in China which can help strength the market position and substantial development for Chinese banks.
The Merger And Acquisition Market
There are two primary mechanisms by which ownership and control of a public corporation can change: merger and acquisition. In both cases, the acquiring entity must purchase the stock or existing assets of the target either for cash or for something equivalent value (such as shares in the acquiring or newly merged corporation). Mergers and acquisitions (M&As) are part of what is often referred to as ‘the market for corporate control' (Berk and DeMarzo, 2007).
The global merger and acquisition market is highly active, averaging more than $1 trillion per year in transaction value. Global announced mergers and acquisitions (M&As) rose to a new high of US $4.8 trillion in 2007, up 23% from US $3.9 trillion in 2006. Since the increasingly fierce competitions in banking sector during this time, banks conducted large-scaled M&A within and across the border in aim to expand operation and enhance competitiveness. In 2007, M&As in banking increased 32.5% from US $543 billion in 2006, was again the most active sector for M&As (Francis and Hasan, 2008). But M&A activities correlate with bull markets and are often greater during economic expansions than during contractions.
During the past year when global banking industry was suffered by the credit crunch which originated in the US sub-prime market, M&A activities were hit notably in financial service sector. The worldwide M&A volume decreased significantly during the end of 2007 and the beginning of 2008. But, against all the negative factors, the trend of selective M&A still exists as ‘acquirers take advantage of the lower market valuations to strengthen their business with synergistic deals' and ‘the ongoing strength of emerging market' (Financial Times, 2009b). From August 2008 by monthly comparison, the M&A volume increased again increased gradually and kept vibrating. On average, the announced deal number is around 3,000 and total rank value is about $160,000 million per month (Figure 6).
Thomson ONE Banker Database)
M&A Activities In UK Banking Industry
Banking industry is the most active sector in M&A market. And UK, despite of US, is more involved in M&A process than any other EU country in financial service industry (Ahammad and Glaister, 2008).
During 1999 to 2007, the number of UK incorporated banks which include commercial banks, investment banks, foreign owned banks and banks operated by retail companies declined from 200 to 157 mostly due to the severely competition and M&As in its banking industry (IFSL, 2008). Also, the rapid economic expansions during this period drive peaks in both domestic and cross-border M&A activities in UK banking industry (Harford, 2008). At the end of 2007, the total announced M&A deals in UK financial industry was 1,806, which represented as 38.3% of all M&A deals announced in this year. And among 1806 ones, UK banking industry contributed significant amount of deal values comparing to other financial sectors.
However, the following global financial crisis hit the UK banking industry significantly. Although the number of M&A activity is less during economic contractions than expansions, it still has its market due to various business strategies. There were 79 M&A deals totally in UK banking industry for 2008, and the number for 2009 as so far is 35. According to Economics Outlook, the post-crisis restructuring, recapitalization and seeking for re-strengthening business positions in UK banking industry will provide substantial opportunities for M&A activities both domestically and internationally. This will produce an urgent demand for identifying right potential merger and acquisition targets which gives great practical value for this study.
M&A Activities In Chinese Banking Industry
While the economic market get more international and worldwide, to be globalization has become an irreversible trend for all nations in the world, especially for developing countries (Bonin and Hasan, 2005). Although Chinese banking industry has achieved sustainable development during these years even in the global financial crisis, it's still far away from globalization.
Entry to WTO in 2001 and new policy for foreign investment in Chinese banking industry in 2003 brought momentums for encouraging Chinese banks to participate in global competitions. To strength their own positions in global financial market, most Chinese banks took actions of within or cross border M&A activities which led by Industrial and Commercial Bank of China (ICBC) acquiring 20% share of South Africa's Standard bank for $5.6 billion, the largest M&A deal in Chinese banking industry in 2007 (Munroe, 2008). Further opening up in Chinese banking industry also attract strategic foreign investors invest in Chinese domestic banks through M&As. At the end of 2007, foreign financial institutions invested in 25 domestic banks totally through partially acquisition (People's Bank of China, 2008).
With the rapid development and increased opening-up of Chinese banking sector, banks' M&A activities will be increasingly active. According to People's Bank of China, effort will be made to vigorously support qualified commercial banks to conduct M&A and create favourable policy environment based on international experiences to encourage cross-border M&As in China. Moreover, the ongoing strength of Chinese economic will help to fund or attract domestic and overseas acquisitions in China (Wilson, 2008). It can be expected that China will be the most active M&A area in the emerging market in future.
To sum up, both UK and Chinese banking industry have the great potential for future M&A activities. To secure the successful performance in the present and future M&A waves in order to obtain sustainable business growth, identifying the right potential acquisition targets is crucial (Rossi and Volpin, 2004). However, investigating the relationship between bank and the features of market where banks are within and the likelihood of being a right potential acquisition targets in banking industry is relatively under-researched(Pasiouras et al., 2007). And it will be interesting to investigate and compare the determinants of potential M&A targets between UK and China's banking industries which have significant differences in characteristics, and also for two special periods which are before and during the global financial crisis.
Research Objectives And Value
The purpose of this study is to identify the determinants of bank's domestic and cross-border acquisitions in China and UK, and detailed analysis will also be provided for the difference of determinants between China and UK, the two relatively different banking sectors. Moreover, it will compare the difference between two time periods which are before global financial crisis (2005-2007) and on crisis (2008- August 2009).
Three aspects will be examined for the likelihood of domestic bank M&As which included bank's characteristics, market characteristics and management incentives. Another two aspects which are geographic factors and regulatory barriers will be also considered in cross-border bank M&As. Same Logit analysis model will be employed to analyze the domestic and cross-border M&As for the purpose that the comparisons have the same base.
This study has the value for bank managers in China and UK to identify the most suitable targets or to check if their own bank has developed a profile that similar to typical target. And also, it's meaningful for policy makers to understand which factors would increase the possibility for bank acquisitions (Scott, 2007).
The study has the originality in three aspects. First, it combines all the important factors that will influence the likelihood of acquisitions in banking industry and distinguishes the domestic and cross-border acquisitions according to the gaps in the literatures. Second, this study investigates and compares two important banking sectors (China and UK) which haven't done by any researchers before. Last but not the least, this research focus on bank acquisitions over a most recent time period, from 2005 to August 2009, which is the period includes prior global financial crisis when M&As rose to a new high record and on-crisis period. This differs from the prior studies that focused on earlier time periods and will reflect the new trend for M&As. According to Hagendorff et al. (2008), the more recent M&As may be qualitatively different from those in earlier periods which suggests that studies focusing on more recent M&As may provide more relevance to likely future takeovers.
The relationship between the bank and features of the market where banks are within and the probability that a bank will be a potential target remains an open question (Cyree et al., 2000; Wheelock and Wilson, 2000). Few studies in the literature have examined the major features of banks which are acquired by other organizations (Hannan and Rhoades, 1987; Moore, 1996; Hadlock et al., 1999; Wheelock and Wilson, 2000, 2004) are focused on the US market, while Pasiouras et al.(2006), Shen and Lin (2007) and Hernando et al.(2009) have studied the Greece, Asia and EU banking industry respectively.
Gaps In Previous Studies
Hanna and Rhoades (1987) examined the likelihood of an acquisition based on the banks performance using a sample which was consisted 1046 acquired and non-acquired banks in Texas between 1970 and 1982. The results showed that market concentration and high capital asset ratios have negative relationship with the probability of bank's acquisition. Moore (1996) also investigated the characteristics of US banks acquired between 1993 and 1996 using multinomial logit analysis. However, both studies mainly focused on financial characteristics of banks, but ignoring the external factors such as regulations. Based on these, some other studies focused on the search of the best predictive variables included bank characteristics, market features and regulatory factors (Bartley and Boardman, 1997; walter, 1998; Cudd and Duggal, 2000).
Hadlock et al. (1999) researched a sample of 84 acquired and non-acquired US banks during 1982 and 1992 by employing both univariate and multivariate methods to identify the determinants of acquisitions. However, the variables they analyzed mainly focused on the management incentives. In more recent studies for US banking sector, Wheelock and Wilson (2000, 2004) used proportional hazard models and a two-part hurdle model by collecting massive amount of available data and employing relatively comprehensive variables included financial, market and regulatory factors to investigate the determinants of likelihood of bank's acquisitions. They found that regulatory approval process and market concentration are negatively related to the likelihood of M&A activities, while management incentives, location, banks' size, and capital strength are positively related. However, they didn't identify the difference of determinants between domestic and cross-border acquisitions.
More recently, Pasiouras et al.(2006) investigated the Greece banking industry to analyze the determinants of bank acquisitions, but they ignored the management incentives and corporate governance factors due to lack of data available. In a later study, Parisouras and Gaganis (2007) also investigated the financial characteristics of bank acquisitions covering the 5 principal EU banking sectors (France, Germany, Italy, Spain and UK). However, they didn't distinguish domestic from cross-border takeovers in their studies.
Shen and Lin (2007) studied the determinants of financial institutions which engaged in cross-border M&A activity before and after the 1997 Asia financial crisis. They found that regulation barrier and market opportunities have less impact on the takeovers after crisis while geographic factors are important determinants both prior to and post Asian crisis.
Hernando et al. (2009) analyzed the determinants of bank acquisitions both within and across 25 members of the European Union during the period 1997 to 2004. Their results suggested that determinants of domestic and cross-border takeovers appear to be different in several aspects such as market concentration and profitability of banks performance. However, they examined all the variables according to the experiential model which generated from the US banking sector. The model can be argued if it is applicable to the EU banking industry. Other studies about the determinants of bank acquisitions mainly focused on the search of the most effective empirical method for the development of the prediction models (Cheh et al., 1999; Doumpos et al., 2004; Espahbodi, 2003).
This study is based on previous research, and will cover the above identified gaps which include examining domestic and cross-border M&As respectively by using the same Logit analysis model, studying the two typical and representative banking markets in UK and China, and analyzing all the various factors typically found to be the most likely determinants of bank M&A activities in the literature. And these factors will be detailed in the following section related to previous studies.
Possible Determinants Of Banking M&As In Literature
Seven factors of bank features are mainly analyzed in the literature which are related to the likelihood of being acquired.
The main motive underlying acquisitions is the target bank is underperforming. The inefficient management hypothesis (Manne, 1965) argues that if management can't maximize the shareholders' wealth by using the resource it has, then the firm is more likely to be acquired so the inefficient management will be replaced. Then there will be the space for the acquirer to improve the performance and efficiency of target and increase total profitability. Therefore, indicators of bank performance should contain explanatory power on the likelihood of being required. But the empirical results are mixed. Hannan and Rhoades (1987) found no evidence to support this hypothesis while Moore (1996), Focarelli et al. (1999), Wheelock and Wilson (2000), Pasiouras et al. (2006) and Hannan at el.(2007) found that less efficient and profitable banks are more likely to be acquired.
However, while underperforming banks may provide greater opportunities for further improvement of profitability, they are also more risky, especially if the source of the underperformance is a high level of loans (Hernando et al, 2009). Hannan and Rhoades (1987) demonstrated that a high level of loans would indicate the aggressive business strategy of target bank and a penetrated and strong established client networking which will make it more attractive as a target while a bank with a low level of loans due to its conservative management may also be attractive to the acquirers since acquirers can use more aggressive way to increase returns of the target. And they found loan activity was negatively related to acquisition likelihood but not ‘statistically significant'. But Moore (1996) found a negative and significant relationship for both in-market and out-of-market acquisitions. Moreover, the results of Wheelock and Wilson's studies (2000, 2004) were mixed. They found it depended on the specification of the estimated model, in some cases it was negatively related but in others, it was positively related with not always statistically significance.
As stated by Pasiouras et al. (2007), liquidity is an additional factor that can affect the attractiveness of banks as targets since ‘the process of managing assets and cash flow to maintain the ability to meet current liabilities as they come due' is an important decision for managers. This argument is supported by Wheelock and Wilson (2000) in their study. But it is also possible that some banks be acquired because they have liquidity issues and turn to help to acquirers. In the study of Pasiouras et al.(2007), it found no significant correlation between liquidity and likelihood of being acquired.
Another important bank characteristic for likelihood of being acquired is the capital strength while there are different hypotheses associated with this (Hernando et al, 2009). Several hypotheses predict a positive relationship between banks' capitalization and the likelihood of involving into acquisitions. One is that if high capitalization indicates inefficiency of a bank to diversify its assets, then better diversified acquirers will be attracted by such banks. Another one is that if acquirers face regulatory pressure of capital requirement, they may seek highly capitalized targets. Finally, banks with high capital ratios may be operated further below their potential profit due to less pressure to managers. While on the other hand, some hypotheses predict a negative relationship. One of them is an acquisition by a well capitalized acquirer might be stimulated by the supervisor if the target has low level of capitalization. And Hanna and Prilloff (2007) also argue that ‘acquirers prefer low capitalized targets because it enables them to maximize the magnitude of post-acquisition performance gains relative to the cost of achieving those gains'. The empirical results for this are mixed as well. Akhigbe et al. (2004) found a positive relationship between capitalization and the likelihood of being a target in study of publicly traded banks in the U.S. While most studies found the relationship is negatively related (Hanna and Pilloff, 2007; Lanine and Vander, 2007; and Pasiouras et al., 2007).
Bank's size is another characteristic which may influence the likelihood of being acquired. Smaller banks may be more attractive to the acquirers since it's easy to finance and even integrate after the acquisition. However, if the acquirer is seeking economies of scale or market power through acquisition, especially for the cross-border acquisitions, size may have a positive influence on the likelihood of being acquired. Hannan and Rhaodes (1987) and Moore (1996) have not found a significant relationship between bank size and the probability of being a target while Wheelock and Wilson (2000), Focarelli and Pozzolo (2001) and Hannan and Pilloff (2007) find that larger banks are more likely to be acquired when they estimate their model using full samples. Lanine and Vander (2007) and Pasiouras et al. (2007) also have positive results in their studies. But Hanna and Pilloff (2007) also point out that for the acquisitions by smaller banks, larger banks are less likely to be acquired due to the difficulty of post-acquisition integration.
Market share is an additional variable for the reasons of M&A activities. Bodie et al. (2008) argues that market share is one of the most important factors which impact the acquirer's decision in domestic and cross-border acquisitions in banking industry. It is similar to the variable of bank's size, but provides a relative standard to evaluate the target comparing with others in the same industry. In the banking industry, a bank with small share is likely to be acquired since only banks with substantial market share can compete effectively and the assets of banks with smaller shares will be more valuable after being acquired by the large bank. But regulatory concerns about anti-monopoly for banks with large market share will give the negative effect on the likelihood of being acquired and high market share. The empirical results for this factor are mixed as well. Moore (1996) and Pasiouras et al. (2007) found that it is significantly and negatively related with the probability of acquisition in in-market M&As while Hanna and Rhoades (1987) found it has positive impact on the acquisition probability.
Finally, prospects of bank's future growth can affect the acquirers' M&A decision as well. Banks which experience high growth may be more attractive to the acquirers as potential gains raised from the expanding markets after acquisition can be expected more than before. Consistent with this hypothesis, Hannan and Rhoades (1987) and Cheng et al. (1989) find that the likelihood of acquisition is positively related to the potential growth rate of the assets of the target banks in their studies of U.S. banks in 1980s. However, Moore (1996) argues that slower growing banks may attract a buyer who is looking to increase the target's growth rate through efficient management. Together with Moore (1996), Pasiouras et al. (2007) find a negative relationship between bank's growth rate and the acquisition probability. But Hanna and Pilloff (2007) and Lanine and Vander (2007) do not include this variable into their studies.
Three main independent factors are discussed in literature about market characteristics which may influence the acquisition probability of a particular bank.
First one is market concentration. Through the impact on bank competition, the degree of bank market concentration potentially affects the likelihood of acquisitions. Increased concentration may increase the attractiveness of the target banks in that market. But buyers that would want to increase the concentration further may face the pressure from anti-monopoly authorities (Hernando et al, 2009). Hanna and Rhoades (1987) found there is a negative relationship between market concentration and the takeover probability for out-of-market acquisitions while it is significantly positive for in-market acquisitions. Moore (1996) found no statistically significant relationship between them for in-market takeovers but a positive sign for out-of-market ones. Hannan and Pilloff (2007) also fail to find any statistically significant evidence that market concentration is a determinant of takeover targets. However, Pasiouras et al. (2007) stated a significantly negative coefficient on the five large banks concentration ratio in their sample of European takeovers.
Harford (2008) argues that there is a correlation between industry profitability and M&A activities within this industry. M&A deals are often greater in more profitable industries than those less profitable ones. According to Thomson ONE Banker database, there are less M&A deals in UK banking industry in 2008 than ones in 2007 when the whole banking industry was experienced the global financial crisis. This also gives the evidence that the level of industrial profitability has the positive relationship with the likelihood of acquisitions. However, Ali-Yrkko (2002) points out that in the beginning of 1990s, the entire banking was restructured due to the extremely deep bank crisis with large bankrupts. Low profitability may be one of the main reasons which lead to higher level of acquisitions in banking industry in an attempt of banks to be restructured.
Another main factor which was discussed in literature is the market growth. As discussed in Chapter 1, M&A activities correlate with bull markets and are often greater during economic expansions than during contractions (Francis and Hasan, 2008). Audretch (1989) and Schoenberg (1999) also indicate that firms have been more attracted to make acquisitions within industries which have higher growth rates.
It is possible that managers put their own interests over shareholders. To the extent that managers in banks may lose the job positions or, at least, may suffer a reduction in their executive autonomy, they may oppose takeover bids even if the deals are maximizing the shareholders' value. Or if there is an extremely ‘lucrative severance package' guaranteed to managers in the event that the firm is taken over, then managers have more incentives to let their firm be acquired (Cuthbertson and Nitzsche, 2009). Thus, management incentive is another factor which will affect the probability of being acquired for a specific bank. Moreover, Hadlock et al. (1999) argues that banks with higher levels of management ownership are less likely to be acquired, especially in acquisitions where managers in target banks should leave from the organization post-acquisition.
It is reasonable that the regulatory restrictions towards M&A activities by the local authority of a particular country can be a critical factor which affects the attractiveness of the potential acquired targets, especially in cross-border M&As. On the one hand, putting the foreign investment limits on the capital or blocking single takeovers would definitely reduce the number of cross-border M&As. And moreover, the extremely strict regulations on anti-monopoly will reduce the attractiveness of potential targets in that market if acquires want to gain scale of economies through M&As (Straub, 2007). On the other hand, regulatory restrictions would reduce the degree of information asymmetry such as the more transparent information of banks' financial information. And this will make banks in such environment more attractive to buyers. Or because of the difference for regulations exists in different countries, banks with relative advantage positions of regulations will have higher likelihood to be acquired by those banks who would likely have a great incentive to expand the business abroad in order to bypass their own country's regulatory restrictions (Galindo, Micco and Serra, 2003). Shen and Lin (2007) also argue that the regulatory differences between home and host countries have positive effects on cross-border M&A deals in banking industry.
It is relatively difficult to measure the regulatory factors quantitatively. Two categories of regulatory restrictions are used in the previous studies. First category where the regulatory restrictions are considered in a broad sense includes the rule of law, institutional quality and government effectiveness. The proxies for this category are legal origin (La Porta et al., 1997,1998), regulatory burden, corruption, rule of Law and government efficiency (Kaufmann et al., 2000). According to the previous studies, countries with a relatively more judicial system are more attractive to the foreign investors since the market there is guaranteed as more fair and transparent (Focarelli and Pozzolo, 2000; Shen and Lin, 2007).
The second category of regulatory restrictions comprises restrictions on banking activities in securities, insurance, commodity and real estate which is taken from the study of Barth et al. (2000). Shen and Chang (2005) argue that these restrictions may have harmful impact on the performance of banks, and prevent other banks abroad enter into the country by M&As. Barth et al. (2001, 2004) point out that stricter restrictions actually reduce the number of acquisitions cross-broadly. Claessens and Klingebiel (2000) also find that fewer regulatory restrictions permit the utilization of economies of scale and scope which will increase the likelihood of acquisitions. Focarelli and Pozzolo (2001) find the similar results as well.
Geographic factors including distance, language, culture, and currency are studied by Berger et al. (2000) for their relationship with the likelihood of being a target in banking industry in Europe. Buch and Delong (2001) also examine three geographic factors which include distance, language and common legal system and they find that M&A activities happen more frequently for cases that managers in acquired and acquiring banks have the same language and are also easy to communicate with each other in terms of time zones and geographical distance.
Some previous studies have found that these geographical factors are negatively linked with the probability of being acquired for banks unless the both target and takeovers are well known by each other about whole transaction behaviours and post-acquisition restructuring processes as well (Dunning, 1998; Wei et al., 1999, Shen and Lin, 2007). Most literature which study for the likelihood of banking acquisitions include no geographic factors into their research (Hanna & Rhoades, 1987; Moore, 1996; Weelock and Wilson, 2000, Wansley and Boehm, 2000, Pasiouras et al., 2006, 2007).
This study is based on quantitative analysis, using collected data to investigate the determinants of domestic and cross-border bank acquisitions. Logit analysis model will be employed to examine the related variables in both domestic and cross-border M&As in UK and Chinese banking industry. The whole process of research design is showed as Figure 7. It will begin from data collection and selection, follow by defining all the variables for domestic and cross-border acquisitions respectively, then examine all the variables by employing the same model (for the purpose that all the following analysis and comparisons have the same base), and finally analyze the results and make comparison and explanation.
This study employs Logit analysis to investigate the determinants of both domestic and cross-border bank acquisitions. All the variables in this model can be numerically expressed.
In previous studies, most of them have used multivariate statistical and econometric techniques such as discriminant analysis (Barnes, 1990; Stevens, 1973) and Logit analysis (Barnes, 1998, 1999; Powell, 2001; Pasiouras et al., 2006, 2007) and only more recently the parametric nature and the statistical assumptions of those approaches have led researchers to the application of alternative techniques such as artificial neural networks (Cheh et al., 1999).
However Barniv and Mcdonald (1999) summarized some of the problems related to the use of discriminant analysis (DA) and logit analysis (LA). They pointed out that DA is generally sensitive to departure from normality and requires strong assumptions. Hopwood et al. (1998) also showed that DA has less accuracy than LA in certain situations. Moreover, Manel et al. (1999) assessed the performance of DA, LA and artificial neural networks (ANN) and pointed out that LA correctly predicted more cases (82%) than DA (69%) and ANN (73%) in a ‘leave one out' data partitioning approach and in other data partitioning procedure, all the methods performed similarly. They also demonstrated LA is more efficient in the use of computer time than ANN, and also more straightforward in providing testable hypotheses. Furthermore, Barros and Hirakata (2003) tested Poisson regression (PR) and LA in the research of determinants of cross-border acquisitions, and found that results from PR for cross-border cases do not departure significantly than the results from LA although PR provides correct estimates and the prevalence ratio is more interpretable and understandable for non-specialists than odds ratio. And Hernando et al. (2009) point out that it's better to use the same model to analyze domestic and cross-border M&As since it gives the same mathematical base to compare the results. Thus, in this study, Logit analysis will be employed to examine the various determinants of both domestic and cross border acquisitions in banking industry.
According to Pasiouras et al. (2007), for any given year, there will be two mutually exclusive events for a specific bank which are:
(1) It may be acquired by another organization within or cross border, or
(2) It may not be acquired.
Then the Logit model can be developed into this specific situation. Dependent variable whether the bank is acquired or not in year t in this model is defined as following: y=0 for all non-acquired banks in year t, and y=1 for acquired banks. And independent variables which include bank features, market characteristics, regulatory restriction factors and geographical factors are as value X in this model. Also, it is argued that since the bank acquisitions would take some time to complete, the independent variables using in the model should be picked as data in year t-1 for the acquisitions that completed in the year t (Hannan and Rhodes, 1987; Wheelock and Wilson, 2004).
Then the Logit analysis model can be explained as the following:
Where y=1 if the bank is acquired in year t, and y=0 if the bank is not acquired in year t; E(y) =P(acquired)=X, and X stands for the probability that y=1 in year t; b0 is the intercept term and b1, b2, …, bn are the coefficients of independent variables; and x are the independent variables in year t-1.
Data And Data Source
Sample Selection In UK banking Industry
It is necessary to understand the structure and assets distribution of UK banking industry in order to select the fair samples which can represent the whole UK banking industry. According to IFSL (2008) and Bank of England (2008), foreign banks held the majority assets of UK banking sector since 2000, and the number of foreign banks also increased during these years. In 2007, 58% of total assets in UK banking sector was held by foreign banks while UK incorporated banks had 42% of that. And foreign banks accounted for around 59% of total UK deposits. Also, the number of foreign banks at the end of 2007 reached 254, where the number of UK incorporated banks declined into 157 from 200. Thus, the samples should include representatives from both incorporated banks and foreign banks in UK. Moreover, the top 30 banks which include foreign banks in UK (ranked by the total assets according to BankScope database) have a market share around 78% in terms of total assets, and slightly higher shares in terms of lending and deposits. Based on these facts, these 30 banks can form a fairly representative set of samples for the universe of banks operating in UK according to Pasiouras et al. (2006, 2007).
To conduct the final acquired or non-acquired samples based on the initial sample set of 30 banks, three principles are set up in this study:
(1) Banks should have annual reports since 2003;
(2) Targets which are affiliated with the acquirers before the acquisitions (such as parent company acquires subsidiary companies) should be excluded (Shen and Lin, 2007; Pasiouras et al., 2007; and Hernando et al., 2009);
(3) All government bailed out cases are excluded from this study since purpose for such acquisitions is significantly different;
(4) All financial ratios from 2003 can be obtained from BankScope database in order to have same base for comparisons.
Finally, based on the final sample set and the data of M&A deals which obtained from Thomson ONE Banker for UK banking industry in the time period from 2005 to August 2009, the final dataset of acquired and non-acquired banks which consists of a total of 117 observations is shown as the Table 1.
Sample Selection In Chinese Banking Industry
Similarly to the sample selection process of UK banking industry, for Chinese banking industry, first step is to understand the structure and assets distribution of the whole sector. Chinese banking industry is mainly dominated by the big five commercial banks. As of the end of 2007, the assets of big five commercial banks accounted for 53.25% while the market share of lending was 49.87%. Joint stock commercial banks (JSCBs) and rural commercial banks have the market shares in terms of assets increased into 13.78% and 10.64% respectively, and the lending shares as 14.44% and 11.29% respectively (People's Bank of China, 2008). The details of structure for Chinese banking sector are presented in Table 2.
In this study, banks in ‘others' category are excluded from the analysis since most of banks in others are policy-related banks, postal savings banks and foreign banks which are mainly focused on the niche market in China and the features of these banks may can't directly compare with other commercial banks on the same basis. Moreover, top 30 banks in Chinese banking industry represent 81.3% market share in terms of total assets in Chinese banking industry which include large commercial banks, all JSCBs, and banks from rural commercial banks and city commercial banks. Thus, the initial samples for this study are these 30 banks in China.
According to the same selection principles which applied for UK banking industry, the final sample set for acquired and non-acquired banks in Chinese banking industry can be conducted. Table 3 presents the final data set for all 142 research observations in Chinese banking sector within specific time periods.
According to literature review, variables which reflect bank's characteristics, market features, regulatory factors and geographic factors will be analyzed in this study. However, management incentives will not be included in the study because of data unavailable. Table 4 lists all the variables which will be used in the analysis.
For bank's characteristics, the proxy for target's performance in this study is cost-to-income ratio where higher value indicates greater cost inefficiency. Thus, the expected sign on coefficient for this ratio is positive if underperforming banks are the more attractive targets.
The level of loan activity is measured by the net loans-to-total assets ratio according to the previous studies. But there is no assumption about its relationship with likelihood of bank acquisitions.
This study will use the ratio of liquid assets to customer and short term funding to measure the liquidity capacity. This ratio measures the percentage of short term debt could be met if they are withdrawn suddenly. And higher this ratio, higher liquidity capacity the bank has (Atrill and Mclaney, 2006). But the sign of coefficient for this factor can't be expected as well.
The capitalization of the target is measured as the ratio of equity to total assets. As discussed before, the sign for coefficient will be expected as mixed which depends on which effects dominated. It will be positively related if: the target is believed less diversified and less efficient than acquirer, or acquirers face regulatory pressure of capital requirement, or less management incentive to increase the performance due to high capitalization level. And the relationship will be negative if: banks that are near or below supervisory minimums are more likely to be acquired, or the acquirers will pay higher premiums as a proportion of equity for less well capitalized banks.
The proxy for a bank's size is the logarithm of its total assets as in previous studies. The expected sign on this variable is positive if acquirers view the larger targets can provide the business scale and scope to strengthen market position. However, the relationship can be negative to the extent that ‘the acquirer's primary concern if the difficulty in combining the target's operations with that of the acquirer' (Hernando et al, 2009).
And the proxy for market share is the ratio of deposits of the bank to the total deposits of the banking sector over the same year. This variable is expected to carry a negative sign if acquirers face the regulatory pressures for acquisition. And the sign will be positive if acquirers want to have synergy effect after acquisition.
Growth rate is measured as the annual growth rate of the target's assets in this study. It can be expected that the coefficient sign will be positive if growing targets provide the potential of more gains post-acquisition while it can be negative if slower growing targets provide the acquirers better opportunities to increase profitability through improving management.
For market features, this study will use the five large banks concentration ratio which is total assets of the five largest banks to total assets of the banking industry in the same market to measure the market concentration. But there is no prediction about its relationship with the acquisition likelihood.
The proxy for market profitability is the average cost-to-income ratio for banks in the same market. It is expected that the coefficient sign for this variable is positive before 2007 credit crunch if higher profitability in banking industry will bring more M&A deals. And the sign will possibly be negative in this study for on crisis period, since the banking industry may has to be restructured for better market profit and position.
And the market growth in this study is measured by the annual change of total assets in the specific banking industry. Similarly to market profitability, a positive linked relationship between market growth and the possibility of being acquired for a specific bank in this market before global financial crisis is expected while a negative relationship is expected for on-crisis period in banking industry.
For regulatory factors, two sets of regulatory variables are used in this study to measure the regulatory factors quantitatively. The first set is related to government governance. KKZ's indices of corruption, rule of law, quality of regulation and government efficiency are the measures in this set which adopted from Kaufmann, Kraay and Zoido's study (KKZ, 2002). In this study, it is assumed that these four factors are carrying the same weight for government governance, and then the proxy for this is the weighted KKZ indices. It is expected a positive relationship between governance and the likelihood of being acquired for banks in corresponding market.
Another set of regulatory variables includes the regulation restrictions on banking industry for the activities involving securities, commodities, insurance, real estate and non financial sectors according the study of Barth et al. (2000). In this study, it is also assumed these variables carry the same weight for regulation restriction factor, and then similar to governance, the weighted index which combines the all variables is the proxy for this factor. And it is expected negative relationship between these regulatory restrictions and the likelihood of being a target for a specific bank in the market.
For geographical factors, adopted from Shen and Lin (2007), three geographic variables will be analyzed in details which are language, common religion and common region in this study. These three variables will be processed as dummy variables, which means if both acquirer and target share the same language or religion, or are in the same region, then the dummy is the same unity; otherwise, it will be zero. Since these variables are only applicable to acquired banks, and can't measure for non-acquired banks, they are excluded from the Logit analysis, and will be analyzed individually.
Results For UK Banking Industry
Tables 5 and 6 show the descriptive statistics and univariate analysis for the differences between characteristics of acquired banks and non-acquired banks for period 2005-2007 which is before the global financial crisis and period 2008-August 2009 which is considered as on crisis.
For the period before global financial crisis, it is obviously that banks being acquired within border have higher cost-to-risk ratios than ones in cross border acquisitions, and even much higher than banks which are not acquired. This finding is consistent with Campa and Hernando (2006) who found that targets present a worse financial performance on average. Also, banks which are the targets of domestic acquisitions have highest liquidity level as well. Moreover, it can be seen in the above table that non-acquired banks have significantly high level of capital strength than acquired banks. For the factors of bank assets and market share, acquired banks have notably greater value than non-acquired ones, especially for cross border acquisitions. All variables related to market features are similar here for all banks, and it will further be analyzed in the Logit analysis. From this descriptive analysis, banks in UK which are underperforming, with weaker capital strength, and have higher assets and market share are more likely to be acquired. And this is more significantly for cross border acquisitions. But this is under the condition that the market is before global financial crisis.
Characteristics for banks being acquired during the crisis are different in four aspects than before crisis. First, the cost-to-income ratio for acquired banks during this period is much lower than the non-acquired banks, which is opposite to the period before crisis. Secondly, the liquidity level for all banks is lower in crisis than before, especially for banks which are acquired within border. Thirdly, capital strength is also weaker for banks during the crisis than before, but banks with lower capitalization are still more likely to be targets. Finally, while the average market share for non-acquired banks is lower than which is before the crisis, it is notably high for acquired banks during the crisis, particularly for cross-border acquisitions. It suggests that the determinants on bank's characteristics for the likelihood of being acquired are not the same for the two specific periods, and such differences may due to the various approaches for motivations of acquisitions. According to this descriptive analysis, market features as well as regulatory factors do not have significant difference before or in the crisis, but as said before, these will be further studied in the Logit model.
Geographic factors are been simply analyzed for cross-border acquisitions. For 10 and 6 acquisitions which announced before crisis and in crisis, 9 and 6 acquisitions respectively do not share the same language between acquirers and targets. Moreover, 40% and 50% of these acquisitions respectively are not in the same region for targets and buyers, and only 4 and 3 acquisitions are considered as having same religion. These figures demonstrate that geographic factors are not the important determinants for cross border acquisitions when the acquirers making their decision. And this is understandable since UK is one of the biggest financial centres world widely, and the increasing level of globalization makes the geographic factors even less impacted (Brealey and Kaplanis, 2000).
Totally, 12 variables which consist of 7 bank characteristics, three market features and two weighted regulatory factors will be analyzed in the Logit analysis. Correlations between these independent variables are firstly examined in this study. Table 7 presents the results of correlations between 12 variables.
It is very obvious that two variables which measure bank's size and market share have high correlations. And one variable measures stands for market feature has significant correlation with another regulatory restriction factor. Two sets of these correlated variables are: LOGAS and SHARE, MACI and BINDEX.
According to Pasiouras et al. (2007), four models were produced to by including the correlated variable one at a time in order to ‘reduce the multi-colinearity among the independent variables to the interpretation of results'.
The results of the four Logit models are presented in table 8 and 9 which are represented time period for before and in global financial crisis respectively. For time period 2005-2007, all models are statistically significant at the 1% level with Logit pseudo R2 in this study is between 0.41 in Model 2 and 0.63 in Model 4, which means the data generally fit into the presumed underlying distribution quite well. Similarly, for another time period which is during the crisis, pseudo R2 is between 0.43 (Model 1) and 0.65 (Model 4) which also present the good data fit for the underlying distribution.
Results For UK Banking Industry Before Crisis
The results of estimating the determinants on likelihood of being acquired for UK banking industry before global financial crisis are presented in the table 8.
For the time period before global financial crisis, consistent with Hadlock et al. (1999), Hannan and Pilloff (2007) and Pasiouras et al. (2007), bank performance in terms of management efficiency does not have a statistically significant relationship with the likelihood of being a target in acquisitions, but it is positively linked with the possibility of being acquired as expected both for domestic and cross-border acquisitions.
The bank's net loans to total assets ratio has a negative relationship with the likelihood of acquisitions which indicate that banks with low level of loans may be more attractive to the acquirers who can use more aggressive way to change the loan activities of targets and increase the business return finally. This finding is consistent with Hannan and Rhoades (1987) and Pasiouras et al. (2007), but different from Moore (1996) since his study states a significantly negative relationship between loan activities and the possibility of being acquired for banks.
For LIQ and EQAS which stand for liquidity and capital strength respectively, this study does not find a strong relationship between the likelihood and these two factors which is same in the Pasiouras et al. (2007) and Shen and Lin (2007)'s studies. But it does show a negative relationship for both domestic and cross border acquisitions.
Bank's size as measured by LOGAS in this study shows a converse relationship with the likelihood of being a target for domestic and cross-border acquisitions. The coefficient for domestic acquisitions is negative while it has positive sign for cross-border acquisitions. This indicate that foreign investors seek for achieving economies of scale and penetrating into the new market abroad quickly through acquiring the large cross border banks. Wheelock and Wilson (2000), Focarelli and Pozzolo (2001) have the similar results in their studies. But the purpose for domestic acquisitions may not focus on quickly occupying the domestic market, and because of certain anti-monopoly regulations, smaller banks may more attractive to acquirers in the same country. And this result is consistent with Hanna and Pilloff (2007). However, both relationships are not statistically significant.
As analyzed before, SHARE which stands for market share in terms of bank's total deposits has great correlation with LOGAS, the results in table 8 show the same relationship as LOGAS with acquired likelihood for domestic and cross border acquisitions. But SHARE is more significant than LOGAS based on statistics. It is consistent with the findings of Moore (1996) and Pasiouras et al. (2007) for domestic acquisitions while similar with Lanine and Vander (2007) for cross border acquisitions.
The annual growth of bank's total assets (GROWTH) has statistically significant relationship with the likelihood of being acquired. It is positively related for domestic acquisitions, which indicates that banks with high growth rate are more attractive to the potential buyers in the same market with the confidence that potential gains can be expected after acquisitions (Cheng et al., 1989; Hernando and Nieto, 2009). And the sign for coefficient is negative for cross border acquisitions. It suggests that M&A deals are more likely to happen for foreign investors when they can see the chance to increase the value and profits of targets which is underperforming due to the various reasons.
In terms of market characteristics, A5 and MGROWTH are statistically significant for the possibility of being acquired. A5 which measures for market concentration is negatively related to the possibility of acquisitions notably. This may due to two reasons. Firstly, if the industry is very concentrated which means the market is dominated by several banks, then large banks will be under pressure of regulators for anti-monopoly reasons if they have incentives to acquire other banks. And small banks may still face strong competitive from large banks even after they go for acquisitions, thus this reduces their incentives for acquisitions. Secondly, UK banking industry is highly developed, and high level concentration may give even smaller business opportunities for other banks which will harmfully affect the likelihood for investors to acquire banks in UK (Wheelock and Wilson, 2004).
MGROWTH has a highly positive relationship with the likelihood of banks being acquired for both domestic and cross border acquisitions. High market growth gives investors the confidence to further expand their business for promising profits, and this is one of the important factors which drive the M&As in banking industry.
Concerning the regulatory factors, the results in table 8 reveal that WKKZ has a significant negative relationship with the possibility of domestic acquisitions while BINDEX is negatively related with the acquisition likelihood in both domestic and cross border acquisitions. WKKZ is the weighted KKZ index which measures the government governance includes government effectiveness, regulatory quality, rule of laws and control of corruption in UK. The coefficient for this variable is negative, and particularly significant for domestic acquisitions which indicates that probability of domestic acquisitions will decrease if the government has different requirements which increase the complex of transactions for M&As such as documentary requirements, legal compliance etc. BINDEX is the index measuring regulatory restrictions based on Barth et al. (2001, 2004). It shows a slightly negative relationship for domestic acquisitions while more significant for cross border ones. This indicates that foreign investors pay more attention to the local regulations in the market where the targets located. Strict regulations for M&As in target market will certainly has a negative influence on foreign investors when they make decisions for cross border acquisitions.
In sum, for the time period from 2005 to 2007 which is before the global financial crisis, bank's characteristics including market share and annual growth rate have significant impact on the likelihood of being acquired for UK banking industry. And market features including concentration and growth rate are also the determinants of acquisition likelihood. Regulatory factors are negatively related to the probability of being acquired, but only significantly for domestic acquisitions.
Results For UK Banking Industry During Crisis
Results For Chinese Banking Industry
Similarly to UK banking industry, table 10 and 11 present the differences for characteristics among within-border acquired banks, cross border acquired banks and non-acquired banks for two specific periods which are before and in the global financial crisis.
For the time period from 2005 to 2007 which is before the crisis, it can be seen that acquired banks have higher cost-to-income ratio than the non-acquired banks, especially for cross-border acquisitions. Moreover, market share for cross-border acquired banks are much higher than non-acquired banks
Cite This Dissertation
To export a reference to this article please select a referencing stye below: