Emerging Markets Review 19 (2014) 18–48

Contents lists available at ScienceDirect

Emerging Markets Review

j ourna l homepage : www.e lsev ie r .com/ locate /emr

Idiosyncratic volatility and mergers andacquisitions in emerging markets

PengCheng Zhu a,1, Vijay Jog b,2, Isaac Otchere b,⁎a School of Business Administration, University of San Diego, 5998 Alcalá Park, San Diego, CA 92110, USAb Sprott School of Business, Carleton University, 1125 Colonel By Dr, Ottawa, Ontario K1S 5B6, Canada

a r t i c l e i n f o

⁎ Corresponding author. Tel.: +1 613 520 2600x2E-mail addresses: pzhu@sandiego.edu (P. Zhu), v

1 Tel.: +1 619 260 2382.2 Tel.: +1 613 520 2600x2377.

http://dx.doi.org/10.1016/j.ememar.2014.04.0011566-0141/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

Article history:Received 20 January 2014Received in revised form 23 March 2014Accepted 3 April 2014Available online 13 April 2014

Given the recent findings in the literature that idiosyncratic vola-tility reflects stock price informativeness, we analyze the impact ofidiosyncratic volatility on many acquisition parameters. We find thatidiosyncratic volatility is positively related to acquisition premium; therelationship is more significant in deals that occurred in information-poor economies where acquirers have difficulty gathering informationabout the targets. These deals typically involve bidders from emergingmarkets and those that have less experience in the target country.Idiosyncratic volatility is also positively related to acquisition comple-tion rate, the likelihood of the bidder acquiring majority control, but isnegatively related to takeover probability.

© 2014 Elsevier B.V. All rights reserved.

JEL classification:D82G15G34

Keywords:Idiosyncratic volatilityMergers and acquisitionsEmerging markets

1. Introduction

The role of idiosyncratic volatility has received considerable attention in the literature in terms of its effecton market efficiency and stock price in both developed and emerging markets with conflicting conclusionsespecially in relation to emerging markets. For example, beginning with Morck et al. (2000), studies haveconcluded that higher idiosyncratic volatility is a reflection of better stock price informativeness andefficiency of the market in incorporating private information into the stock prices (for example, Chen et al.,2007; Durnev et al., 2003, 2004; Ferreira and Laux, 2007; Jin and Myers, 2006; Wurgler, 2000). Theseresearchers claim that through informed trading, private information (i.e., firm specific information) isimpounded into stock prices thereby ensuring that the stock price better reflects the fundamental value of the

731; fax: +1 613 520 2247.jog@ccs.carleton.ca (V. Jog), isaac.otchere@carleton.ca (I. Otchere).

19P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

company. However, several other authors argue that idiosyncratic volatility is merely the consequence ofinvestor sentiments and that the phenomenon reflects market inefficiency (Jiang et al., 2009; Rajgopal andVenkatachalam, 2011).

Many of these authors arrive at their conclusions by focusing on country level institutional environmentand the variation in stock price synchronicity in different markets using data on country market indices. Inthis paper, we take a different approach and examine the effects of firm level idiosyncratic volatility inmergers and acquisitions (M&A). We argue that if idiosyncratic volatility reflects the firm's stock priceinformativeness (or noise), then it would play a significant role in various M&A parameters since privateinformation about the target firm is critical to the managers of bidding firms when making acquisitiondecisions. Accordingly, the central research question we explore in this study is whether idiosyncraticvolatility is perceived as valuable information or noise by “corporate” investors — acquiring firms.

Emerging markets provide an interesting setting for this research since it has been documented thatthese markets exhibit higher stock price synchronicity and a very distinct institutional environmentcompared to that observed in developed markets. Emerging countries are characterized by complexownership structures such as concentrated ownership structure, cross-holdings, business groups andpyramidal ownership structure; therefore the controlling owners could extract private benefits of controlat the expenses of outside investors (Bertrand et al., 2002; Johnson et al., 2000). Under such conditions,owners of firms have incentives to withhold and/or selectively disclose value-relevant, privateinformation to outside investors in order to conceal the valuation implication of their self-servingbehaviors (Fan andWong, 2005; Kim and Yi, 2006). As a result, the cost of acquiring private information islikely to be higher and the profitability of informed trading is likely to be lower in emerging markets ascompared to developed markets. This suggests that stock prices will be less informative in emergingmarkets due to the lack of informed trading of private information by arbitragers.

The study is also motivated by three additional observations. First, prior studies primarily focus on theempirical connections between idiosyncratic volatility and various information efficiency measures (forexample, Durnev et al., 2004; Morck et al., 2000). These authors argue that private information can beimpounded into stock prices through informed trading by arbitragers. However, there is no empiricalevidence that shows whether informed traders such as corporate investors actually value and take actionsto use this private information in their investment decisions. We argue that one of the informed tradingchannels is mergers and acquisitions where informed traders have to act on this information. Acquiringfirms' managers have stronger motivation for collecting target firm specific information before making anM&A decision in emerging markets. Second, many studies have been conducted at the country level andthe argument has been that the lack of legal protection of informed trading can affect informationalefficiency, resulting in negative correlation between stock return synchronicity and property rightprotection and economic development of a country. However, until recently, none of the studies focusedon cross sectional variations in idiosyncratic volatility among firms. Huang et al. (2013) examine theimpact of financial market liberalization on the pricing of idiosyncratic risk in emerging markets andNartea et al. (2011) studied the impact of idiosyncratic volatility on stock returns in the Southeast Asianmarkets. Our study complements these recent studies by providing results on the phenomenon using firmlevel granularity focusing on the information role of idiosyncratic volatility in the corporate mergers andacquisition decisions. Third, we extend the literature on idiosyncratic volatility in decision making to“corporate” investors such as acquiring firms by testing whether acquirers value idiosyncratic volatility oftarget firms and incorporate it into acquisition decisions.

Our key results can be summarized as follows. Using acquisition data from twenty emerging countriesover a seventeen year period, we find that idiosyncratic volatility of target firms plays an important role inmergers and acquisitions in emerging countries. Consistent with the private information and informedtrading explanation of idiosyncratic volatility, we find a positive relationship (a high correlation) betweentarget firms' idiosyncratic volatility and merger premiums. The relationship is significant mostly inacquisitions made by acquiring firms that face information disadvantage in collecting firm-specificinformation on the target firm. Our results suggest that stock price informativeness is more valuable toacquiring firms that are subject to higher information constraints. We further show that target firms'idiosyncratic volatility is also positively correlated with acquisition completion rate and negativelycorrelated with the time taken to consummate the deal. We argue that transparency in the target firm'sstock price, which results from informed trading of private information by market participants and

20 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

reflected in the target firm's idiosyncratic volatility, helps facilitate the acquisition process.3 It may alsoencourage the acquiring firm to make quick decision to prevent competing bidders from exploiting theprivate information source in the target firm.4

Next, after controlling for the target's country, firm and deal characteristics, we find that bidders aremore likely to acquire majority control in the target firm when the target firm's idiosyncratic volatility ishigher. This observation is consistent with the assertion that higher stock price informativeness (asreflected in higher idiosyncratic volatility) provides more impetus for the bidder to gain a controllinginterest in the target firm. In addition, we investigate the impact of idiosyncratic volatility on acquisitionprobability and find that idiosyncratic volatility is inversely related to the likelihood of a companyreceiving a bid. This result supports the argument that informed traders may have lower incentive tocollect private information when the target stock price is more informative. It might reduce the incentivefor investors to acquire additional private information beyond what is already reflected in the target stockprice. In summary, our results indicate that firms with higher idiosyncratic volatility in emerging marketsare less likely to be taken over, but once bidders decide to take over such firms, they pay high premiums.Acquirers that are likely to do this are those that are informationally disadvantaged (i.e., bidders fromemerging markets and bidders who have less experience in the target country).

Our study makes some unique contributions to the literature. First, to the best of our knowledge, this isthe first study that examines the role of idiosyncratic volatility in mergers and acquisitions decisions inemerging countries using firm level data. Given the recent findings in the literature on idiosyncraticvolatility, our study provides an alternative view of idiosyncratic volatility as a measure of stock priceinformativeness and firm specific information in the context of mergers and acquisitions. Second, firmlevel idiosyncratic volatility in emerging markets is a relatively under researched area where the focus hasbeen on a single market (e.g. China and Chile) while examining the role of private information andgovernance and their relationship with idiosyncratic volatility (Gul et al., 2010; Khanna and Thomas,2009). By focusing on idiosyncratic volatility in many emerging countries we are able to provide moregeneralizable results. Finally, many M&A studies focus mostly on acquiring firm characteristics. Relativelyfewer studies examine the impact of idiosyncratic risk on target firms' characteristics in mergers andacquisitions (Chatterjee et al., 2012). We extend the literature on emerging markets by examining theimpact of idiosyncratic volatility on the M&A process in these countries. In addition, we conduct manyrobustness tests (including analysis of acquired firms and non-acquired firms along different dimensions,as well as analyzing the firms that withdrew their acquisition proposals) to provide a comprehensiveanalysis of the effect of idiosyncratic volatility on acquisition decisions in emerging markets.

The rest of the paper is organized as follows. In the next section, we review the related research onidiosyncratic volatility, focusing mostly on mergers and acquisitions in emerging markets. We develop ourhypotheses in Section 3. Our data sources, sample, hypotheses and the statistical methods employed forthe empirical analyses are described in Section 4. In Section 5, we present our main results along with theresults from our robustness tests. We conclude the paper in Section 6.

2. Literature review

Since our paper deals with three sub-areas of the literature on idiosyncratic volatility, we group ourreview into three sections. First we review the literature on the effects of idiosyncratic volatility on stockprice informativeness, and this is followed by a review of the evidence on idiosyncratic volatility inemerging markets. Finally we review of the literature on the relationship between idiosyncratic volatilityand mergers and acquisitions.

3 If the stock price of the target firm reflects private information and the stock is tracking its fundamental value, then negotiationscould be facilitated. However, if the stock price is not reflecting the true information about the company (as reflected in lowidiosyncratic volatility), the acquiring firm and the target firm may have to spend a lot of time negotiating the right offer price.

4 For example, if the acquiring firm has some strategic reason for the takeover, and once it has acquired sufficient amount ofprivate information the acquiring firm could proceed quickly to consummate the deal, and it could even pay higher acquisitionpremium to avoid competition from other bidders, thereby reducing the time to complete the deal.

21P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

2.1. Idiosyncratic volatility and stock price informativeness

The importance of idiosyncratic volatility in corporate decision and governance at the country level hasreceived considerable attention in recent years. Morck et al. (2000) find that countries with betterproperty rights protection tend to have lower stock return synchronicity (i.e., higher idiosyncraticvolatility). They argue that strong property rights promote informed arbitrage trading, which ultimatelyleads to more firm-specific information being impounded into stock prices and thus leading to highermarket efficiency. Jin and Myers (2006) find that idiosyncratic risk increases with a country's accountingtransparency. Wurgler (2000) also shows that the efficiency of capital allocation across countries ispositively correlated with idiosyncratic risk in domestically traded stock returns.

Most of the firm level studies are based on the U.S. market. Durnev et al. (2003) show that idiosyncraticrisk is positively correlated with stock price informativeness. Durnev et al. (2004) show further that firmsthat exhibit higher levels of idiosyncratic risk tend to use more external financing and allocate capitalmore efficiently. The authors argue that firms with higher firm-specific price variation will attract theattention of informed arbitrageurs who trade on their private information. Consequently, the firms' stockprices will thus track their fundamental values more closely. This in turn will reduce informationasymmetry problems that impede external financing and distort capital spending decisions. Similarly,Chen et al. (2007) show that managers of firms with higher idiosyncratic risk incorporate moreinformation about their stock prices into their investment decisions. Ferreira and Laux (2007) find thatfirms with fewer governance provisions that restrict openness to takeover offers are associated with highlevels of idiosyncratic risk and private information flow. They argue that openness to the market forcorporate control and well-developed corporate governance mechanisms encourage the collection of, andtrading on, private information by investors.

However, some recent studies question the use of idiosyncratic volatility as a measure of marketefficiency (e.g., Ashbaugh-Skaife et al., 2006; Dasgupta et al., 2010; Griffin et al., 2010; Kelly, 2007; Teoh etal., 2007). These studies have attempted to test the connection between idiosyncratic volatility and priceinformativeness and couldn't find a positive relationship between them. Some other papers (Lee and Liu,2011; Xing and Anderson, 2011) propose a nonlinear relationship between idiosyncratic volatility andstock price informativeness. However, we do not know how idiosyncratic volatility is valued and used byexternal corporate investors, such as acquirers. The existing literature does not provide any evidence onthe impact of price transparency and the degree of idiosyncratic volatility on key parameters in theacquisition process such as premium, degree of ownership control, and bid probability. Most importantly,none of the studies examines firm level idiosyncratic volatility and mergers and acquisitions in emergingcountries. Our study provides an out-of-the-sample test of the role of idiosyncratic volatility in a verydifferent institutional and information environment from that of the developed market.

2.2. Idiosyncratic volatility in emerging countries

Existing evidence mostly based on studies using stock market indices indicates that idiosyncraticvolatility varies significantly across countries. Morck et al. (2000) and Bartram et al. (2012) find thatidiosyncratic volatility tend to be lower in emerging countries. Gul et al. (2010) argue that loweridiosyncratic volatility in emerging markets stems from two primary sources. First, while many emergingmarkets have disclosure regulations that are of similar quality to those in developed markets, theseregulations are often not fully enforced (Ball, 2001; Chan and Hameed, 2006). Less information disclosurecould lead to higher stock return synchronicity (i.e., less idiosyncratic volatility as argued by Morck et al.,2000). Second, in emerging markets it is quite common to observe concentrated ownership structure,divergence of cash flow rights and voting rights, cross-holdings, business groups and pyramid ownershipstructure. These complicated ownership structures are conducive to managerial entrenchment andprovide controlling owners with incentives and opportunities to extract private control benefits at theexpenses of outside investors (Bertrand et al., 2002; Johnson et al., 2000). In this environment, thecontrolling owners have incentives to withhold (or selectively disclose) value-relevant, privateinformation to outside investors in order to conceal the valuation implication of their self-servingbehaviors (Fan andWong, 2005; Kim and Yi, 2006). As a result, the cost of acquiring private information islikely to be higher, and therefore the profitability of informed trading is likely to be lower, in emerging

22 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

markets as compared to developed markets. This suggests that stock prices will be less informative inemerging markets due to the lack of informed trading of private information by arbitragers.

Although the information disclosure and governance environments are generally weak in emergingcountries, there is significant firm-level variation observed in these countries. For example, in a number ofemerging countries some firms cross list their stocks in developed stock markets, which enforce muchhigher information disclosure and governance standards on those firms (Karolyi, 2012). However, thereare very few studies that examine the cross-sectional variation of idiosyncratic volatilities in emergingcountries. Gul et al. (2010) examined a sample of publicly listed Chinese firms and found that the amountof earnings information reflected in stock returns is lower for firms with high synchronicity. Their resultsconfirm the information efficiency and informed trading explanation of idiosyncratic volatility. Huang etal. (2013) studied the impact of financial market liberalization on the pricing of idiosyncratic risk inemerging markets. Nartea et al. (2011) studied the impact of idiosyncratic volatility on stock returns in theSoutheast Asian markets. None of these studies examined the information role of idiosyncratic volatility inthe corporate mergers and acquisition decisions.

2.3. Idiosyncratic volatility in mergers and acquisitions

A few studies have investigated idiosyncratic volatility and its impact on mergers and acquisitions inthe United States. Moeller et al (2007) find that acquiring firm's idiosyncratic volatility has a negativeimpact on the firm's returns in stock financed acquisition of publicly traded target firms in the US. Using asample of US firms between 1995 and 2004 to examine target firms' idiosyncratic volatility on acquiringfirms' returns, Officer et al. (2009) find significant and substantially higher announcement period returnsto stock-swap acquirers when the target firms' idiosyncratic volatility is high. They attribute these resultsto information asymmetry and the difficulty in evaluating the target firms' assets.

The foregoing studies focused on acquiring firms' returns and financing arrangement. Only a fewstudies examine the impact of idiosyncratic volatility on target firms.5 Based on a sample of 2381 USacquisitions between 1984 and 2004, Chatterjee et al. (2012) find that takeover premium is positivelyrelated to the target's idiosyncratic volatility. Using idiosyncratic volatility of target firms as a measure ofinvestor opinion diversity, the authors conclude that higher idiosyncratic volatility (i.e., more investoropinion diversity) can induce higher acquisition price paid to the target firms.

2.4. Summary of literature review

Overall, the review of existing literature leads us to two key conclusions. First, most of the studies focuson US acquisitions and from the viewpoint of the acquirers. Second, none of the studies has used andinterpreted idiosyncratic volatility as a measure of informed trading and price transparency. Our studyconnects these two strands of literature and provides new insights into the role of idiosyncratic volatilityin mergers and acquisitions. In particular, our study examines the influence of idiosyncratic volatility onmany important parameters in the M&A decision process, including acquisition completeness rate, time tocompletion of the deal, percentage ownership acquired, as well as acquisition probability. We also believethat ours is the first study to examine these important issues in a large sample of emerging country firms.

3. Hypotheses development

3.1. Acquisition premium

Our main objective is to examine the impact of target firms' idiosyncratic volatility on acquisitionpremium. We argue that if idiosyncratic volatility serves as a proxy for asymmetric information, then itwill be negatively correlated with acquisition premium. This expectation is based on the notion that

5 A review of the recent research on mergers and acquisitions in emerging countries (including, Bhagat et al., 2011; Chi et al.,2011; Kohli et al, 2013; Nagano, 2013; Zhu and Jog, 2012; Zhu et al., 2011), shows that to the best knowledge of our knowledge, noneof the studies examines the impact of target firm idiosyncratic volatility on mergers and acquisitions decisions in the emergingcountries.

23P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

acquirers will discount the value of target firms and pay lower premiumwhen they perceive the targets tohave opaque information and more noise in the market valuation. On the other hand, the privateinformation and informed trading explanation posits a positive relationship between target firms'idiosyncratic volatility and acquisition premium. Better information transparency, which results from theincorporation of private information into the stock price of the target firms, may elicit a higher premiumfrom acquiring firms. Following from the foregoing and the stock price informativeness and informedtrading argument, we present the following hypothesis on acquisition premium:

Hypothesis 1. Target firm's idiosyncratic volatility has a positive impact on the acquisition premium.

3.2. Deal completion

Second, we examine the impact of idiosyncratic volatility on deal completion rate and completion time.If idiosyncratic volatility serves as a proxy for information asymmetry, we would expect it to adverselyaffect the completion rate and lengthen the deal completion time as a result of the informationopaqueness and the difficulty in evaluating and negotiating with the target firm, thus leading to a negative(positive) relationship between the target firm idiosyncratic volatility and deal completion rate (dealcompletion time). However, under the private information explanation of idiosyncratic volatility, weexpect that higher idiosyncratic volatility would be associated with higher stock price transparency andmore informative earnings prospect of the company. These features may facilitate the informationcollection and analyses and ease the valuation and negotiation process. Thus, we expect that it will takeless time for the acquiring firm to complete the deal when the target firm has higher price transparencyand informativeness. The correlation between idiosyncratic volatility and the time to completion is thenexpected to be negative. We summarize the two hypotheses (namely information asymmetry (2.1) andinformed trading (2.2)) as follows:

Hypothesis 2.1. Target firm's idiosyncratic volatility has an adverse (positive) impact on the likelihood ofcompleting the acquisition (time taken to complete the acquisition).

Hypothesis 2.2. Target firm's idiosyncratic volatility has a positive (negative) impact on the likelihood ofcompleting the acquisition (time taken to complete the acquisition).

3.3. Degree of control

We examine the degree of ownership control acquired in the target firm in the presence of target firmidiosyncratic volatility. Chari et al. (2010) show that developed countries' acquirers tend to benefitfinancially from acquiring majority control ownership in targets in emerging countries. If idiosyncraticvolatility reflects stock price transparency, we would expect acquiring firms to have more confidence inacquiring higher ownership in the target firm or gain majority control of the company. Given the generallyweak legal environment and rule of law in emerging markets, it is also important for the acquiring firms todefine property rights clearly through the acquisition of majority ownership position instead of a sharedownership position in the emerging markets. Once the acquirer decides to make an acquisition in theemerging market, the best way to deal with possible opportunistic behavior and potential property rightconflicts with the target firm's managers is to acquire a majority control ownership in the target firm.Based on this argument, we expect a positive relationship between idiosyncratic volatility and probabilityof acquiring majority control in the target firm.

Hypothesis 3. Target firm's idiosyncratic volatility has a positive impact on the likelihood of bidderacquiring majority control in the target firm.

3.4. Bid probability

Lastly, we examine the impact of the target firms' idiosyncratic volatility on the probability of receivingtakeover bids. More private information reflected in the target firm's price through informed tradingmeans less information advantage held by the others and the lower the likelihood that the firm will be

24 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

targeted by other investors. Unless the acquiring firms have strong strategic reasons for the investment orstrong confidence in their private information advantage, they are more likely to shy away from targetfirms that have more informed trading and private information. Thus, we expect an inverse relationshipbetween the target firms' idiosyncratic volatility and the probability of receiving bids from other firms.Accordingly, we hypothesize that:

Hypothesis 4. Target firm's idiosyncratic volatility has a negative impact on the firm's probability ofreceiving a bid from other firms.

4. Data and variables

4.1. Data

In this section, we introduce the sample and empirical methods to test each of the hypotheses. Oursample consists primarily of completed acquisitions from 1990 to 2007.6 Data on completed acquisitionsand deal characteristics come from the Security Data Corporation (SDC) database.7 The emerging countriesincluded in the study were selected from the Morgan Stanley Emerging Market Index. One problem ofusing such a source is that some countries are maintained in the index only for continuity purposes, eventhough these countries can be considered as “developed”; therefore we excluded three of such countries(namely, Taiwan, South Korea, and Israel) from the sample. In addition, we excluded Jordan and Moroccofrom the study due to the very small number of acquisitions in these countries. In addition, to be includedin the study, we require stock price and financial statement data for the target firms to be available in theDatastream/Worldscope database.

The initial sample consisted of 8041 mergers and acquisitions in the twenty emerging countries from1990 to 2007 (comprising 5071 domestic acquisitions and 2970 cross-border acquisition). To be includedin the study, the firmmust have stock price and financial statement data from the aforementioned sources.After merging the SDC database with the Datastream database, the sample size dropped to 3156observations (comprising 2029 domestic M&As and 1127 cross-border M&As). Finally, after obtaining firmlevel financial variables (such as firm size and market to book ratio) from the Worldscope database, thesample reduced further. Our final list consists of 742 acquisitions involving publicly traded firms from 20countries including India, Brazil, Poland, Thailand, Indonesia, South Africa, Argentina, Malaysia, the CzechRepublic, China (mainland), Mexico, Hungary, Philippines, Chile, Peru, Russia, Turkey, Colombia, Egypt andPakistan.8 After merging the SDC dataset with the Datastream/Worldscope database, we found 132withdrawn deals with completed financial information. We then collected country-level data from theWorld Development Indicator (WDI) database, Central Intelligence Agency (CIA) World Fact Book,Transparency International, and theWorld Bank. A detailed list of measures and data sources can be foundin Appendix A.

4.2. Independent variable: idiosyncratic volatility measures

The most important variable in the study is idiosyncratic volatility, which we measure in three ways.First, based on the market model (Rit = αi + βiRmt + εit), and as per Durnev et al. (2004) we estimatedfor each target firm in the estimation period (120 days to 64 days prior to the takeover announcement)the variance of the residuals, ε2it, as our proxy for idiosyncratic risk. In this model, Rit is the daily stock

6 We selected this period in order to avoid the effects of the global financial crisis and the sell off that occurred in the emergingcountries. According to the World Investment Report (2013), global cross-border mergers and acquisitions reached peak of$1 trillion in 2007 and dropped thereafter because of the financial crisis. Despite the fact that the global financial crisis occurred in2008, global M&Amarket has not fully recovered by 2013. We study the stable period of 1990 and 2007 to avoid the abnormally highvolatility during the crisis period and the subsequent sell off in the emerging financial market, which potentially could distort theresults.

7 Following Betton and Eckbo (2000), we include transactions that are described in the SDC database as mergers, acquisition ofmajority interest, acquisition of partial interest, and acquisition of remaining interest.

8 Seven of these countries (China, Brazil, Russia, Chile, India, Indonesia, and Columbia) are listed in the top twenty destination offoreign direct investment in 2013 (World Investment Report, 2013). Thus, the sample provides a general representation of theleading corporate investment markets in the emerging countries.

25P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

return for firm i on day t, Rmt is the daily return of the main stock market index in the target country on dayt. All return data were converted into US dollar returns based on the daily closing foreign exchange rate.The βi is estimated from the estimation period data by using the OLS regression. We take ε2it as theidiosyncratic risk that is not explained by the market return (Rmt). In order to ensure that the idiosyncraticvolatility is not affected by the overall volatility, we standardize the idiosyncratic volatility by the totalreturn volatility of the target firm in the estimation window. More specifically, we estimate sigma ratio,

defined as, θi ¼σ2

e;i

σ2t;iwhere σe,i

2 is the market model residual variance obtained from the estimation period

for each target firm and σt,i2 is the total variance of the daily stock returns in the estimation period.

As a robustness check, we also use as our measure of risk (1 − R2) obtained from the market modelabove estimated for each target firm in the estimation period (120 days to 64 days prior to the takeoverannouncement). Due to the econometrically undesirable characteristic of the (1 − R2), which is boundedbetween zero and one, and to be consistent with prior studies, we use the log transformation as suggestedby Durnev et al. (2004)9:

9 Witmay aff10 Rec

ψi ¼ ln1−R2

i

R2i

!:

Durnev et al. (2004) and Ferreira and Laux (2007) use this measure to capture the firm-specificvariation of return and they interpret it as stock price informativeness. Finally, we also use the rawmeasure of the variance of the residuals, ε2it, as another measure of idiosyncratic risk for robustness checkof our results.

4.3. Dependent variables

4.3.1. Acquisition premiumAcquisition premium is estimated as the difference between the offer price and the target firm's stock

price four weeks before the acquisition announcement (as per SDC definition). For emerging countries thedata obtained from SDC has many missing values, including the offer price. Out of the 742 acquisitions, weobtained acquisition premium data from the SDC Platinum database for only 359 firms (48%). To increasethe sample size, we adopt Schwert's (2000) approach and estimate the takeover premium using the targetfirm's abnormal stock returns. Following Schwert, we estimate the takeover premium as the cumulativeabnormal returns of the target firm stocks from 63 days before to 126 days subsequent to the takeoverannouncement date.10 The abnormal returns are defined as the difference between the actual returns andthe expected returns in the event window. The returns are estimated in US dollars; therefore, fluctuationsin the exchange rate are reflected in the calculation of the acquisition premium. Using the market model,we estimate expected returns as follows: Rit = αi + βiRmt + εit, where Rmt is the return on the broadestmarket index in the target's country, Rit is the daily stock return of the target firm. The coefficients αi and βi

are estimated using observations from 120 days to 64 days before the takeover announcement. Thecoefficients are then used to compute daily expected returns around the acquisition announcement date.Since our sample includes both full and partial acquisitions, we adjust the two measures of premiumabove by multiplying the premium by the percent of shares acquired. We then correlate the stock returnbased premium with the SDC premium data and find the correlation to be as high as 0.7. Since the twomethods provide a consistent measure of acquisition premium, we use both methods to identify thesample size. As a result, we are able to increase the sample size significantly. We also conduct robustnesstest by using either the SDC premium data or the stock return based premium data separately and we findthat the results are robust to both measures of acquisition premium.

hout the transformation, the distribution of the variable, which is bounded between zero and one, would be “truncated” andect the OLS regression results.ent studies including Bargeron et al. (2008) and Fu et al. (2013) also used the CAR measure to proxy for acquisition premium.

26 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

4.3.2. Deal completionWe measure deal completion using two variables. First we use a dummy variable that equals 1 if the

deal was successfully completed and the observation has a completion date in the SDC database, and 0 ifthe deal is coded as “withdrawn” in the SDC database for deal completion. We use this variable to estimatethe logistic regression to test if the target firm idiosyncratic volatility has a significant impact on the dealcompletion probability. Second, we focus on the time taken to complete the deal. Based on theannouncement date and the completion date of each deal obtained from the SDC database, we calculatethe elapsed time (in number of days) between the two dates. We take the logarithm of the elapsed timemeasure in the models. Due to the left-censored distribution of the time measure, we use the Tobitregression to test whether the target firm idiosyncratic volatility affects the deal completion time.

4.3.3. Degree of controlWe use a dummy variable to differentiate majority control deals from minority control deals. If the

acquiring firm gains more than 50% ownership in the target firm after the acquisition, we consider theacquisition as majority control acquisition and code it as 1; and if the acquired ownership is less than orequal to 50%, we consider the deal as minority control deal and code it as 0. We then use logistic regressionto examine the impact of the target firm's idiosyncratic volatility on the decision to acquire majoritycontrol in the target firm. As a robustness check, we also use the continuous variable of target firmownership acquired by the bidder to measure the degree of control acquired. The variable ranges from 5%to 100% and is collected from the SDC Platinum database. We use the OLS regression to analyze the impactof the target firm idiosyncratic volatility on ownership control.

4.3.4. Bid probabilityTo examine the impact of the target firms' idiosyncratic volatility on the probability of receiving a bid

from other firms, we expand the sample to include the non-acquired firms in the analysis. To do this, weobtain a list of all publicly listed firms available in the Datastream/Worldscope database. We then mergedthe dataset with the SDCmergers and acquisitions dataset and identified firms that received an acquisitionbid during the sampling years in our dataset. If a firm receives a bid from another firm in year t, we codethe dummy variable Pi,t = 1; if there was no acquisition bid made to acquire a firm in year t, we code thedummy variable Pi,t = 0. For each firm year observation Pi,t, we extract the idiosyncratic volatility of thefirms in the previous calendar year (t − 1).11 The calculation of the idiosyncratic volatility is based on asimple market model estimated by regressing the firm's daily stock returns on the local market indexreturns in the sample year t − 1. Similar to the sigma ratio specified in Section 4.2, we standardize thevariance of the residuals in the model by the total variance of the company's stock returns in the year. As arobustness check, we also calculate the price informativeness measure as log (1 − R2) / R2 to capture theidiosyncratic volatility of the firm in year t − 1. We then regress the lagged idiosyncratic volatilitymeasures on the dummy variable of the acquisition probability in year t.

4.4. Control variables

In this section we describe the control variables and the measures used in this study. Pastor andVeronesi (2003) find a negative relation between volatility and firm age and a positive relation betweenvolatility and market-to-book. They also show that firm size, measured by the logarithm of total assets, isnegatively related to volatility. Irvine and Pontiff (2005) and Comin and Philippon (2005) suggest thatmarket competition and R&D may explain the increase in idiosyncratic volatility of firms. In a sample ofinternational firms, Bartram et al. (2012) also find that lagged R&D and profit margins (which areinversely related to competition) as well as lagged leverage ratio are the most economically importantdeterminants of idiosyncratic volatility. Therefore, we control for the effects of these factors by includingfirm age, market to book ratio, firm size, R&D expense to sales ratio, industry Herfindahl index, profitmargin and leverage ratio on idiosyncratic volatility in all the models. These firm level variables arecollected from the Worldscope database.

11 The estimation is based on a calendar time dataset. The estimation of the idiosyncratic volatility in the previous sections is basedon M&A event time dataset (where the M&A announcement dates are the event dates).

27P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

For the country level variables, we control for the effect of economic development using per capitaGDP. Morck et al. (2000) find that stock return idiosyncratic volatility tends to be higher in economieswith high per capita GDP. We use Kaufman et al. (2007) rule of law index as a proxy for country risk andpolitical stability. As a proxy for shareholder protection and corporate governance we use theanti-self-dealing index from Djankov et al. (2008).12 Higher values of the index are associated withgreater obstacles to self-dealing and hence better shareholder protection and governance. We also use theindex of creditor rights from Djankov et al. (2007) as a measure of investor protection and creditor rights;higher values of this index represent better creditor rights. Finally, we use the ratio of marketcapitalization of publicly traded firms to GDP as a proxy for overall financial development. The countrylevel variables are collected from various sources, including the World Development Indicator, the WorldBank survey and the online data sources provided by the studies referred above.

We also control for several deal-specific variables in the models. We included acquiring firm's statusmeasured by a dummy variable that takes a value of 1 if the acquirer is publicly listed and 0 otherwise. Wecontrol for related versus unrelated acquisitions in the models (Gugler et al., 2003). In line with earlierstudies (e.g., Morck et al., 1990), we capture the effect of industry relatedness using a dummy variableequal to 1 if the acquiring and target firms have the same 4-digit SIC code, and zero otherwise. Wedifferentiate cash deals from the stock or mixed payment deals using a dummy variable that equals 1 if theacquiring firm used cash to acquiring the target firm and equals 0 for other forms of payments are used inthe acquisitions. We also control for different deal characteristics, including deal size, tender offers (Jarrellet al., 1988) and friendly versus hostile acquisitions (Moeller, 2005). In addition, some researchers haveemphasized that the presence of competing bidders influence the magnitude of acquisition premium (e.g.,Laamanen, 2007). We control for the effects of competing bids using a dummy variable categorized as 1 ifthere is a competing bidder, and 0 otherwise. Finally, we control for the target country corruption level,which is found to have an impact on acquisition premium payment (Weitzel and Berns, 2006). For all themodels, we include industry and year dummy variables to control for heterogeneity of themultiple-industry longitudinal dataset. A summary of the variables and their measurements is found inAppendix A.

5. Results

5.1. Descriptive analysis

We present descriptive statistics in Table 1 Panel A. Out of the twenty emerging countries in oursample Malaysia and India have the largest number of M&A deals in the seventeen-year sample period.The average size of the deal is about $USD228 million and the total transaction value is $USD169 billion.Since our study only focuses on publicly listed target firms and since the study requires many-detailedfinancial information available for these firms, our final sample size (742) is relatively small compared towhat is typically used in M&A studies that focus on developed countries. However, our sample iscomparable to that used in other emerging countries M&A studies (e.g., Zhu et al., 2011).

Table 1 Panel B shows the trend in emerging countries M&A deals over the study period. Thesecountries experienced an increase in acquisition deals around the year 2000. Subsequently we have seenanother surging trend of M&A transactions in the late 2000s. Because of the reforms taking place indeveloping countries, mergers and acquisitions and privatizations have become important corporaterestructuring vehicles used to liberalize the emerging countries' economies. In Panel C, we present theM&A transactions by industry using Fama–French industry classification criteria. Out of the 742acquisitions, 322 acquisitions (43%) took place in the manufacturing sector.13 Our sample also includesmany large transactions in the transportation, public utilities and financial services sectors, industries thathave experienced a lot of privatization activities in emerging countries.

12 Data are only available for one year but their analysis suggests that the index is very stable over time.13 According to the World Investment Report (2011), the manufacturing sector contributed 48% of the global foreign investment in2010. The ratio is comparable to the ratio of M&A activities in the manufacturing sector in our sample. The manufacturing sector alsoexperienced the fastest recovery from the global financial crisis after 2009. Thus, our sample reflects the major industries thatexperienced the active investment activities after the financial crisis.

28 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

Table 2 shows the pair-wise correlations of the main variables used in the study. As the figures in thefirst two columns show the idiosyncratic volatility measures are highly correlated (with correlationcoefficient of 0.77). Consistent with the private information and informed trading hypothesis, we find that

Table 1Emerging market mergers and acquisitions sample description.

Panel A. Mergers and acquisitions statistics by target country

The panel shows the descriptive statistics of our sample. The table shows the number of completed acquisitions in each countrybetween 1990 and 2007. It also shows the average transaction size, total transaction value and the percentage of totaltransaction value over the total sample value.

Target country Number ofacquisitions

Average transactionvalue (mil. $)

Total transactionvalue (mil $)

% of totaltransaction value

Argentina 14 451.4 6319.3 3.74%Brazil 48 461.7 22,162.2 13.10%Chile 13 303.0 3939.5 2.33%China 83 57.1 4742.7 2.80%Colombia 5 246.7 1233.7 0.73%Czech 7 954.1 6678.9 3.95%Egypt 5 215.4 1077.1 0.64%Hungary 12 190.4 2285.0 1.35%India 106 51.3 5433.9 3.21%Indonesia 46 134.8 6199.5 3.66%Malaysia 158 169.8 26,534.4 15.86%Mexico 12 462.6 5550.6 3.28%Pakistan 6 695.7 4174.0 2.47%Peru 3 821.1 2463.3 1.46%Philippines 41 102.6 4206.7 2.49%Poland 27 304.5 8220.9 4.86%Russia 10 2.554.8 25,548.4 15.10%South Africa 28 358.2 10,030.3 5.93%Thailand 99 104.3 10,327.3 6.10%Turkey 19 618.0 11.742.5 6.94%Grand total 742 228.0 169,170.1 100%

Panel B. Mergers and acquisitions statistics by year

The panel shows the number of completed acquisitions in the twenty emerging countries in each year between 1990 and 2007.It also shows the average transaction size, total transaction value and the percentage of total transaction value over the totalsample value in each sample year.

Year Number ofacquisitions

Average transactionvalue (mil $)

Total transactionvalue (mil $)

% of totaltransaction value

1990 4 239.2 956.7 0.57%1991 6 68.4 410.4 0.24%1992 5 33.8 168.9 0.10%1993 13 63.3 822.5 0.49%1994 5 37.7 188.3 0.11%1995 6 43.4 260.3 0.15%1996 17 212.8 3617.4 2.14%1997 33 285.0 9406.3 5.56%1998 23 76.1 1751.4 1.04%1999 40 151.6 6063.4 3.58%2000 53 349.3 18,514.9 10.94%2001 55 107.2 5894.6 3.48%2002 40 78.5 3139.3 1.86%2003 81 287.0 23,250.8 13.74%2004 68 220.4 14,989.9 8.86%2005 101 209.1 21,113.2 12.48%2006 103 326.0 33,574.0 19.85%2007 89 281.4 25,042.8 14.50%Grand total 742 228.0 169,170.1 100%

Panel C. Mergers and acquisitions statistics by target sector

The panel shows the industry breakdown of completed transactions in the twenty emerging countries between 1990 and 2007.It also shows the average transaction size, total transaction value and the percentage of total transaction value over the totalsample value in each target industry.

Target sectors Number ofacquisitions

Average transactionvalue (mil $)

Total transactionvalue (mil $)

% of totaltransaction value

0l–09 Agriculture, Forestry, Fishing 6 142.5 854.8 0.51%l0–l4 Mining 27 1114.3 30,087.2 17.79%15–17 Construction 14 58.4 817.7 0.48%20–39 Manufacturing 322 124.5 40,078.8 23.69%40–49 Transportation & public utilities 97 530.5 51,460.3 30.42%50–51 Wholesale trade 12 51.9 622.3 0.37%52–59 Retail trade 27 89.8 2425.2 1.43%69–67 Finance, insurance, real estate 165 230.4 38,011.9 22.47%70–89 Services 70 65.6 4593.9 2.72%9l–99 Public administration 2 109.0 218.1 0.13%Grand total 742 228.0 169,170.1 100%

Table 1 (continued)

29P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

both idiosyncratic volatility measures are positively correlated with acquisition premium. Acquiring firmsperhaps perceive idiosyncratic volatility positively and are willing to a pay higher premium for the highprice transparency of these target firms. We also find that idiosyncratic volatility measures aresignificantly related to several acquisition deal characteristics. More specifically, both the percentageownership acquired and the probability of acquiring majority ownership are positively correlated with theidiosyncratic volatility measures, implying that acquiring firms tend to gain more control in highidiosyncratic risk firms. Deal completion time is also negatively related to the target firm idiosyncraticvolatility.

We also examine the correlation between idiosyncratic volatility and firm characteristics. Consistentwith previous literature, we find that smaller firms tend to have higher idiosyncratic volatility. In addition,target firms with low financial leverage, and firms with longer history tend to have higher idiosyncraticvolatility. For the country level characteristics, we observe that idiosyncratic volatility is positivelycorrelated with the country's financial market depth (proxied by the financial market capitalizationdivided by the GDP of the country) and economic development (proxied by the per capita GDP). Ourpreliminary results also confirm the results of earlier studies that document a positive correlation betweenidiosyncratic volatility and the strength of target country's rule of law and investor right protectionenvironment (as measured by the anti-self-dealing index in Djankov et al., 2007). Similarly, we also findthat idiosyncratic volatility is positively correlated with the creditor rights and the anti-corruptionmeasures of the target countries, which supports the notion that idiosyncratic volatility is associated withbetter investor's protection and institutional environment (documented in Morck et al, 2000).

Overall, it seems that both measures (the sigma ratio and the transformed price informative measure)capture the informed trading and private information explanation of idiosyncratic volatility. To ensurethat these (preliminary) results are not driven by other confounding variables, we estimate multiple linearregression models in the following sections.

5.2. Does idiosyncratic volatility affect acquisition premium?

In this section, we examine the impact of idiosyncratic volatility on acquisition premium, whilecontrolling for target firm characteristics, such as firm size, market to book ratio, leverage, R&D expenseratio, industry concentration (Herfindahl index) and firm age. We also include measures of countrycharacteristics, such as financial market development, economic development, industry competition, ruleof law, investor right protection, disclosure, creditor rights, and corruption to control for countrycharacteristics that can affect acquisition premium. Finally, we include some commonly used variables inM&A studies, such as the status of the acquiring firms (public or private), tender offers, deal size, payment

Table 2Completed mergers and acquisitions sample descriptive statistics and correlations. The table shows the Pearson correlation for the variables used in the study. The table also shows the sampledescriptive statistics, including mean, median and standard deviation of each variable in the last three rows of the table.

1 2 3 4 5 6 7 8 9 10 11 12

1 Idiosyncratic volatility (θ) 12 Idiosyncratic volatility (ψ) 0.77a 13 Acquisition premium 0.14a 0.12a 14 Log of transaction value −0.20a −0.16a 0.21a 15 Majority control acquisition 0.12a 0.10a 0.43a 0.28a 16 Percentage ownership acquired 0.16a 0.18a 0.40a 0.31a 0.59a 17 Pure cash payment −0.02 0.01 −0.17a −0.16a −0.17a −0.27a 18 Deal completion time −0.11a −0.09b 0.03 0.05 0.05 0.13a −0.12a 19 Target firm size −0.13a −0.12a −0.03 0.35a −0.02 −0.03 −0.02 −0.09b 110 Target firm leverage −0.08b −0.10a −0.05 0.03 −0.04 −0.15a −0.05 0.03 0.07b 111 Target firm market-to-book −0.06 −0.03 −0.02 −0.13a −0.04 −0.06c 0.06c 0.10a −0.03 −0.01 112 Target firm R&D to sales 0.02 0 0.01 −0.03 −0.02 −0.05 0.02 −0.03 0.01 −0.01 0.03 113 Target firm age 0.13a 0.10a o.11a 0.06c 0.17a 0.21a −0.03 0.01 0.02 0.03 −0.01 −0.0314 Target industry Herfindahl index 0.01 0.02 0.03 0.16a −0.01 0.02 0.06 −0.10a 0.28a 0.03 −0.02 −0.0115 Target country financial market development 0.23a 0.24a 0.02 0.03 0.02 0.07b −0.03 −0.06c −0.34a −0.06c 0.12a −0.0216 Target country per capita GDP 0.09a 0.14a 0.06 0.22a 0.12a 0.25a −0.08b −0.07b −0.29a −0.08b −0.23a −0.06c

17 Target country rule of law 0.16a 0.14a 0.06c −0.06c 0.06 0.14a 0 −0.07c −0.41a −0.08b 0.09b 0.0218 Target country anti-self-dealing index 0.08b 0.11a 0 −0.21a −0.05 −0.03 −0.02 0.11a −0.28a 0 0.31a −0.0319 Target country disclosure index −0.02 0.01 −0.01 −0.16a −0.04 0 −0.02 0.13a −0.24a 0.01 0.23a −0.0520 Target country creditor right index 0.13a 0.12a 0.02 −0.20a −0.05 −0.05 0.01 0.01 −0.11a 0.01 0.29a 021 Target country corruption index 0.15a 0.17a 0.07c 0.08b 0.10a 0.20a −0.05 −0.02 −0.37a −0.07b −0.03 −0.0122 Competing bids −0.01 0.02 0.07b 0.05 0.02 0.02 0.02 0 −0.01 −0.04 0.02 0.0423 Public acquirers 0.06 0.06c 0.07b 0.11a 0.09b 0.21a −0.23a 0.02 0.03 −0.02 −0.03 −0.0124 T ender offers 0.09b 0.09b 0.12a 0.10a 0.17a 0.49a −0.04 0.09b −0.07c −0.15a −0.04 −0.0225 Industry relatedness (4SIC) 0 0.02 0.13a 0.21a 0.15a 0.19a −0.12a −0.04 0.06 −0.08b 0 026 Developed country acquirers −0.02 0.02 0.02 0.19a 0.03 0.06c 0.05 −0.05 0.08b 0 −0.03 0.09b

27 Acquirers' acquisition experience 0.05 0.03 −0.07c 0 −0.02 −0.01 −0.02 −0.11a 0.08b 0 0.02 −0.05Mean 0.89 2.06 2.85 3.59 0.00 0.29 1.00 0.11 14.19 −1.38 0.33 0.00Median 0.83 2.64 7.65 3.63 0.23 0.40 0.93 0.25 14.77 −1.84 0.01 0.19Std dev. 0.17 2.33 11.13 1.94 0.42 0.32 0.25 0.41 2.66 1.38 1.73 2.36

a 1% significance.b 5% significance.c 10% significance.

30P.Zhu

etal./

Emerging

Markets

Review19

(2014)18

–48

13 14 15 16 17 18 19 20 21 22 23 24 25 26 27

1−0.08b 1

0.15a −0.21a 10.10a 0.07c 0.37a 10.20a −0.07c 0.54a 0.53a 10.16a −0.32a 0.61a 0.08b 0.41a 10.09b −0.37a 0.34a 0.12a 0.20a 0.84a 10.18a −0.24a 0.54a −0.11a 0.41a 0.85a 0.64a 10.13a −0.07c 0.65a 0.72a 0.84a 0.39a 0.19a 0.23a 1

−0.05 0.07c 0.04 0.01 0 −0.01 −0.03 0.01 0.01 10.06 0.06 0.03 0.17a 0.11a −0.02 −0.01 0 0.11a 0.02 10.15a 0 0.06c 0.18a 0.20a 0 0 −0.03 0.20a 0.04 0.07c 10.05 0.10a −0.06 0.16a 0.03 −0.15a −0.13a −0.16a 0.06c 0.05 0.28a 0.10a 10.01 0.12a −020a 0.05 −0.01 −0.25a −0.17a −0.17a −0.08b −0.01 0.11 0.08b 0.16a 10.07c −0.01 0.15a −0.03 0.03 0.05 −0.04 0.09b 0 −0.01 0.12b −0.07c −0.04 −0.14a 19.30 0.18 0.48 7.78 0.07 0.65 9.50 2.16 3.27 0.00 0.00 0.00 0.00 0.00 0.009.53 0.26 0.73 7.61 0.11 0.64 8.13 1.89 3.70 0.01 0.38 0.16 0.25 0.21 0.375.14 0.25 0.59 0.81 0.48 0.25 2.34 0.07 1.07 0.09 0.48 0.37 0.43 0.41 0.48

31P.Zhu

etal./

Emerging

Markets

Review19

(2014)18

–48

32 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

method and industry relatedness. Industry and year dummies are also included in the regression, albeit wedo not report these coefficients in the tables.

Table 3 shows the model with only the control variables. We find that public acquirers tend to payhigher premium than private acquirers and that acquisition premium is significantly higher in tender offerdeals than non-tender offers. In addition, we find some marginal, albeit significant evidence that targetfirms in countries with higher disclosure and creditor right protection scores tend to receive higheracquisition premium. In Models 1 and 2 of Table 3 we add the two alternative measures of idiosyncraticvolatility (θ and ψ) in the regression. Both measures exhibit positive and significant coefficients (βθ =1.166, significant at 1%; βψ = 0.075 and is also significant at the 1% level). These results supportHypothesis 1 and confirm the findings in the literature that idiosyncratic volatility is related to higheracquisition premium which can be attributed to diverse investor opinion (e.g., see Chatterjee et al., 2012).This finding is consistent with the private information hypothesis of idiosyncratic volatility since targetfirms with higher idiosyncratic volatility seem to receive higher acquisition premium.

The high positive correlation between idiosyncratic volatility and acquisition premium that wedocument above can be explained in a number of ways. Higher idiosyncratic volatility suggests betterinformativeness and transparency of the target firm's stock price. As the target firm's stock price reflectsmore firm-specific information, it may attract the attention of acquirers, particularly those who have lessinside information who may pay high premium for the target. In addition, when the stock price isinformative due to more informed trading, acquiring firms may spend more resources to acquireadditional private information before making the M&A decision. Given the cost they have to incur inacquiring information, and perhaps due to some important strategic reasons for engaging in theacquisition, bidders may become aggressive in bidding for the target firm by offering higher price topreempt potential competing bids.

To provide more evidence on the private information and price efficiency explanation of theidiosyncratic volatility, we conduct several additional split-sample tests. Specifically, we split the sampleon the basis of variables that are related to private information and strategic decisions.

First, we split the sample into two depending on whether the acquirer is from an emerging country ordeveloped country and present the results in Table 4 Panel A. We find that only acquirers from emergingcountry pay higher premium to target firms with higher idiosyncratic volatility. For the emergingcountries acquirers' regression, the coefficient of idiosyncratic volatility is positive and significant at 1%level (βθ = 1.559). However, for the subsample of developed country acquirers, the coefficient ofidiosyncratic volatility is negative but not significant at conventional levels (βθ = −1.428). Weargue that emerging market acquirers are more information disadvantaged than those from developedcountries. They also tend to have weaker governance and poor information disclosure than theirdeveloped country peers. Target firms with better price transparency and governance may help theacquiring firm bootstrap their own information disclosure and governance standard (Martynova andRenneboog, 2008) and thus create more synergy. Better information environment of the target alsofacilitates the integration process and reduce the opportunistic behavior of managers in thepost-acquisition period. Therefore, the emerging country acquirers could value idiosyncratic volatilitymore positively and are willing to pay a higher premium for this valuable feature of the target thanacquirers from developed countries.

Second, we differentiate acquiring firms that had previous acquisition experience in the local marketfrom those that had no previous acquisition experience and examine how their experience in the localmarket affects the premium that they pay. Table 4 Panel B shows that only the acquirers without prioracquisition experience pay higher premiums to target firms with higher idiosyncratic volatility. We arguethat inexperienced acquirers are more disadvantaged than experienced acquirers in terms of the cost ofcollecting and analyzing the target's firm-specific information. Therefore, they value the informativenessof the target firm's stock price induced by informed trading of private information by market participantsand are willing to pay a higher premium for it.

Our main results confirm the assertion that acquiring firms perceive idiosyncratic volatility positivelyand are willing to pay higher price. Higher idiosyncratic volatility might be perceived valuable particularlyin emerging countries as it reflects the price informativeness and transparency of the firm's shares. Oursubsample analysis further suggests that the feature of price informativeness is more valuable to acquiringfirms that are informationally disadvantaged. Since this is our primary finding, we conduct several

Table 3Impact of idiosyncratic volatility on acquisition premium. The table shows the OLS regression results for the sample firms. The dependentvariable is acquisition premium. The premium measure is based on the percentage difference between the offer price and the target firm'sstock price four weeks before the acquisition announcement. We also follow Schwert (2000) and estimate the takeover premium using theabnormal return of the target firm stock from day −63 to day +126 surrounding the acquisition announcement date. We use this stockreturn based premium measure to supplement the missing value in the SDC premium measure. The key independent variable is theidiosyncratic volatility of the target firm. We estimate the variance of the residual of the market model of each target firm in the emergingcountry sample based on the local market index between days (−120 and −61) relative to the acquisition announcement date. We thenstandardize the residual variance by the total variance of the daily returns of the target firm in the estimation window to estimate theidiosyncratic volatility (θ); alternatively, we follow Durnev et al. (2004) and use the log transformation of (1 − R2) / R2 to estimate theidiosyncratic volatility (ψ).We test bothmeasures inmodel 1 andmodel 2 respectively.Wealso control for deal characteristics (such as tenderoffers and industry relatedness), targetfirmcharacteristics (includingfirm size, leverage, growthpotential, R&D to sales,firmage) and countrylevel variables (financialmarket development, industry Herfindahl index, per capita GDP, rule of law index, anti-self-dealing index, disclosureindex, creditor right index, target country corruption and the public status of the acquiring firms. Finally, we include the industry dummyvariables and year dummy variables in the models. All standard errors are robust for heteroscadasticity and autocorrelation.

Dependent variable: Acquisition premium Control model Model 1 Model 2

Intercept 4.022 13.915 9.965(32.527) (32.483) (32.425)

Idiosyncratic volatility (θ) 1.166a

(0.377)Idiosyncratic volatility (ψ) 0.075a

(0.026)Target firm size −0.410a −0.370a −0.380a

(0.043) (0.045) (0.044)Target firm leverage 0.007 0.006 0.011

(0.040) (0.040) (0.040)Target firm market-to-book −0.099b −0.100b −0.100b

(0.047) (0.047) (0.047)Target firm R&D to sales 0.023 0.022 0.023

(0.022) (0.022) (0.022)Target firm age 0.012 0.009 0.010

(0.013) (0.013) (0.013)Target country financial market development −0.168 −0.161 −0.183

(0.250) (0.249) (0.249)Target industry Herfindahl index −0.084 −0.109 −0.109

(0.295) (0.293) (0.293)Target country per capita GDP −0.868 −1.035 −0.941

(1.035) (1.030) (1.030)Target country rule of law 0.138 2.855 1.772

(7.024) (7.035) (7.011)Target country anti-self-dealing index −0.025 −5.796 −3.670

(12.90) (12.95) (12.89)Target country disclosure index 0.125 −0.774 −0.381

(2.476) (2.477) (2.469)Target country creditor right index 0.276 3.401 2.161

(7.647) (7.666) (7.635)Target country anti-corruption index −0.116 −1.483 −0.945

(3.353) (3.362) (3.348)Competing bids 0.514 0.627 0.540

(0.609) (0.606) (0.606)Public acquirers 0.234c 0.222c 0.217c

(0.121) (0.120) (0.120)Tender offers 0.291c 0.281c 0.279c

(0.155) (0.154) (0.154)Industry relatedness (4SIC) −0.095 −0.102 −0.103

(0.137) (0.136) (0.137)Deal size 0.439a 0.449a 0.447a

(0.038) (0.038) (0.038)Cash payment −0.557b −0.539b −0.567b

(0.220) (0.219) (0.219)Industry dummy Yes Yes YesCountry dummy Yes Yes YesYear dummy Yes Yes Yes

(continued on next page)

33P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

Table 3 (continued)

Dependent variable: Acquisition premium Control model Model 1 Model 2

R square 0.272 0.282 0.281F statistics 4.317a 4.458a 4.426a

N 742 742 742

a 1% significance.b 5% significance.c 10% significance.

34 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

robustness tests on the relationship between acquisition premium and idiosyncratic volatility in a latersection.

5.3. Does idiosyncratic volatility affect deal completion rate?

We examine the relationship between idiosyncratic volatility and acquisition completion rate. As perHypotheses 2.1 and 2.2 we expect a higher probability of deal completion in the presence of higheridiosyncratic volatility. To test this hypothesis, we expand the sample to include withdrawn acquisitionsand estimate a logistic regression model where the dependent variables takes on a value of 1 if theacquisition is a completed deal and 0 if the deal proposal was withdrawn, and using idiosyncratic volatilityas our key independent variable. The results presented in Table 5 Panel A are consistent with the informedtrading hypothesis as idiosyncratic volatility positively impact on the acquisition completion rate. Theregression coefficients of both measures of idiosyncratic volatility are positive and significant (βθ = 1.369,significant at 5% level; βψ = 0.143, significant at 5% level). The results support the private information andinformed trading hypothesis (i.e., Hypothesis 2.1).14

We further examine the impact of idiosyncratic volatility on the time taken to complete the deal. Foreach completed deal, we count the number of days between the acquisition announcement date and theacquisition completion date. We then regress the idiosyncratic volatility measures on the elapsed time tocompletion of the deal. Due to the censored nature of the distribution of the time measure, we use Tobitregression to test the models. In addition, due to the existence of special regulations governing tenderoffers in different countries, e.g. on minimum times to consummate deals in tender offers (such as theWilliams Act in the US), we remove tender offers from the sample for this test. The results, which wepresent in Panel B of Table 5, indicate that the regression coefficients for idiosyncratic volatility arenegative and significant (βθ = −0.217, significant at 1% level; βψ = −0.016, significant at 1% level). Theresults support Hypothesis 2.2 and again confirm our assertion that stock price informativeness facilitatesthe acquisition process andmay ease the negotiation and valuation work for the acquiring firms. Acquiringfirms also have greater incentives to quickly consummate the deal to acquire target firm with transparentprice and good governance, which is reflected in high idiosyncratic volatility. These results are consistentwith the informed trading hypothesis (H2.2).

5.4. Does idiosyncratic volatility affect the ownership acquired in the acquisitions?

Chari et al. (2010) assert that gaining majority control in an emerging market firm benefits thedeveloped country acquirers and thus acquiring majority control of the target in emerging markets is animportant strategic decision in mergers and acquisitions. We extend Chari et al. (2010) study and test theimpact of firm-level idiosyncratic volatility on the degree of control gained in the acquisition using twoapproaches. First, we differentiate majority ownership acquisitions (where the target shares acquired ismore than 50%) from minority ownership acquisitions (where the ownership acquired is equal to or lessthan 50%). We code majority control acquisitions as 1 and minority control acquisitions as 0. A logisticregression is estimated using the dummy variable as the dependent variable and idiosyncratic volatility asthe key independent variable. We control for deal, firm and country variables in the regression using thesame set of control variables used earlier. In addition, we also include transaction size as an additionalcontrol variable as transaction size is highly related to the percentage ownership acquired in the target

14 We also re-estimated the regression using probit technique, but the results remain qualitatively the same.

Table 4Subsample tests of impact of idiosyncratic volatility on acquisition premium. Panel A shows the results of the regression where we split thesample into two based onwhether the acquirer is from a developed country or a developing country. Panel B shows similar regression resultsbased on whether the acquiring firm has prior acquisition experience in the target country or not. The dependent variable is the acquisitionpremium. The independent variable is the residual variance of the market model of the target firm divided by the total variance of the dailyreturns of the target firm in the estimation window (θ). For brevity sake, we do not report the results for idiosyncratic volatility (ψ), whichprovides similar results as the ones presented here. We control for deal characteristics, target firm characteristics, country level variables,industry and year dummy variables in the subsample models. All standard errors are robust for heteroscadasticity and autocorrelation.

Panel A Panel B

Dependent variable:Acquisition premium

Developedcountry acquirers

Developingcountry acquirers

Acquirers withacquisition experiencein the target country

Acquirers withoutacquisition experiencein the target country

Intercept 172.60a −52.75 22.12 19.69(78.42) (41.34) (59.897) (40.816)

Idiosyncratic volatility (θ) −1.428 1.559b 0.365 1.436b

(0.994) (0.441) (0.607) (0.507)Target firm size −0.571b −0.344b −0.507b −0.277b

(0.129) (0.049) (0.074) (0.059)Target firm leverage 0.040 −0.009 0.051 −0.025

(0.100) (0.047) (0.059) (0.057)Target firm market-to-book −0.002 −0.130a −0.054 −0.128a

(0.098) (0.054) (0.080) (0.059)Target firm R&D to sales 0.037 −0.051 −0.564 0.021

(0.022) (0.124) (0.627) (0.023)Target firm age −0.004 0.000 0.011 0.010

(0.035) (0.014) (0.018) (0.018)Target country financial marketdevelopment

−0.638 −0.023 0.194 −0.334(1.078) (0.281) (0.358) (0.410)

Target industry Herfindahl index 0.404 −0.056 0.481 −0.324(0.760) (0.339) (0.485) (0.387)

Target country per capita GDP −6.272a 0.069 −2.512 −0.499(2.61) (1.226) (1.952) (1.276)

Target country rule of law 33.070a −12.920 0.900 6.301(16.21) (9.248) (12.76) (8.877)

Target country anti-self-dealing index −58.080c 18.290 6.848 −11.345(30.32) (16.6) (24.46) (16.30)

Target country disclosure index −12.250a 5.037 −0.388 −1.730(5.722) (3.264) (4.524) (3.115)

Target country creditor right index 37.410a −13.440 −1.933 7.065(17.69) (10.02) (14.08) (9.644)

Target country anti-corruption index −15.810a 6.374 −0.378 −2.930(7.764) (4.46) (6.047) (4.257)

Competing bids 0.110 0.550 −0.746 0.852(1.846) (0.670) (1.181) (0.799)

Public acquirers 0.260 0.183 0.500b −0.003(0.290) (0.140) (0.181) (0.171)

Tender offers 0.727a 0.074 0.630a 0.159(0.316) (0.183) (0.276) (0.197)

Industry relatedness (4SIC) −0.053 −0.083 −0.302 −0.062(0.298) (0.162) (0.229) (0.183)

Deal size 0.603b 0.444b 0.489b 0.432b

(0.102) (0.043) (0.060) (0.051)Cash payment 0.657 −0.620a −0.938b −0.382

(0.648) (0.245) (0.332) (0.295)Industry dummy Yes Yes Yes YesCountry dummy Yes Yes Yes YesYear dummy Yes Yes Yes YesR square 0.558 0.290 0.476 0.287F statistics 2.449b 3.562b 3.476b 2.753b

N 157 585 271 471

a 5% significance.b 1% significance.c 10% significance.

35P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

Table 5Impact on idiosyncratic volatility on deal completion.

Panel A. Impact on deal completion rate

Panel A shows the Logistic regression results for deal completion rate. The dependent variable is a dummyvariable,which equals 1 if theacquisition is completed and0 if the acquisition iswithdrawn by the acquiring firm. There are 132withdrawn deals and 742 completeddeals in our sample. The key independent variable is the idiosyncratic volatility of the target firm (θ) and (ψ). We test thesetwomeasures in model 1 and model 2 respectively. We also control for deal characteristics (such as tender offers and industryrelatedness), target firm characteristics (including firm size, leverage, growth potential, R&D to sales, firm age) and countrylevel variables (financial market development, industry Herfindahl index, per capita GDP, rule of law index, anti-self-dealingindex, disclosure index, creditor right index, target country corruption). We also control for public status of the acquiring firms, theexistence of competing bidders as well as the transaction size. Finally, we include the industry dummy variables and year dummyvariables in the models. All standard errors are robust for heteroscadasticity and autocorrelation.

Dependent variable: Deal completion probability Control Model Model 1 Model 2

Intercept 51.000 −1.828 23.240(34.45) (2.755) (35.85)

Idiosyncratic volatility (θ) 1.369a

(0.621)Idiosyncratic volatility (ψ) 0.143a

(0.069)Target firm size 0.229a 0.154a 0.274a

(0.101) (0.066) (0.107)Target firm leverage −0.247a −0.303b −0.280a

(0.109) (0.111) (0.114)Target firm market-to-book 0.278a 0.089 0.229a

(0.11) (0.074) (0.112)Target firm R&D to sales 0.287 0.252 0.218

(0.306) (0.332) (0.313)Target firm age 0.010 0.027 0.008

(0.030) (0.028) (0.030)Target country financial market development −0.759 0.424 −0.707

(0.506) (0.417) (0.514)Target industry Herfindahl index 0.403 0.970 0.592

(0.698) (0.713) (0.753)Target country per capita GDP −2.959 0.587c −1.161

(2.066) (0.306) (2.146)Target country rule of law 13.330 1.909b 5.664

(12.92) (0.710) (13.41)Target country anti-self-dealing index −24.200 −3.407c −13.700

(20.14) (1.889) (20.93)Target country disclosure index −4.269 0.097 −1.966

(3.594) (0.153) (3.721)Target country creditor right index 11.650 −0.088 5.824

(10.37) (0.451) (10.74)Target country anti-corruption index −9.407 −1.557a −3.818

(9.843) (0.761) (10.22)Competing bids −1.425 −0.809 −1.146

(0.867) (0.891) (0.942)Public acquirers −0.825b −0.661a −0.809b

(0.267) (0.258) (0.273)Tender offers −0.400 −0.071 −0.236

(0.321) (0.317) (0.337)Industry relatedness (4SIC) −0.507 −0.326 −0.430

(0.285) (0.283) (0.293)Deal size −0.513b −0.414b −0.505b

(0.098) (0.089) (0.101)Industry dummy Yes Yes YesCountry dummy Yes Yes YesYear dummy Yes Yes YesPseudo-R2 0.531 0.529 0.548N 874 874 874

(continued on next page)

36 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

Table 5 (continued)

Panel B. Impact on deal completion time

Panel B shows the Tobit regression results for deal completion time. The dependent variable is the elapsed time (in years)between the acquisition announcement date and the acquisition completion date. The analysis is based on 622 completed deals.We exclude the tender offers from the sample. The key independent variable is the idiosyncratic volatility of the target firm (θ)and (ψ). We test these two measures in model 1 and model 2 respectively. We control for deal characteristics, target firmcharacteristics and country level variables. We also control for public status of the acquiring firms, the dummy variable ofexistence of competing bidders as well as the transaction. Finally, we include the industry dummy variables and year dummyvariables in the models. All standard errors are robust for heteroscadasticity and autocorrelation.

Dependent variable: Deal completion time Control model Model 1 Model 2

Intercept 0.933a 1.225b 1.080a

(0.433) (0.437) (0.438)Idiosyncratic volatility (θ) −0.217a

(0.094)Idiosyncratic volatility (ψ) −0.016a

(0.007)Target firm size −0.028b −0.028b −0.029b

(0.008) (0.008) (0.008)Target firm leverage 0.004 0.005 0.003

(0.012) (0.012) (0.012)Target firm market-to-book 0.019c 0.018c 0.019c

(0.010) (0.010) (0.010)Target firm R&D to sales −0.002 −0.002 −0.002

(0.005) (0.005) (0.005)Target firm age −0.002 −0.002 −0.002

(0.003) (0.003) (0.003)Target country financial market development −0.177b −0.171b −0.170b

(0.054) (0.054) (0.054)Target industry Herfindahl index −0.021 −0.014 −0.014

(0.080) (0.080) (0.080)Target country per capita GDP −0.096a −0.093a −0.092a

(0.038) (0.038) (0.038)Target country rule of law −0.116 −0.119 −0.129

(0.085) (0.085) (0.085)Target country anti-self-dealing index 0.931b 0.912b 0.918b

(0.239) (0.238) (0.239)Target country disclosure index −0.019 −0.020 −0.020

(0.017) (0.017) (0.017)Target country creditor right index −0.160b −0.151a −0.150a

(0.060) (0.060) (0.060)Target country anti-corruption index 0.076 0.077 0.079c

(0.047) (0.047) (0.047)Competing bids 0.097 0.076 0.103

(0.198) (0.198) (0.198)Public acquirers 0.024 0.025 0.027

(0.035) (0.035) (0.035)Industry relatedness (4SIC) −0.080a −0.081a −0.080a

(0.039) (0.039) (0.039)Deal size 0.052b 0.047b 0.049b

(0.010) (0.010) (0.010)Cash payment −0.116c −0.128c −0.120c

(0.065) (0.065) (0.065)Industry dummy Yes Yes YesCountry dummy Yes Yes YesYear dummy Yes Yes YesPseudo-R2 0.224 0.227 0.267N 622 622 622

a 5% significance.b 1% significance.c 10% significance.

37P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

38 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

firms. We present the logistic regression results in Table 6 Panel A. The coefficients of the idiosyncraticvolatility measures are positive and highly significant (βθ = 2.982, significant at 1% level; βψ = 0.145,significant at 1% level). The results support Hypothesis 3 and suggest that acquiring firms are more likelyto acquire majority control in target firms that have higher idiosyncratic volatility.15

As a robustness check, we use a continuous measure, namely the percentage ownership acquired(instead of the dummy variable) as the dependent variable and estimate an OLS regression to test theimpact of idiosyncratic volatility on the ownership acquisition decision. Again, we control for transactionsize and other firm, deal and country characteristics in the model. The results presented in Table 6 Panel Bconfirm the positive and significant regression coefficients of the idiosyncratic volatility measures in themodel (βθ = 0.167, significant at 1% level; βψ = 0.015, significant at 1% level). The results suggest thatbidders acquire higher ownership in the target firm to gain control of the private information source in thetarget firm. Given the higher price transparency and informativeness of the target firms' share pricereflecting the market's private information set, the acquiring firms are prepared to acquire larger stakes inthe target firms and reduce the risk of opportunistic managerial behavior.

5.5. Does idiosyncratic volatility affect the bid probability?

Finally, we investigate the impact of the target firm's idiosyncratic volatility on its probability ofreceiving a bid. Although Ferreira and Laux (2007) find that firms with higher idiosyncratic volatility tendto have fewer anti-takeover restrictions in the United States, they don't examine whether such opennessto the market for corporate control actually increases the probability of the firm receiving acquisition bids.We directly test the relationship between the target firm's idiosyncratic volatility and its probability ofreceiving bids in the emerging market context. To do this, we expand our sample to include all publiclytraded firms that were not acquired in the twenty emerging countries during the sampling period. Weobtained a list of all publicly traded firms in the Worldscope database. The stock price information and thefirm level financial information were obtained from the Datastream and Worldscope databases for eachfirm and each year between 1990 and 2007. After matching the databases, we had 36,934 firm-yearobservations with complete financial information at the firm and country levels. We then identified thefirms that received acquisition bids in the year and coded the observations as 1 and the other firm-yearobservations as 0. We use this binary variable as the dependent variable in the logistic regression. Toremove the potential endogenous impact on the econometrics models, we use one-year laggedidiosyncratic volatility measure of the firm as independent variable. The idiosyncratic volatilities areestimated using the daily stock returns of the firm regressed on the daily return of the local market indexin each sample year. The variance of the residuals from the market model is used to estimate thestandardized volatility measures as specified in Section 4. The estimated models also include the one-yearlagged firm level and country level control variables and industry and year dummy variables. The results ofthe regression are presented in Table 7.

Consistent with our hypothesis, we find an inverse relationship between the target firm's idiosyncraticvolatility and acquisition probability. The coefficients of idiosyncratic volatility are negative and significant(βθ = −0.719, significant at 1% level; βψ = −0.038, significant at 1% level). The results supportHypothesis 4. We argue that firms with high idiosyncratic volatility tend to have more price transparency,with the stock price reflecting their fundamental values. Thus, gathering any additional privateinformation on the firm is costly and time consuming. Unless there is a strong strategic reason motivatingthe acquisition or the acquiring firm has very strong private information advantage, bidders will not havethe incentive to collect the additional private information and pay higher price to acquire these targetfirms.

Put together, our results indicate that firms with higher idiosyncratic volatility in emerging markets areless likely to receive acquisition bid, but once bidders decide to take over such firms, they pay highpremiums. Bidders that are likely to do this are those that are informationally disadvantaged (i.e., biddersfrom emerging markets and those that have less experience in the target country).

15 The conclusion remains the same when we use the probit regression technique.

Table 6Impact of idiosyncratic volatility on other acquisition decisions.

Panel A. Impact on majority control acquisitions

Panel A shows the results of the logistic regression for the propensity of the acquirers to make majority control acquisitions. Thedependent variable is a dummy variable which equals 1 if the acquiring firm owns more than 50% ownership in the target firmafter the acquisition and 0 if the ownership acquired is less than or equal to 50%. Our key independent variable is idiosyncraticvolatility of the target firm (θ) and (ψ). We test these two measures in model 1 and model 2 respectively. We control for dealcharacteristics, target firm characteristics and country level variables. We also control for public status of the acquiring firms, thedummy variable of existence of competing bidders as well as the transaction size. Finally, we include industry dummy variablesand year dummies in the models. All standard errors are robust for heteroscadasticity and autocorrelation.

Dependent variable: Majority control acquisition Control model Model 1 Model 2

Intercept −1.049 −2.086 −0.308(2.597) (2.601) (2.607)

Idiosyncratic volatility (θ) 2.982a

(0.760)Idiosyncratic volatility (ψ) 0.145a

(0.045)Target firm size −0.200a −0.212a −0.205a

(0.055) (0.056) (0.056)Target firm leverage −0.004 0.006 0.004

(0.074) (0.075) (0.075)Target firm market-to-book 0.025 0.046 0.030

(0.064) (0.065) (0.065)Target firm R&D to sales −0.084 −0.085 −0.082

(0.181) (0.192) (0.187)Target firm age 0.044b 0.037 0.043b

(0.022) (0.023) (0.023)Target country financial market development −0.705b −0.941c −0.884c

(0.414) (0.430) (0.426)Target industry Herfindahl index 0.422 0.243 0.328

(0.530) (0.535) (0.532)Target country per capita GDP −0.501b −0.555c −0.563c

(0.257) (0.257) (0.258)Target country rule of law −1.063b −0.955 −0.934

(0.597) (0.601) (0.604)Target country anti-self-dealing index −1.909 −1.880 −1.845

(1.441) (1.449) (1.442)Target country disclosure index 0.040 0.074 0.061

(0.106) (0.107) (0.106)Target country creditor right index 0.611 0.591 0.553

(0.380) (0.381) (0.380)Target country anti-corruption index 0.901a 0.865a 0.866a

(0.324) (0.325) (0.328)Competing bids 0.193 0.586 0.326

(1.105) (1.137) (1.079)Public acquirers 0.198 0.174 0.178

(0.230) (0.233) (0.231)Tender offers 0.480b 0.424 0.435b

(0.258) (0.261) (0.259)Industry relatedness (4SIC) 0.214 0.210 0.187

(0.243) (0.246) (0.245)Deal size 0.466a 0.562a 0.522a

(0.076) (0.082) (0.079)Cash payment −0.965a −0.909c −0.977a

(0.373) (0.38) (0.376)Industry dummy Yes Yes YesCountry dummy Yes Yes YesYear dummy Yes Yes YesPseudo-R2 0.198 0.219 0.211N 742 742 742

(continued on next page)

39P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

Table 6 (continued)

Panel B. Impact on percentage ownership acquired

Panel B shows the OLS regression results for the percentage of ownership acquired. The dependent variable is the percentageownership in the target firms acquired by the bidder. The independent variable is idiosyncratic volatility of the target firm (θ)and (ψ). We test these two measures in model 1 and model 2 respectively. We control for deal characteristics, target firmcharacteristics and country level variables. We also control for public status of the acquiring firms, the dummy variable ofexistence of competing bidders as well as the transaction size in the regression. Finally, we include the industry dummyvariables and year dummy variables in the models. All standard errors are robust for heteroscadasticity and autocorrelation.

Dependent variable: Percentage ownership acquired Control model Model 1 Model 2

Intercept 5.345 6.763 6.558(5.483) (5.487) (5.451)

Idiosyncratic volatility (θ) 0.167a

(0.063)Idiosyncratic volatility (ψ) 0.015a

(0.004)Target firm size −0.044a −0.039a −0.038a

(0.007) (0.007) (0.007)Target firm leverage −0.017c −0.017c −0.016c

(0.006) (0.006) (0.006)Target firm market-to-book 0.005 0.005 0.005

(0.008) (0.007) (0.007)Target firm R&D to sales −0.004 −0.004 −0.004

(0.003) (0.003) (0.003)Target firm age 0.004c 0.004b 0.004b

(0.002) (0.002) (0.002)Target country financial market development −0.008 −0.007 −0.011

(0.042) (0.042) (0.041)Target industry Herfindahl index 0.023 0.019 0.018

(0.049) (0.049) (0.049)Target country per capita GDP −0.265 −0.289b −0.280

(0.174) (0.174) (0.173)Target country rule of law 0.810 1.199 1.143

(1.184) (1.188) (1.178)Target country anti-self-dealing index −1.771 −2.598 −2.515

(2.175) (2.188) (2.168)Target country disclosure index −0.267 −0.396 −0.370

(0.417) (0.418) (0.415)Target country creditor right index 0.908 1.356 1.293

(1.289) (1.295) (1.283)Target country anti-corruption index −0.346 −0.542 −0.515

(0.565) (0.567) (0.563)Competing bids −0.031 −0.014 −0.025

(0.102) (0.102) (0.101)Public acquirers 0.066a 0.064a 0.063a

(0.020) (0.020) (0.020)Tender offers 0.326a 0.324a 0.323a

(0.026) (0.026) (0.026)Industry relatedness (4SIC) 0.006 0.005 0.004

(0.023) (0.023) (0.023)Deal size 0.054a 0.056a 0.056a

(0.006) (0.006) (0.006)Cash payment −0.239a −0.236a −0.241a

(0.037) (0.037) (0.036)Industry dummy Yes Yes YesCountry dummy Yes Yes YesYear dummy Yes Yes YesR square 0.494 0.499 0.502F statistics 11.26a 11.28a 11.45a

N 742 742 742

a 1% significance.b 10% significance.c 5% significance.

40 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

41P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

5.6. Further robustness tests

5.6.1. Alternative measures of idiosyncratic volatilityIn this section, we carried out robustness test on our measure of idiosyncratic volatility. First, in the

above analyses, we have presented the results based on two commonly used idiosyncratic volatilitymeasures, namely the standardized idiosyncratic volatility measure or the sigma ratio (θ) and thelog-transformed idiosyncratic (1 − R2) measure (ψ). We standardize both measures by the total varianceor explained variance of the market model to capture the firm-level variation of the idiosyncratic volatilitymeasure. Although not reported here, we also used the unstandardized idiosyncratic volatility measure —

Table 7Impact of idiosyncratic volatility on bid probability. The table contains the results of the regression that tests the impact of a firm'sidiosyncratic volatility on the probability of the firm receiving a takeover bid in the emerging country. We include both acquiredfirms and non-acquired firms in the analysis. The dependent variable is a firm-year observation, Pi,t, which equals 1 if firm i receivesan acquisition bid from another firm in year t, and equals 0 if we do not observe any acquisition bid for firm i in the sample year t. Thesampling period is from year 1990 to 2007. Our key independent variable is idiosyncratic volatility of the target firm (θ) and (ψ). Wetest these two measures in model 1 andmodel 2 respectively. We control for target firm characteristics (including firm size, leverage,growth potential, R&D to sales, firm age) and industry and country level variables (such as financial market development, industryHerfindahl index, per capita GDP, rule of law index, anti-self-dealing index, disclosure index, creditor right index, target countrycorruption). We also include the industry dummy variables and year dummy variables in the models. We use firm fixed effect modelto estimate the regression coefficients. All standard errors are robust for heteroscadasticity and autocorrelation.

Dependent variable: Acquisition probability Control model Model 1 Model 2

Intercept −3.569a −2.950a −3.618a

(0.843) (0.860) (0.844)Idiosyncratic volatility (θ) −0.719a

(0.190)Idiosyncratic volatility (ψ) −0.038a

(0.010)Target firm size 0.069a 0.065a 0.062a

(0.011) (0.011) (0.011)Target firm leverage 0.504a 0.554a 0.533a

(0.134) (0.135) (0.135)Target firm market-to-book 0.051a 0.048a 0.049a

(0.007) (0.007) (0.007)Target firm R&D to sales 0.179b 0.128 0.154

(1.434) (1.468) (1.45)Target firm age 0.004 0.005 0.005

(0.006) (0.006) (0.006)Target country financial market development 0.001 0.002c 0.001

(0.000) (0.000) (0.000)Target industry Herfindahl index 0.133 0.171 0.177

(0.129) (0.130) (0.130)Target country per capita GDP 0.150a 0.150a 0.169a

(0.056) (0.057) (0.057)Target country rule of law 0.599a 0.598a 0.546a

(0.130) (0.131) (0.132)Target country anti-self-dealing index 2.057a 1.949a 1.986a

(0.375) (0.375) (0.375)Target country disclosure index −0.128a −0.137a −0.141a

(0.027) (0.027) (0.027)Target country creditor right index −0.566a −0.501a −0.493a

(0.091) (0.093) (0.094)Target country anti-corruption index −0.444a −0.436a −0.423a

(0.072) (0.072) (0.073)Industry dummy Yes Yes YesYear dummy Yes Yes YesPseudo-R2 0.128 0.129 0.129N 36,934 36,934 36,934

a 1% significance.b 5% significance.c 10% significance.

Table 8Robustness check for correcting the sample selection bias. The table shows the results of the robustness test aimed at correcting forsample selection bias in the completed deals sample of emerging country firms. We re-estimate three different models in thisanalysis. The first model is based on acquisition premium, the second model is for the propensity of making majority controlacquisitions, and the last model is for the percentage ownership acquired in the target firm. The dependent variable in all theregressions is idiosyncratic volatility (θ). The results are also robust if we use the idiosyncratic volatility (ψ). We follow Heckman(1979) to correct the sample selection bias by introducing the Inverse Mill's ratio based on the estimated target firm probability froma probit regression. We then control for this estimated target firm probability (i.e., Inverse Mill's ratio) in the models. We account forthe effect of deal characteristics (such as tender offers and industry relatedness), target firm characteristics (including firm size,leverage, growth potential, R&D to sales, firm age) and country level variables (financial market development, industry Herfindahlindex, per capita GDP, rule of law index, anti-self-dealing index, disclosure index, creditor right index, target country corruption).We also control for the public status of the acquiring firms, deal size and cash payment and the existence of competing bidders andtransaction size. Finally, we include industry, country and year dummy variables in the models. All standard errors are robust forheteroscadasticity and autocorrelation.

Dependent variables Acquisition premium Majority control acquisition % of ownership acquired

Intercept 13.908 −2.377 6.594(32.51) (2.616) (5.487)

Inverse Mill's ratio −0.022 −7.568 −0.527(2.692) (5.183) (0.454)

Idiosyncratic volatility (θ) 1.165a 2.746a 0.156b

(0.381) (0.773) (0.064)Target firm size −0.370a −0.192a −0.037a

(0.045) (0.058) (0.007)Target firm leverage 0.006 0.009 −0.017b

(0.040) (0.075) (0.006)Target firm market-to-book −0.100b 0.069 0.007

(0.048) (0.067) (0.008)Target firm R&D to sales 0.022 −0.091 −0.004

(0.022) (0.199) (0.003)Target firm age 0.009 0.030 0.004c

(0.013) (0.023) (0.002)Target country financial market development −0.161 −1.052b −0.008

(0.249) (0.441) (0.042)Target industry Herfindahl index −0.109 0.141 0.017

(0.293) (0.542) (0.049)Target country per capita GDP −1.034 −0.510b −0.283

(1.031) (0.260) (0.174)Target country rule of law 2.854 −0.918 1.167

(7.042) (0.605) (1.188)Target country anti-self-dealing index −5.792 −1.705 −2.493

(12.97) (1.455) (2.190)Target country disclosure index −0.774 0.054 −0.383

(2.480) (0.108) (0.418)Target country creditor right index 3.399 0.594 1.300

(7.677) (0.381) (1.295)Target country anti-corruption index −1.482 0.863a −0.526

(3.365) (0.328) (0.567)Competing bids 0.627 0.543 −0.015

(0.606) (1.149) (0.102)Public acquirers 0.222c 0.159 0.064a

(0.120) (0.234) (0.020)Tender offers 0.281c 0.423 0.325a

(0.154) (0.261) (0.026)Industry relatedness (4SIC) −0.102 0.239 0.006

(0.137) (0.247) (0.023)Deal size 0.449a 0.570a 0.057a

(0.038) (0.083) (0.006)Cash payment −0.539b −0.892b −0.235a

(0.219) (0.382) (0.037)Industry dummy Yes Yes YesCountry dummy Yes Yes YesYear dummy Yes Yes YesR square 0.282 0.222 0.500

42 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

Table 8 (continued)

Dependent variables Acquisition premium Majority control acquisition % of ownership acquired

F statistics 4.378 – 11.130N 742 742 742

a 1% significance.b 5% significance.c 10% significance.

43P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

the variance of the market model residuals — in testing the main models. The results, which are availableupon request, are similar to what we reported earlier.

Second, a challenge that characterizes the estimation and analysis of idiosyncratic volatility for firms inemerging markets is thin trading. Although our initial mergers and acquisitions transactions in the twentyemerging countries is about 8041 observations, as a result of missing daily stock return data for many ofthe firms in the Datastream database our final sample reduced to 742 observations. In our main analysis,we followed Bartram et al. (2012) and used firms that have less than 30%missing values in the daily returndata to calculate the idiosyncratic volatility. To ensure that our results are not driven by this selectionprocedure, we relaxed the selection threshold to 20% and 10% missing information in the daily stockreturns and re-estimated our main regressions but the results remain the same.

5.6.2. Sample selection biasExtant literature on takeovers in developed market shows that firms that takeover targets have

substantially different characteristics than firms that are not acquired (e.g., Palepu, 1986). Thus, it ispossible that acquiring firms in our sample pay a high premium under information asymmetry because ofthe distinct characteristics of the target firms in emerging markets. Ignoring the not-acquired firms couldintroduce sample selection bias, which can affect the overall conclusions of the study. We followHeckman's (1979) procedure to control for the sample selection bias. We obtained data on all publiclylisted firms in the twenty emerging countries over the sampling period (1990 to 2007) from theDataStream database. We then estimated a probit regression using the same dataset as the one used in theacquisition probability models. The dependent variable in this regression is a binary variable which equalsone if the firm was acquired in a sample year, and zero otherwise. We include the same firm and countrycharacteristics as specified in the acquisition probability models reported in Table 7 as independentvariables. These variables have also been found to be important characteristics of takeover targets (seePalepu, 1986; Rossi and Volpin, 2004). Following Heckman's method, we compute the “Inverse Mill'sratio” (IMR) for each target firm based on the estimated probit model and then include the IMR as anadditional control variable in our main regression models (particularly the acquisition premium,transaction size, and percentage ownership acquired models). This procedure effectively corrects thepotential effects of sample selection bias.

Table 8 presents the results of the regressions corrected for sample selection bias. The IMR variable issignificant in some of the models, suggesting that sample selection bias does exist in our dataset.Comparing the results in Table 8 with those presented in Tables 3 and 6, we find consistent results foridiosyncratic volatility (θ). We confirm the early findings that idiosyncratic volatility has a positive andsignificant impact on acquisition premium. We also show that acquiring firms tend to acquire moreownership in the target firms and gain majority control of the firms in the face of higher idiosyncraticvolatility. We tested the models on the alternative idiosyncratic volatility measure (ψ) and found similarresults, but for the sake of brevity, we do not show these results in the paper.

5.6.3. Simultaneous relationship in deal arrangementsIn the previous sections, we demonstrate that the degree of idiosyncratic volatility is positively related to

acquisition premium. We also show that idiosyncratic volatility affects the degree of control acquired.However, some acquisition outcomes may be interrelated with each other. For example, the acquisitionpremium could depend on the desire to acquire majority. Higher idiosyncratic volatility results in higherprobability of the bidder acquiring majority control ownership in the target firm (as we have documented).However, gaining controlling ownershipmay also require paying higher premium to the target shareholders.

Table 9Robustness check for interdependence of deal arrangements. The table presents the results of the regression that tests the effect ofidiosyncratic volatility on acquisition premium and majority control by considering the simultaneous relationship between the twoacquisition outcomes. Equation 1 models the simultaneous impact of majority control on the acquisition premium, and equation 2models the simultaneous impact of acquisition premium on majority control. The key independent variable in the respective modelsis the idiosyncratic volatility (θ). We control for the other deal characteristics (such as deal size, cash payment, tender offers andindustry relatedness), target firm characteristics (including firm size, leverage, growth potential, R&D to sales, firm age) and countrylevel variables (financial market development, industry Herfindahl index, per capita GDP, rule of law index, anti-self-dealing index,disclosure index, creditor right index, target country corruption). We also control for the status of the acquiring firms and theexistence of competing bidders. All standard errors are robust for heteroscadasticity and autocorrelation.

Simultaneous equation model Equation 1: Acquisition premium Equation 2: Majority control acquisition

Intercept −2.982a 0.464b

(0.924) (0.248)Majority control acquisition 2.086a

(0.127)Acquisition premium 0.149a

(0.009)Idiosyncratic volatility (θ) 1.108a 0.113

(0.323) (0.087)Target firm size −0.074a −0.009

(0.026) (0.007)Target firm leverage 0.010 −0.006

(0.037) (0.009)Target firm market-to-book −0.033 0.012

(0.033) (0.008)Target firm R&D to sales 0.022 −0.005

(0.021) (0.005)Target firm age −0.015 0.010a

(0.010) (0.002)Target country financial market development −0.175 −0.056

(0.155) (0.041)Target industry Herfindahl index 0.327 −0.076

(0.235) (0.062)Target country per capita GDP −0.277c 0.012

(0.119) (0.032)Target country rule of law 0.242 −0.146c

(0.271) (0.072)Target country anti-self-dealing index 0.210 −0.287

(0.762) (0.203)Target country disclosure index 0.059 −0.001

(0.053) (0.014)Target country creditor right index 0.126 0.058

(0.196) (0.052)Target country anti-corruption index −0.033 0.083c

(0.145) (0.038)Competing bids 0.454 −0.022

(0.568) (0.151)Public acquirers 0.149 −0.020

(0.112) (0.030)Tender offers 0.068 0.060

(0.145) (0.038)Industry relatedness (4SIC) −0.099 0.051

(0.127) (0.033)Deal size 0.196a 0.019c

(0.033) (0.009)Cash payment −0.124 −0.103b

(0.210) (0.056)R square 0.212 0.207N 742 742

a 1% significance.b 10% significance.c 5% significance.

44 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

45P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

Thus, the positive relation between idiosyncratic volatility and acquisition premiummight be confounded bymajority control acquisitions. To address this potential dependency issue, we run a simultaneous regressionmodel with both majority control and acquisition premium as simultaneously determined variables in thesystem of equations. The results of the regression, which we report in Table 9, indicate that the majoritycontrol variable has a highly positive and significant impact on acquisition premium. The finding supports ourconjecture that these deal three variables are highly interrelated. More interestingly, we find that theidiosyncratic volatility measures still remain positive and highly significant in explaining the acquisitionpremium even after controlling for other deal characteristics (βθ = 1.108, significant at 1% level), results thatare consistent with the private information and informed trading interpretation of the measure.

In the second equation, we find that the impact of idiosyncratic volatility on majority controlacquisitions is not significant anymore after modeling the simultaneous relationship with the acquisitionpremium. The results suggest that acquisition premium and acquiring majority control in the target firmsmight indeed be simultaneously determined. Acquirers in emergingmarkets pay higher premium to targetfirms with higher idiosyncratic volatility not simply because they are acquiring controlling ownership butalso perhaps because of the more informative nature of the stock price. Overall, the results confirm theprivate information and stock price informativeness explanation of the idiosyncratic volatility in emergingmarkets.

6. Conclusions

In this paper we investigate the impact of idiosyncratic volatility on various key parameters intakeovers, including acquisition premium, transaction size, controlling ownership acquired, dealcompletion rate, completion time and acquisition probability. Using a sample of 742 completedacquisitions in twenty emerging countries during the period 1990 to 2007, we find that acquisitionpremium is positively related to idiosyncratic volatility of the target firms. We also find that the degree ofcontrol acquired is positively related to idiosyncratic volatility. After controlling for the simultaneousimpact of idiosyncratic volatility on acquisition premium and majority control measure, we find thatidiosyncratic volatility still influences these parameters. Our analysis also shows that target firm'sidiosyncratic volatility is positively related to acquisition completion rate, and negatively related to dealcompletion time and the probability that the target will receive acquisition bid. Our results relating to theidiosyncratic volatility and bid probability are consistent with the information efficiency and informedtrading hypothesis, i.e., when a firm's stock price reflects its fundamental value, the cost of collectingadditional private information increases. The more private information that is reflected in the target firm'sshare price, the less information advantage held by others and the lower the likelihood that the firm willbe the target of takeover, all things being equal.

Put together, our results indicate that firms with higher idiosyncratic volatility in emerging markets areless likely to be taken over, but once bidders decide to take over such firms, they pay higher premiums.Acquirers that are likely to pay higher premiums are those that are informationally disadvantaged (i.e.,bidders from emerging markets, bidders who make unrelated acquisitions, and those that have lessexperience in the target country). Our results are robust with respect to the choice of idiosyncraticvolatility proxies, estimation period and sample selection. Overall, our results support the informedtrading and private information explanation of idiosyncratic volatility.

Some interesting extensions to our work can be considered for future research. First, our study focusesonly on idiosyncratic risk and its impact on acquisition decisions variables. It would be instructive toextend the study to the post merger period to examine how idiosyncratic volatility interacts with theacquisition variables to impact on the long-term performance and changes in the informationenvironment of the combined firm, especially when the target firm is of a comparable size to theacquiring firm. Second, it may be interesting to test whether the relative valuation differences between theacquiring firm and the target firm's market plays a role in the timing and the amount of premium and theidiosyncratic volatility of the acquiring firm. Third, it is possible that the private information role reflectedin idiosyncratic risk of targets in emerging markets could also be found in developedmarkets. Future studycan expand the analysis and test the findings in a large sample of firms engaged in takeover activities indeveloped countries.

46 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

Appendix A. Variable measurement and data source

Variable name Measurement Data sources

Idiosyncratic volatility (θ) Variance of the residuals in the market modeldivided by the total variance of the daily stockreturns of the target firm

Datastream database

Idiosyncratic volatility (ψ) Logarithm of (1 − R2) / R2 estimated based on themarket model of the target firm stock returns

Datastream database

Acquisition premium Difference between the deal offer price and thetarget stock price in 4 weeks before theacquisition; alternative measure is based on theCAR (−63. +126) of the market model of thetarget firm stock return (Schwert, 2000)

SDC Platinum database orcalculated from theDatastream database

Deal size Transaction value of the acquisition deal in millionUSD

SDC Platinum database

Majority control acquisition Dummy variable which equals 1 if the acquiringfilm acquires more than 50% ownership in thetarget film and otherwise equals 0

SDC Platinum database

Percentage ownership acquired Percentage of target firm ownership acquired bythe acquiring firm

SDC Platinum database

Cash payment Dummy variable which equals if the acquisition ispaid in pure cash; it equals 0 for pure stockpayment or mixed payment

SDC Platinum database

Deal completion time Elapsed time (in years) between the acquisitionannouncement date and the completion date

SDC Platinum database

Target firm size Log of total revenue of the target firm in the fiscalyear before the acquisition year

Worldscope database

Target firm leverage Total debt to assets ratio of the target firm in thefiscal year before the acquisition year

Worldscope database

Target firm market-to-book Market to book ratio of the target firm in the fiscalyear before the acquisition year

Worldscope database

Target firm R&D to sales R&D expense to sales ratio of the target firm in thefiscal year before the acquisition year

Worldscope database

Target firm age Number of years since the initial listing year inDatastream to the acquisition year

Worldscope database

Target industry Herfindahl index Herfindahl index of the target industry calculatedbased on the market share of each firm in theindustry in the year before the acquisition

Worldscope database

Target country financial market development Total capitalization of the target country's stockmarket dividend by the target country GDP in theyear before the acquisition

World DevelopmentIndicator database

Target country per capita GDP Target country per capita GDP in the year beforethe acquisition

World DevelopmentIndicator database

Target country rule of law Rule of law measure of the target country World Bank databaseTarget country anti-self-dealing index Anti-self-dealing index of the target country Djankov et al. (2008)Target country disclosure index Average of the disclosure index of the target

country between 2000 and 2009World Bank database

Target country creditor right index Average of the target country's creditor right index Djankov et al. (2007)Target country anti-corruption index Average of the target country's corruption

perception indexTransparencyInternational database

Competing bids Dummy variable which equals 1 if there is anycompeting bid in the acquisition, otherwise 0

SDC Platinum database

Public acquirers Dummy variable which equals 1 if the acquiringfirm is publicly listed, otherwise 0

SDC Platinum database

Tender offers Dummy variable which equals 1 if the acquisitionis a tender offer, otherwise 0

SDC Platinum database

Industry relatedness (4SIC) Dummy variable which equals 1 if the acquiringindustry and the target industry have the same4-digit SIC code, otherwise 0

SDC Platinum database

Cross-border acquisitions Dummy variable which equals 1 if the acquisitioninvolves a foreign acquirer, it equals 0 if theacquiring firm is a domestic firm

SDC Platinum database

(continued on next page)

(continued)

Variable name Measurement Data sources

Developed country acquirers Dummy variable which equals 1 if the acquirercomes from a developed country (defined as highincome country by theWorld Bank) and otherwise 0

SDC Platinum database

Acquires acquisition experience Count of number of prior acquisitions made by theacquiring film in the target country before thecurrent acquisition

Calculated from the SDCPlatinum database

Appendix A (continued)

47P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

References

Ashbaugh-Skaife, H., Gassen, J., LaFond, R., 2006. Does stock price synchronicity represent firm-specific information? Theinternational evidence. Working Paper, Columbia University.

Ball, R., 2001. Infrastructure requirements for an economically efficient system of public financial reporting and disclosure. Brook.Pap. Financ. Serv. 2001, 127–182.

Bargeron, L., Schlingemann, F., Stulz, R., Zutter, C., 2008. Why do private acquirers pay so little compared to public acquirers? J.Financ. Econ. 89, 375–390.

Bartram, S., Brown, G.W., Stulz, R.M., 2012. Why do foreign firms have less idiosyncratic risk than U.S. firms? J. Financ. 67,1329–1370.

Bertrand, M., Mehta, P., Mullainathan, S., 2002. Ferreting out tunneling: an application to Indian business groups. Q. J. Econ. 117 (1),121–148.

Betton, S., Eckbo, B., 2000. Toeholds, bid jumps, and expected payoffs in takeovers. Rev. Financ. Stud. 13 (4), 841–882.Bhagat, S., Malhotra, S., Zhu, P., 2011. Emerging country cross-border acquisitions: characteristics, acquirer returns and cross-

sectional determinants. Emerg. Mark. Rev. 12 (3), 250–271.Chan, K., Hameed, A., 2006. Stock price synchronicity and analyst coverage in emerging markets. J. Financ. Econ. 80, 115–147.Chari, A., Ouimet, P.P., Tesar, L.L., 2010. The value of control in emerging markets. Rev. Financ. Stud. 23 (4), 1741–1770.Chatterjee, S., John, K., Yan, A., 2012. Takeovers and divergence of investor opinion. Rev. Financ. Stud. 25 (1), 227–277.Chen, Q., Goldstein, I., Jiang, W., 2007. Price informativeness and investment sensitivity to stock prices. Rev. Financ. Stud. 20 (3),

619–650.Chi, J., Sun, Q., Young, M., 2011. Performance and characteristics of acquiring firms in the Chinese stock markets. Emerg. Mark. Rev.

12 (2), 152–170.Comin, D., Philippon, T., 2005. The rise in firm-level volatility: causes and consequences. NBER Macroecon. Annu. 20, 167–201.Dasgupta, S., Gan, J., Gao, N., 2010. Transparency, price informativeness, and stock return synchronicity: theory and evidence. J.

Financ. Quant. Anal. 45, 1189–1220.Djankov, S., McLiesh, C., Shleifer, A., 2007. Private credit in 120 countries. J. Financ. Econ. 84, 299–329.Djankov, S., La Porta, R., Lopez-de-Silanes, F., Shleifer, A., 2008. The law and economics of selfdealing. J. Financ. Econ. 88, 430–465.Durnev, A., Morck, R., Yeung, B., 2003. Value enhancing capital budgeting and firm-specific stock returns variation. J. Financ. 59,

65–106.Durnev, A., Morck, R., Yeung, B., Zarowin, P., 2004. Does greater firm-specific return variation mean more or less informed stock

pricing? J. Account. Res. 41, 797–836.Fan, J., Wong, T.J., 2005. Do external auditors perform a corporate governance role in emerging markets? Evidence from East Asia. J.

Account. Res. 43, 35–72.Ferreira, M.A., Laux, P.A., 2007. Corporate governance, idiosyncratic risk, and information flow. J. Financ. 62, 951–989.Fu, F., Lin, L., Officer, M.S., 2013. Acquisitions driven by stock overvaluation: are they good deals? J. Financ. Econ. 109 (1), 24–39.Griffin, J.M., Kelly, P.J., Nadari, F., 2010. Do market efficiency measures yield correct inferences? A comparison of developed and

emerging markets. Rev. Financ. Stud. 23, 3225–3277.Gugler, K., Mueller, D.C., Yurtoglu, B.B., Zulehner, C., 2003. The effects of mergers: an international comparison. Int. J. Ind. Organ. 21,

625–653.Gul, F.A., Kim, J.B., Qiu, A., 2010. Ownership concentration, foreign shareholding, audit quality, and stock price synchronicity:

evidence from China. J. Financ. Econ. 95, 425–442.Heckman, J.J., 1979. Sample selection bias as a specification error. Econometrica 47 (1), 53–161.Huang, B., Wald, J., Martell, R., 2013. Financial market liberalization and the pricing of idiosyncratic risk. Emerg. Mark. Rev. 17, 44–59.World Investment Report, 2013. Global Value Chains: Investment and Trade for Development United Nations Conference on Trade

and Development (UNCTAD) 978-92-1-112868-0.Irvine, P.J., Pontiff, J., 2005. Idiosyncratic return volatility, cash flows, and product market competition. Working Paper. Boston

College, MA. unpublished paper.Jarrell, G.A., Brickley, J.A., Netter, J.M., 1988. The market for corporate control: the empirical evidence since 1980. J. Econ. Perspect. 2,

49–68.Jiang, G., Xu, D., Yao, T., 2009. The information content of idiosyncratic volatility. J. Financ. Quant. Anal. 44, 1–28.Jin, L., Myers, S., 2006. R2 around the world: new theory and new tests. J. Financ. 79, 257–292.Johnson, S., La Porta, R., Lopez-de-Silanes, F., Shleifer, A., 2000. Tunneling. Am. Econ. Rev. Pap. Proc. 90, 22–27.Karolyi, A., 2012. Corporate governance, agency problems and international cross-listings: a defense of the bonding hypothesis.

Emerg. Mark. Rev. 13 (4), 516–547.

48 P. Zhu et al. / Emerging Markets Review 19 (2014) 18–48

Kaufman, D., Kraay, A., Mastruzzi, M., 2007. Governance matters VI: governance indicators for 1996–2006. World Bank PolicyResearch Working Paper No. 4280.

Kelly, P.J., 2007. Information efficiency and firm-specific return variation. Working Paper, University of South Florida.Khanna, T., Thomas, C., 2009. Synchronicity and firm interlocks in an emerging market. J. Financ. Econ. 92 (2), 182–204.Kim, J.B., Yi, C.H., 2006. Ownership structure, business group affiliation, listing status, and earnings management: evidence from

Korea. Contemp. Account. Res. 23, 427–464.Kohli, R., Jit, B., Mann, S., 2013. Analyzing the likelihood and the impact of earnout offers on acquiring company wealth gains in India.

Emerg. Mark. Rev. 16, 203–222.Laamanen, T., 2007. Research notes and commentaries on the role of acquisition premium in acquisition research. Strateg. Manag. J.

28, 1359–1369.Lee, D.W., Liu, M.H., 2011. Does more information in stock price lead to greater or smaller idiosyncratic return volatility? J. Bank.

Financ. 35 (6), 1563–1580.Martynova, M., Renneboog, L., 2008. Spillover of corporate governance standards in cross-border mergers and acquisitions. J. Corp.

Financ. 14 (3), 200–223.Moeller, T., 2005. Let's make a deal! How shareholder control impacts merger payoffs. J. Financ. Econ. 76 (1), 167–190.Moeller, S., Schlingemann, F., Stulz, R., 2007. How do diversity of opinion and information asymmetry affect acquirer returns? Rev.

Financ. Stud. 22 (6), 2047–2078.Morck, R., Shleifer, A., Vishny, R.W., 1990. Do managerial objectives drive bad acquisitions? J. Financ. 45 (1), 31–48.Morck, R., Yeung, B., Yu, W., 2000. The information content of stock markets: why do emerging markets have synchronous stock

price movements? J. Financ. Econ. 58, 215–260.Nagano, M., 2013. Similarities and differences among cross-border M&A and greenfield FDI determinants: evidence from Asia and

Oceania. Emerg. Mark. Rev. 16, 100–118.Nartea, G., Ward, B., Yao, L., 2011. Idiosyncratic volatility and cross-sectional stock returns in Southeast Asian stock markets. Account.

Financ. 51 (4), 1031–1054.Officer, M., Pulsen, A., Stegemoller, M., 2009. Target-firm information asymmetry and acquirer returns. Eur. Finan. Rev. 13 (3),

467–493.Palepu, K.G., 1986. Predicting takeover targets, a methodological and empirical analysis. J. Account. Econ. 8 (1), 3–35.Pastor, L., Veronesi, P., 2003. Stock valuation and learning about profitability. J. Financ. 58, 1749–1790.Rajgopal, S., Venkatachalam, M., 2011. Financial reporting quality and idiosyncratic return volatility. J. Account. Econ. 51, 1–20.Rossi, S., Volpin, P., 2004. Cross-country determinants of mergers and acquisitions. J. Financ. Econ. 74 (2), 277–304.Schwert, G.W., 2000. Hostility in takeovers: in the eyes of the beholder? J. Financ. 55 (6), 2599–2640.Teoh, S.H., Yang, Y., Zhang, Y., 2007. R-squared: noise or firm-specific information. Working Paper, University of California, Irvine.Weitzel, U., Berns, S., 2006. Cross-border takeovers, corruption, and related aspects of governance. J. Int. Bus. Stud. 37 (6), 786–806.World Investment Report, 2011. Non-equity Modes of International Production and Development United Nations Conference on

Trade and Development (UNCTAD) 978-92-1-112828-4.Wurgler, J., 2000. Financial markets and the allocation of capital. J. Financ. Econ. 58, 187–214.Xing, X., Anderson, R., 2011. Stock price synchronicity and public firm-specific information. J. Financ. Mark. 14 (2), 259–276.Zhu, P., Jog, V., 2012. Impact on target firm risk-return characteristics of domestic and cross-border mergers and acquisitions in

emerging markets. Emerg. Mark. Financ. Trade 48 (4), 79–101.Zhu, P., Jog, V., Otchere, I., 2011. Partial acquisition in emerging markets: a test of the strategic market entry and corporate control

hypotheses. J. Corp. Financ. 17, 288–305.