Motives, governance, and long-term performance of mergers and acquisitions in Asia

Abstract This study aims to examine the long-term performance of post-Mergers and Acquisitions (M&A) based on the motive’s category. This study also examines whether the governance at firm-level and country-level affects the relationship between motives and long-term performance of M&A. This study uses a sample of 301 completed M&A transactions of Asian companies in 11 countries from 2002 to 2012 and analyses the data using the cross-sectional moderated regression method. This study measures and categorizes M&A motives into synergy and agency categories using two alternate ways, which first using the stock market reaction data of acquirer and target at the M&A announcement period estimated by applying an event study methodology, and the second using the combination of the stock market reaction data and some accounting variables through logistic regression analysis. This study finds that synergy-motivated M&A results in significantly higher long-term performance post-M&A than agency-motivated M&A. It also proves that the positive effect of M&A motives on the long-term performance post-M&A is significantly higher when companies have better firm-level governance quality, but a similar result cannot be proved for country-level governance. This study provides a cross-country M&A study in the Asian region that has an essential role in global M&A activity in recent years. Considering the possibility of changes in company structure and environment post-M&A, this study contributes to provides an insight related to the role of the governance quality post-M&A as a monitoring tool in the realization of M&A predicted gain that can improve M&A performance.


PUBLIC INTEREST STATEMENT
M&A decisions change company's performance and value. This is one of the most highlighted focuses of this study. This study focuses on one of the basic characteristics in every M&A. Based on the agency theory, there is a possibility that M&A is completed not to maximize the value of the company and shareholders, but rather it is for personal interests. This study aims to examine the impact of these different M&A motives on changes in company performance post-M&A. It explores the role of corporate governance that reduces agency problems in realizing the benefits of M&A in the following period. This study was conducted using cross-country M&A transactions in 11 Asian countries. This study provides evidence to all stakeholders to explore more information on the reasons for each M&A decision made by management as it determines the benefits generated from these M&A decisions. In addition, this study also shows how corporate governance mechanisms maximize the benefits of M&A transactions.

Introduction
Mergers & Acquisition (M&A) has an essential role in the efficient allocation of resources in the economy, causing M&A to attract many years of research interest (Bhabra & Huang, 2013). M&A is an external expansion strategy by acquiring other companies so that the company can proliferate (Alam & Ng, 2014;Akgöbek, 2012). The M&A topic is significant for researchers of finance and strategy, especially the impact of M&A's decisions on long-term company performance and shareholder value. Rao-Nicholson et al. (2016) indicate that there are many research results on the effect of M&A's decision on the company's long-term performance after M&A. Some studies found that M&A can increase long-term performance (Healy et al., 1992;Lau et al., 2008;Rahman & Limmack, 2004), but other studies also found that M&A has an adverse impact to company performance (Mantravadi & Reddy, 2008;Yeh & Hoshino, 2002), while the rest did not find a significant effect of M&A decision to company performance after M&A (Dutta & Jog, 2009;Kumar, 2009).
The grouping of M&A motives can be one of the factors that suggest why not all M&A transactions result in improved financial performance. Every M&A decision taken by management has different primary motives, which cannot be ignored when investigating the impact of M&A decision on company performance (Markelevich, 2003). There are several types of motives that can underlie the M&A decisions made by a company. Piesse et al. (2013) summarize the reason for takeovers such as efficiency, agency, free cash flow, market power, diversification, information, bankruptcy avoidance, and accounting/tax hypothesis. Berkovitch and Narayanan (1993) simplify M&A motives into three major groups, namely: synergy, agency, and hubris. Synergy motives occur when M&A produces a joint company whose value exceeds the value of each target and acquirer company separately. Agency motives are when M&A is carried out because the manager's drive to maximize their interests is a burden or risk on shareholders. Whereas the hubris motives behind M&A occurs because management is too overconfident in evaluating the target company, which results in an overpayment. Berkovitch and Narayanan (1993) classify M&A transactions into these three major groups by looking at the correlation between return for the target company with the total return obtained from the company resulting from the merger at the time of the M&A announcement.
Previous studies examined the effect of M&A motives on the company's long-term performance with different method, sample and results. Markelevich (2003) and Liu et al. (2010) found that M&A with synergy motives significantly produce higher long-term financial performance than M&A with agency motives. Both of them use the M&A motive groupings proposed by Berkovitch and Narayanan (1993) that measured M&A motives based on market reaction at M&A announcement period for acquirer companies, target companies, and a combination of both. Besides using market reactions, Markelevich (2003) and Liu et al. (2010) developed Berkovitch and Narayanan (1993) measurement by adding data from several accounting variables that can represent opportunities for synergy and agency problems found in each M&A transaction. In contrast with those researches, Paulone (2013) and He and Wang (2014) found no significant influence of M&A motives on company performance after M&A. Paulone (2013) used qualitative data to measure M&A motives that are generated from content analysis of management words in company publications related to M&A. It classified M&A motives into 4 groups: synergy, market power, market discipline, and diversification. Similar to the method used by Paulone (2013), He and Wang (2014) also used qualitative information to determine M&A motives group separated into four categories that are synergy motive, acquiring market share, acquiring specific assets and accelerate growth. In their research, He and Wang (2014) proved that on average M&A had a positive effect on longterm shareholder value. Based on the motive category, He and Wang (2014) study proved that M&A with a synergy motive significantly increase company value for shareholders (with a two-year lag) but in the other side, M&A with motives for acquiring market share, acquiring specific assets and motives to accelerate growth do not have a significant impact on the value of shareholders in the long-term performance.
This study aims to re-examine the effect of M&A motives on the company's long-term performance following M&A within the Asian region. Asian countries were chosen as the research sample for several reasons. First, the Asian market is playing an increasingly important role in global M&A activity in recent year. M&A value and number data in the worldwide during 2000-2016 from Institute of Merger, Acquisition, and Alliance (IMAA) shows that even though M&A number and value in Asia is lower than America and Europe, but Asia region has the highest average growth rate of M&A number (6.37% each year) and value (14.15% each year) among the three regions. In 2015, M&A in Asia covered 24% of M&A transaction value in total, and it has the potential to keep growing in the future considering the rapid growth of domestic market demand. The M&A study in Asia particularly related to performance outcomes of M&A decision is pertinent with the current M&A in Asia.
This study likewise aims to include governance quality in examining the relation between M&A motives and long-term performance. According to Shleifer and Vishny (1988), the agency problem is one dominant explanation of the inability M&A to improve post-event performance. Corporate governance is a mechanism that protects the shareholder right by reducing the agency problem in the company. This study includes governance quality variables (both for the firm and country-level) in the following period of M&A that represent the control and monitoring mechanism of the post-M&A management activity, primarily the empowerment of resources acquired from M&A. Some previous studies (Carline et al., 2009;Rani et al., 2013;Yen & Andre, 2007;Yen et al., 2013) proved that corporate governance is effective tools to reduce agency conflict in the company and give the positive effect to operating performance post-M&A.
It is necessary to provide a study-related to M&A motive in the Asia region, which is still relatively limited, and it has inconclusive results to provide a detailed capture of M&A motive and performance in Asian countries and to confirm the previous research results. This study contributes to the literature related to long-term performance after M&A in several ways. First, it examines the impact of M&A motives (synergies and agencies) using cross-country data in the Asia region that includes both developed and developing countries. Second, this study includes the corporate governance variable (at firm and country-level) in examining the relationship between M&A motives and long-term performance after M&A that is not included in previous studies. The research outcomes in the different regions might be due to dissimilarities of the characteristics of each region, so it is necessary to address this M&A study limitation to provide more valid and suitable research results for the Asian country's environment. The remaining of this study will be outlined as follows. Section 2 explains about prior literature and the hypothesis development. Section 3 describes the research method used in this study. Section 4 presents research results and discussion, and Section 5 the last section, conclusion.

Theoretical framework on M&A motives
M&A decisions are often driven by complex set of motive that cannot be summarized easily (Geiger & Schiereck, 2014). Kumar (2009) summarizes that generally, there are two theories underlying M&A motives: (1) the neoclassical theory or value-maximizing theory which assumes that management takes M&A as a reaction of technology or economic environment changes to improve company performance and create positive synergy (Harford, 2005). This theory explains that managers do M&A as their effort to create and maximize the value of the company and shareholders' interest. Value-maximizing theory in M&A explains synergy motivated M&A in this study, which suggests that managers only make an M&A deal if it creates synergy from the combined firm. Value-maximizing explanation can be distinguished into the source of takeover gains such as efficiency theory, monopoly theory, raider theory, and valuation theory (Trautwein, 1990). Based on the efficiency theory, synergy can be generated from operating synergy, financing synergy, and managerial synergy. Operating synergy can be obtained from the efficiency of operational or production activities through the utilization of economies of scale, reduction of duplication processes, or the existence of inter-company knowledge transfer (Rabier, 2017;Trautwein, 1990). Financing synergy is the benefits arise from financial structure combination between acquirer and target that includes access to the internal capital market, lower cost of capital, diversification of cash flow streams, and also potential tax saving (Alhenawi & Krishnaswami, 2015;Rabier, 2017). And the last, managerial synergy is realized when the acquirer has superior management skills and abilities that benefit the target's performance (Geiger & Schiereck, 2014;Trautwein, 1990).
The second theory that underlying M&A motives is the managerial theory or agency theory, which shows mergers as extensions of the manager's interests and creates wealth transfer from acquirer shareholders by acquirer management (Geiger & Schiereck, 2014;Trautwein, 1990). M&A is conducted because of several reason or management objective such as the diversification of management's portfolio (Amihud & Lev, 1981), acquiring assets that enhance the company's dependence on the management (Shleifer & Vishny, 1988) and using of company's free cash flow to increase the company size to fulfil the professional ego of management to exercise and exert power and authority (empire-building theory) (Jensen & Meckling, 1976). The managerial theory gives a view that the objective of M&A deals is not to maximize the shareholder benefit but instead for manager self-interest, so M&A deals do not have the expected additional benefit and value for company and shareholder.
The agency-motivated M&A is caused by the agency problem that arises from the separation of ownership (shareholder) and control (management). The conflict of interest and the asymmetry information between shareholders and management induce the management to take some opportunistic actions that enhance management self-interest but destroy the company and shareholder value (Jensen & Meckling, 1976). Agency-motivated M&A is one of the samples of opportunistic decisions made by management. This study limits the separation of M&A motives into two big categories that are synergy motives, and agency motives using the investor perception about expected synergy benefits of M&A reflected in stock market reaction of company M&A announcement refers to (Barragato & Markelevich, 2008;Berkovitch & Narayanan, 1993;Liu et al., 2010;Markelevich, 2003). This study analyses whether synergy and agency motives in M&A creates different performance significantly in the long-term period.

Hypothesis development
The linkage of M&A motive and the performance outcomes of M&A is necessary and has been proved by some previous studies (Hassan et al., 2018;He & Wang, 2014;Liu et al., 2010;Markelevich, 2003;Paulone, 2013;Rabier, 2017) with various methods and results. Referring to previous studies (Berkovitch & Narayanan, 1993;Fernández & Baixauli, 2003;Kiymaz & Baker, 2008;Seth et al., 2000), this study groups M&A motives based on market reactions at M&A announcements period then examines the effects of M&A motive differences on the company long-term performance. This study limits the analysis of the M&A motive only for two categories that are synergy motive and agency motive. The synergy motive is derived from the valuemaximizing theory that assumes M&A as managers "efforts in maximizing the value of the company and shareholders" interest (Trautwein, 1990). Synergy in M&A means the ability of the combined company to be more profitable than the ability of both acquirer and target individually. Operating synergy can create revenue enhancement and cost reduction while financial synergy can create a possibility to reduce the cost of capital for a combined company (Gaughan, 2007). All those kinds of synergy are expected to be realized in the following period of M&A so that it can enhance the company's performance post-M&A.
While agency motives are applications of managerial theory in M&A, which shows that M&A is an extension of the interests of managers to improve personal welfare like avoiding managerial risk with diversification (Amihud & Lev, 1981) and increasing popularity and prestige over the management of larger entities after M&A (Brealey et al., 2008;Roll, 1986;Trautwein, 1990), there is no additional benefit expected from this kind of M&A that can increase shareholder value. The agency-motivated M&A is caused by the agency problem in the company as a result of the existence of information asymmetry and the conflicting interest between management and shareholders, so management decides something that tends to benefit itself and harm the company or shareholders. Management can use company resources for their consumption, not for the benefit of shareholders (Jensen & Meckling, 1976). Management can have more detailed information about the condition of the acquirer and the target that is unknown to shareholders so that the information asymmetry between management and shareholders allows management to hide the information and make M&A decisions that benefit management privately but harm shareholders.
In the other way, M&A motives studies that used quantitative data in measuring M&A motives (both market reaction data and accounting variable data that represent the expected synergy gain from M&A), Markelevich (2003) and Liu et al. (2010) prove that M&A motives have a significant positive effect on the company's long-term performance after M&A. This suggests that synergymotivated M&A is proven to have higher long-term performance compared to agency-motivated M&A. Those findings support the hypothesis of the relationship between the M&A motives to the long-term performance outcomes of M&A. Synergy-motivated M&A will have better performance along with the realization of synergy or efficiency benefits that arise from the combined firm. Whereas in agency-motivated M&A, the increase in company performance produced after M&A will be lower, none or even decrease due to lack of efficiency or synergy that can increase the company's added value.
Based on the description above, the hypothesis can be proposed as follows: H1: The M&A motive has a positive effect on long-term financial performance post-M&A.
The next purpose of this study is to examine the ability of corporate governance variables to moderate the influence of M&A motives variables with long-term performance. Some previous studies (Carline et al., 2009;Rani et al., 2013;Yen & Andre, 2007;Yen et al., 2013) have examined the direct effect of corporate governance at firm-level o the company performance after M&A. Using corporate mergers sample in the UK during 1985-1994, Carline et al. (2009) prove a curvilinear and multidimensional effect of board ownership, the negative impact of larger boards, and the positive effect of outside block-holders on the operating performance changes of M&A. Yen and Andre (2007) examined the impact of ownership concentration and separation of ownership and voting rights to the long-term performance of M&A using M&A sample in some Englishorigin countries and found that ownership concentration has a non-linear relationship to the M&A performance while the separation of ownership and voting rights leads the lower performance. Yen et al. (2013) did similar studies with Yen and Andre (2007) using M&A sample in emerging economies countries and concluded that the presence of professional outside directors has a positive impact, ownership concentration has a non-linear relationship and audit committee has a negative impact on the long-term performance of M&A. Rani et al. (2013) examined the effect of firm-level corporate governance on the company performance of M&A with short term analysis and proved that companies with higher rank for corporate governance score have better shortterm performance companies derived from M&A transaction.
There were no previous studies that examined the moderating effect of corporate governance quality (both for firm-level and country-level) to the long-term performance after M&A. One related study is conducted by Craninckx and Huyghebaert (2015) that examined the moderating effect of the larger shareholders as corporate governance proxy to the negative impact of agency attitudes and management overconfidence in making M&A decisions on owner welfare (stock returns). The results proved that large institutional shareholders effectively limit the adverse effects of management's overconfidence when conducting M&A on returns obtained by owners. It explained that corporate governance should influence the relationship between M&A motives on company performance after M&A. This study has a premise about the ability of corporate governance quality to influence the positive effect of M&A motive on the long-term performance after M&A. Performance improvement that is created by synergy-motivated M&A is suspect to be higher when it is supported by the implementation of better governance practices. By implementing good governance, management will be monitored more closely so that management will be more careful in every action post-M&A, including in managing assets that have just been acquired through M&A. Although management initially carried out M&A with agency motives with the close supervision could force management to maximize acquired assets management more carefully and efficiently so that the negative impacts of agency motive on the company's long-term performance post-M&A can be minimized.
Based on the description, the hypothesis can be arranged as follows: H2a: Corporate-level governance practices strengthen the positive effect of M&A motive on the long-term financial performance post-M&A. Bhagat et al. (2011) explain that there are two dimensions of governance, one is companyspecific, and the other is country-specific. Country-level governance includes the existence of regulations that are made related to corporate governance obligations, which are then practiced, the level of law enforcement, culture, and conformity of accounting principles used with applicable international accounting standards. Yen et al. (2013) have tested the effect of differing levels of country-specific legal protection on M&A value creation in some emerging countries. Using some country-level variable such as legal origin, public enforcement index, anti-director rights index, anti-self-dealing index and extra-legal enforcement index, Yen et al. (2013) proved that companies in stronger investor protection countries (anti-director rights index) can enhance performance post-M&A while public enforcement index is proved to have negative impact to the performance in emerging countries sample. In the other study, Yen and Andre (2007) also proved that greater investor protection has a positive impact on the operating performance of M&A in English-origin countries.
Besides, in company-level scope, the improvement in governance quality at the country-level will give the protection of shareholders' rights from agency problems related to M&A transactions that can result in performance improvement. When M&A is carried out on companies in a county with better protection of shareholders' rights, monitoring of company management activities will increase and more effective so that it can reduce the problems of company agencies and ultimately improve company performance after M&A (Thenmozhi & Narayanan, 2016). Based on that explanation, in the analysis of country-level governance role to the relation of M&A motives and long term performance, this study supposes that high quality of governance in the countrylevel can maximize the positive effect of synergy-motivated M&A in the long-term performance because of the effective monitoring in the M&A expected gain realization. And from the agencymotivated M&A view, high quality in the country-level governance is expected to reduce the negative effect of agency-motivated M&A to the long-term performance following M&A because it can limit the opportunistic behavior of management in the allocation of resource acquired from M&A. Based on the description, the hypothesis can be arranged as follows: H2b: Country-level governance practices strengthen the positive effect of M&A motive on the long-term financial performance of acquirer companies post-M&A.

Sample and data collection
The sample of this study is M&A transactions announced and effective in 11 Asian countries covering China, Hong Kong, India, Indonesia, Japan, South Korea, Malaysia, Philippines, Singapore, Thailand, and Taiwan from 2002 to 2012. 11 sample countries were selected based on the availability of score data country-level corporate governance issued every two years (starting in 2003 as a result of a 2002 survey) by the Asian Corporate Governance Association (ACGA). Additional sample criteria are as follows: (1) M&A must resulting in the acquisition of control over the target, which is characterized by the percentage of ownership before M&A less than 50%, then becomes above 50% after M&A. This condition is needed to capture the performance of the combined firm in consolidation financial report when the acquirer has control of the target; (2) both acquirers and target companies are publicly listed company for at least three years before M&A, and afterward, the merged firm must continue to list for a minimum three years.
This condition is needed because this study uses market value data in M&A motive and performance variable measurement, and because this study needs to capturing the long-term performance changes from M&A transaction, (3) both acquirer and target companies are not in financial industry because it has specific accounting and regulatory requirement, (4) M&A sample is non-overlapping M&A, i.e., the acquirer companies acquired other company in zero years only in a window of (−3 to +3) year to measure the performance resulted from one M&A transaction accurately, (5) M&A transactions also must have complete data such as stock price data, financial data, annual reports and other data needed in calculation of research variables. Long-term performance is measured by subtracting performance in three years following M&A with performance three years preceding M&A so that the financial data collected covers 7-year data for each one observation or covers data from 1999 until 2015.

M&A motives
In this study, M&A motives variable is measured by two approaches, that are: 1) based on market reactions that represent investors' perceptions (Barragato & Markelevich, 2008;Berkovitch & Narayanan, 1993;Markelevich, 2003;Seth, Song, & Pettit, 2000), and 2) based on the combination of accounting variables related to the characteristics of M&A transactions with market reactions (Liu et al., 2010;Markelevich, 2003). The first measurement of M&A motives is done by capturing market reactions reflected in the acquirer and target company stock price during the M&A announcement period. M&A motive is represented by the Cumulative Abnormal Return (CAR) value generated by the acquirer and the target around the date of the M&A announcement (a five-day window started from two days before until two days after M&A announcement). The abnormal return is calculated by the prediction error (the difference between actual return and predicted return) of the market model estimated using stock price data for 200 trading days' period starting from the 31st day before the occurrence of M&A. Barragato and Markelevich (2008) explained that the acquirer (target) CAR is interpreted as representing the stock market's estimates of whether M&A conducted by the company is in the interest of the acquirer (target) shareholder. Total CAR is the combined CAR of the acquirer and the target, weighted by their market value on the sixth day before M&A announcement.
M&A motives are divided into the following (Berkovitch & Narayanan, 1993): As shown in Table 1, this study classifies M&A transaction into the motives category where the M&A transaction in which both the acquirer CAR and the total CAR are positive implies a synergy motive since the M&A is in line with the acquiring shareholder's interest and creates value. While the M&A transactions which have negative the acquirer's CAR and total CAR are classified as agency motive M&A since the stock market's reaction to the M&A announcement suggests a diminution in value and is not in line with the interest of acquirer's shareholders. M&A transaction with acquirer CAR negative and total CAR positive represent hubris motive, where management overpays for target because of overconfidence or excessively optimistic synergy estimation resulting in negative shareholder value.
The second measurement of M&A motives is adopted from Markelevich (2003), and Liu et al. (2010), which combined the first measurement that only uses market reaction data with some accounting variables considered related to M&A motives. With this measurement, research samples are divided equally and randomly into two groups. The first group is used to form a prediction function for the M&A motives category using discriminant or logistic regression (depend on data normality). The prediction function is developed using accounting variables in Table 2 to predict the M&A motives category (1 for synergy motives, and 0 for agency motives). The prediction of probability resulted then will be used as M&A motives proxy, where higher value means that M&A is in shareholder interest and higher expected synergy benefit, and a lower value means that M&A is not in shareholder interest and represent agency motives.

Long-term performance
Long-term performance is measured using the operating cash flow return (OCFR) (Carline et al., 2009;Healy et al., 1992;Markelevich, 2003;Rao-Nicholson et al., 2016). The definition of operating cash flow here refers to Healy et al. (1992), which is operating income before being reduced by depreciation expenses and non-recurring items. To avoid the impact caused by the different methods of recording M&A transactions (purchase method or pooling of interest method) on the company's book value, then company performance is calculated by operating cash flow deflated by the company's sales. Operating Cash Flow Return (OCFR) was calculated for six periods covering three years before M&A and three years after M&A. The OCFR in the period before M&A is calculated on a weighted average based on the market value assets of the acquirer and the target company in the year before the M&A whereas OCFR in the period after M&A is calculated from companies that have joined. OCFR calculations are adjusted to a median industrial value, which has the same two-digit SIC code in the same year and same country. The median value is used as a substitute for the average value as an industry benchmark because the composition of the companies of each industry varies in each year and country; the use of medians could avoid the effect of existing outlier values. The long-term performance (LTP) variable calculated as the difference between average OCFR before M&A and after M&A. Source: Barragato and Markelevich (2008, p.307)

Corporate governance
The Corporate governance (CG) variable in this study covers firm-level and country level. The measurement of governance in country-level (CGCOUNTRY) is calculated by the average of each country CG score issued by the Asian Corporate Governance Association (ACGA) for three years following M&A. The measurement of firm-level corporate governance (CGFIRM) was adopted from Cao et al. (2015), which used the loading factor value resulted from factor analysis of several indicators that are often used in corporate governance studies. The use of composite index values from several different indicators has the advantage of being able to measure a variable more comprehensively (Cao et al., 2015;Larcker et al., 2007;Shan, 2015). This study uses several CG indicators from Cao et al. (2015) as shown in Table 3. The CGFIRM variable is calculated each year during three years after M&A.

Control variables
The controlling variables in this study consists of company-level variables such as company size (SIZE), leverage level (LEV) and cross border M&A (DCBMA) categories, and country levels control variables such as acquirer domicile country category (DDEVAC) and target domicile country (DDEVTR). SIZE variable is measured by using the natural logarithm of the total assets of the  (2003)

Research model
The objectives of this study are to examine the effect of M&A motives on the long-term performance post-M&A, and how the effect of corporate governance (firm-level and country-level) on it. This study uses the ordinary least square (OLS) regression method with cross-sectional data to test the magnitude of the influence between M&A motives and the long-term performance post-M&A. The regression model used in this study includes the moderating variable (moderated regression analysis) and tests the moderating hypothesis (H2a and H2b) by includes the interaction variable (between M&A motive as independent variable and governance as moderating variable) into the model. This study estimates the following model: Hypothesis 1 is proven through α 1 which shows the effect of M&A motives on the long-term performance and it is expected to have positive and significant value. Hypothesis 2 related to the moderating effect of corporate governance are proven through the value of α 3 for firm-level corporate governance (H2a) and α 5 values for country-level governance (H2b). The value of α3 (α5) is expected to be positive and significant so its mean that better governance practice implementation in the company (in the country) will strengthen the positive influence of M&A motive to long-term performance.
The elimination of data outliers is done for data that have values more than the average ± three times the standard deviation. Testing the classical assumption was performed using Jarque-Beta for normality, white testing for heteroscedasticity, and Variance Inflation Factor (VIF) value for multicollinearity.

Descriptive statistics
Based on the sampling and data collection process, this study obtains 401 M&A transactions that had complete market data needed. The following in Table 4 shows grouping the M&A transactions based on the M&A motives criteria expressed by (Berkovitch & Narayanan, 1993). Table 4 shows that the majority (51.87%) of the total sample M&A transactions in this study are included in the synergy motives, while agency motives are 35.16% of the sample M&A transactions. Hubris's motives are also proven to occur in 9.23% of the M&A transactions in this study. The results of M&A motives grouping support the results of previous studies (Berkovitch & Narayanan, 1993;Mukherjee et al., 2004;Porter & Singh, 2010;Seth et al., 2000) which show that M&A transactions are carried out predominantly by synergy motives. In the next analysis, this study only uses M&A transactions with synergy and agency motives as a research sample, so totally, it has 349 M&A in the beginning. Then, some samples must be eliminated because of do not have enough financial data, or it has extreme value. It results in 301 M&A transactions as the final sample. Table 5 indicates that almost all cumulative abnormal return (CAR) values generated by the acquirer, target, or total both of them around the date of the M&A announcement have a positive value that is significantly different from 0. It means that there is a significant reaction from investor related to the M&A announcement from the company that was captured through the cumulative abnormal return value, which will be used as a proxy for the M&A motives. The insignificant results are only obtained in the CAR acquirer with an 11 day time window (−5.5), which indicates that acquirer investors respond to the M&A information quickly (less than eleven days). However, this condition is slightly different for the target company's investors who still have a significant reaction in a longer time.
Next, Table 6 shows M&A samples description from some characteristics. In panel A, it is seen that the M&A samples are dominated by companies in Japan (65.45%) then followed by India and South Korea that each covers 7.31% of the total M&A samples. The highest average long-term    Table 7 presents the descriptive statistics of the research variables that will be used in the analysis of the research model.

Developing of M&A motive equation
As explained before, one of the M&A motive measurements used in this study is by combining market reactions and related accounting variables that represent the possibility of synergy and agency motives in M&A decision. The combination is done by using accounting variables as estimators of the M&A motives group (synergy or agency). To develop the estimation equation, this study uses a half of total 338 M&A data (349 data M&A transactions which include in synergy and agency category minus 11 M&A that do not have complete accounting variable data).   (2019) shows descriptive statistics of accounting variables used in the M&A motives estimation equation for 169 M&A transactions. Table 8 shows that there are several variables such as TRROA, DRELATED, ACFCFINV, RELSIZE and DPAYMENT which have abnormal data distribution. Pohar et al. (2004) have conducted a study that compared the use of discriminant analysis methods and logistic regression in forming categorical variable classification models. According to Pohar et al. (2004) study, discriminant analysis is better used if all assumptions are fulfilled, whereas if it does not fulfill the assumption, logistic regression can provide adequate results.
Based on these conditions, this study uses a logistic regression analysis tool to form an estimator equation for M&A motives group. The results of the logistic regression analysis from this sample group are described in Table 9. Table 9 shows the logistic regression results, which have 74.6% accuracy of prediction. Some variables are proven have significant effect to M&A motives category. First, TRROA, it positively affected the M&A motives category. It shows that M&A of high profitability targets result more benefits and add values to companies than targets with low profitability caused by the operational efficiency of the high profitability target can complete and reduce the operational inefficiency in acquirer company. Second, DRELATED also has a significant positive effect that means M&A carried out between companies in similar business line is more valuable for companies and shareholders than M&A between non-similar industries. This is consistent with the synergy theory that can be  (2019) generated by resource sharing, technology integration, elimination of duplicate processes, or economies of scale benefits resulted from M&A between companies in same industry. Positive effect of TRCASH indicates that the cash amount of target company can be used as an additional funding for acquirer company. DLEV and ABSDLEV represent a financial or funding synergy that can be obtained when M&A is carried out. Both variables have a significant positive coefficient result which means that when the acquirer and target leverage difference is higher, the more likely the benefits of financial synergy generated from M&A are made. ACDEBTOP that represents the level of debt owned by the acquirer company in the period before M&A was carried out, has significant negative effect, which indicates that when the acquirer debt level is higher, it will get a negative response from acquirer and assume that M&A is carried out with agency motives. Next, the comparison of the size of the target company with the acquirer (RELATIVESIZE) into one significant factor can be used to detect M&A motives. The larger size of the target company purchased by acquirer illustrates the greater the synergy benefits that can be generated. And the last, DPAYMENT has a significant positive effect, which means that when M&A payments are settled with cash, investors respond positively and represent the synergy motives in management M&A decision. Table 9 is used to estimate the DMOTIVE variable in the second sample group (test group) based on the value of the accounting variables of each transaction. The logistic regression equation produces an estimate of the probability value of a DMOTIVE variable of 1 for each transaction sample in the second group, which then becomes the alternate proxy of the MOTIVE variable value in testing the research model.

Univariate analysis
The first hypothesis was tested using univariate analysis by testing the difference in the performance of M&A between synergy-motivated M&A and agency-motivated M&A. Table 10 shows the average value of company performance during the first year until the third year after M&A for all samples and each motives category (synergy and agency). The M&A motives categories in Table 10 are separated based on the classification of the acquirer and the total abnormal market return. The benefits or added values generated from M&A require time to be realized, so assessing M&A performance will be more appropriate in long-term analysis than short-term analysis. The average LTP value of the synergy group statistically significantly higher than the LTP in the agency group, which provides an early indication of the empirical evidence that supports hypothesis 1 (H1) that states synergy motives create higher long-term performance post M&A than agency motives.

Multivariate cross-sectional regression analysis
The results of regression analysis are presented in Table 11. Panel A presents the results with the market reaction as M&A motives proxy, while Panel B is the regression results with using M&A motive predicted based on accounting variables using logistic equation. Table 11 shows that the coefficients of MOTIVE variable are positive and significant in all panel and model (Model 1 to 6). It concludes that hypothesis H1 is accepted, which means that the M&A motive significantly influences the long-term performance of M&A companies in a positive     (2) and (3), and using the expected probability resulted from the logistic regression model in Equation (5) and (6) Source: Authors' own calculation (2019).
direction. It means that increase in the MOTIVE variable (represent higher synergy or lower agency motive) significantly creates improvement in the long-term performance after M&A. Although this study focuses more on continuous measurement in a MOTIVE variable, in Table 11 also included model testing if dummy measurement is used. And both using dummy or continuous scale, the results consistently show that M&A motive has a positive impact on long-term performance after M&A. This finding is consistent with the research findings of Liu et al. (2010) in China and Markelevich (2003) in the USA that also proved the higher long-term performance resulted from synergy-motivated M&A than agency-motivated M&A. This study result proves the application of agency theory that addresses the conflict of interest between management and shareholders, especially in M&A transactions. In each M&A decision, there is a risk of management's hidden intentions, which may be not by the intention of the shareholders. The conflict of interests and the information asymmetry between management and shareholders induce management to do some opportunistic actions such as M&A that do not maximize shareholder profits (agency motive).
Although with different motives categories, He and Wang (2014), that categorized M&A motives based on qualitative data, supported the findings that synergy-motivated M&A creates higher value for shareholders than other M&A motives (acquiring market share, acquiring specific assets and accelerate growth motives). Rabier (2017) also proved that the operating synergies-motivated acquisitions have higher returns than the financial synergy-motivated M&A. All those studies conclude the same result that the difference in M&A motives creates different performances or return resulted from M&A transactions. On the other side, this study was inconsistent by having insignificant influence of M&A motive on performance post-M&A as was concluded by Paulone (2013). The inconsistent result might be caused by the less number of the sample used by Paulone (2013) that only consisted of 35 M&A transactions in the United States.
Table 11 also shows that the MOTIVExCGFIRM coefficient in Model 3 has a positive and significant value. It means that hypothesis H2a is accepted. This finding suggests that firm-level governance significantly strengthens the positive influence of M&A motive on the long-term performance post-M&A. In a higher quality of governance company, synergy-motivated M&A significantly creates higher post-M&A long-term performance than the company in a lower quality of governance. This result proves the application of agency theory, conflicts of interest arising from the separation between owners and controllers can be suppressed by monitoring functions resulting from the application of corporate governance (Jensen & Meckling, 1976). With increasingly rigorous monitoring after M&A, it induced management to be more efficient and optimal in the utilization of resources obtained from the target companies, so it results in higher performance. To be considered, the significant result of the MOTIVExCGFIRM interaction variable is not proven in Model 4 when the accounting variables were included in the analysis and only used 87 samples of M&A. Although there was no similarities with the previous studies, the finding related to moderating effect of firm-level governance in this study supports the results of Craninckx and Huyghebaert (2015) that proved the ability of large shareholders (as corporate governance proxy) to mitigate the negative effect of management's overconfidence in M&A decision on shareholder welfare. Both Craninckx and Huyghebaert (2015) and this study proved that firm-level governance mechanisms effective in maximizing the performance of M&A through better monitoring and controlling of management activities so it can reduce the opportunistic action of management. Table 11 shows that the coefficients of MOTIVExCGCOUNTRY both in Models 3 and 6 are negative and insignificant value. It indicates that country-level governance does not have a significant moderating influence on the relationship between M&A motives and long-term performance post-M&A, so H2b is rejected. This finding shows that country-level governance is not an effective tool in reducing agency conflicts in M&A decision and do not has a moderating effect on the relation of M&A motives and long-term performance. In Model 3, this study also proves the negative effect of country-level governance on the long-term performance of M&A directly by the value of the CGCOUNTRY coefficient, which is negative and significant (−0.0016**), this result is inconsistent when compared to result in Model 6. With fewer sample and using the different measurement of M&A motive, the coefficient of CGCOUNTRY is positive and insignificant, while the coefficients of MOTIVExCGCOUNTRY is negative and insignificant, like results in Model 3. From the results in Model 3, it shows that country-level governance only has the primary role as an independent variable that directly influences the performance, but it does not have significant additional role as moderating variable. Focus on main objectives in this study, from the results in both Models 3 and 6, it concludes that there is no significant difference in the relation between M&A motive and the long-term performance between M&A conducted by acquirer domicile in the high and low country-level governance quality. Because of the limitation of similar previous study, the comparison of the moderating results (H2a and H2b) cannot be performed perfectly. Considering the close previous study about the relation of country-level governance of acquirer in M&A performance, this study might be related and consistent with the results of Thenmozhi and Narayanan (2016) that proved country-level governance of acquirer companies were not dominant factor in determining post-M&A performance for both of cross border M&A between developed acquirer-emerging target and cross border M&A between emerging acquirer-developed target in separate analysis. But in addition, Thenmozhi and Narayanan (2016) also showed that the dominant factor in determining post-M&A performance for cross-border M&A between an emerging acquirer and developed target is target country-level governance. It can be explained that the governance quality of target's country can give more significant effect in the performance of M&A than the governance quality of acquirer's country, while this study focuses only in country-level governance quality of acquirer.
The other explanation of the insignificant moderating effect of country-level governance quality might be due to the lack of variation in the country-level governance scores of each country and each observation year in this study. The score of country-level governance in this study is ranged between 39 and 67 with an average in 55.05, and the research sample is dominated by companies in Japan Country (65,45%). So the difference in country-level governance between the sample is not significant enough to create different impact to the long-term performance.

Conclusion
Based on the analysis of 301 M&A transactions that occurred in eleven Asian countries from 2002-2012, this study proves that the M&A motive significantly affected long-term performance after M&A. M&A transaction with synergy motive produces higher long-term performance than M&A with agency motive. It is caused by the realization of added value or the benefits obtained from synergy-motivated M&A that is not found in agency-motivated M&A. Next, by using moderated regression analysis, this study also proves that firm-level governance implementation strengthens the positive relationship between M&A motives and long-term performance after M&A. A company that conducts synergy-motivated M&A implements better corporate governance mechanisms creates a higher long-term performance level than before. This result proves the implementation of agency theory in M&A decision. There is a risk for a shareholder that management takes M&A decision strategy with the hidden reason that gives a negative value for shareholders (agency motive).
A different result is concluded for country-level governance quality variable that shows the insignificant of moderating effect on the relationship between M&A motives and long-term performance post-M&A. That insignificant result interprets that country-level governance mechanism is less powerful than firm-level governance mechanisms as the determinant factor of M&A performance especially in the Asian countries that were sampled in this study. This study has several implications especially for Asian countries that were sampled in this study. First, this study reveals that the M&A motive is one of the main factors to predict and determine the benefit resulted from M&A transactions in Asian countries. Therefore, shareholders must build a critical and cautious attitude when management makes the decision to conduct M&A. Shareholders must give more concern and effort to get information about the true motives behind management decision for conducting M&A because the motive determines the benefits that shareholders can obtain in the future. Second, the results suggest that the market response at M&A announcement date can be used as an "early warning" of M&A motives that shows the presence of added value that will be generated from the M&A conducted. As is generally known, it is difficult for the shareholder to detect what is the true reason that management has in each M&A decision especially in the Asian developing environment that commonly has lower technology, information system, governance, and legal quality. This study shows that stock market reaction data and related accounting variables can be used to detect the ex-ante M&A motive that can be used to predict the long-term performance of M&A. When the market response is mostly negative, shareholders can better supervise activity post-M&A by improving corporate governance practices in the company, which is proven to have the ability to reduce the negative impact of management's opportunistic actions in M&A decisions. Third, for the regulator, this study reveals that country-level governance in Asian countries' sample of this study is still less effective in influencing the M&A performance outcomes. It can be input for regulators in Asian countries to develop and implement a better governance system than before. Lastly, considering the possibility of changes in company structure and environment post-M&A, this study contributes to provide an insight to the companies who intend to conduct M&A activities by relating to the role of the governance quality post-M&A as a monitoring tool in the realization of M&A predicted gain that can improve M&A performance.
There are some limitations of this study; first, this study only uses accounting-based performance in three years' observation period to measure long-term performance variable. This study cannot provide other aspects of company performance such as stock market performance, market power or productivity that might be the focus for some companies or investors. This study cannot capture the gain or loss that creates from M&A that is just realized in longer periods than three years in the measurement of the long-term performance. Further research can consider other aspects of performance or can extend the period of performance measurement so that it is expected to better capture the level of success of M&A done by the company. Next, the sample in this study only M&A involving public target firms, so there are limitations that there are differences in measurements and the generalization of this research results on M&A involving the target companies that have private status.