Does investor sentiment impact premiums in M&A transactions?

Hervé, Fabrice; Rouine, Ibtissem; Thraya, Mohamed-Firas; Zouaoui, Mohamed

doi:https://doi.org/10.59876/a-hy8x-rmwg

There is extensive literature discussing the determinants of the premium paid to acquire potential targets (Hayward and Hambrick, 1997; Eckbo, 2009; Jost et al., 2022). If investors are rational, in line with the expectations of the efficient markets hypothesis (Fama, 1970), the bid premium should reflect the value of potential synergies. Those Synergies may arise in M&A transactions for different raisons such as the gain from the increased efficiency in the form of improved target asset utilization, management efficiency or from wealth transfers as a result of the business combination. This paper relaxes the view of strict investor rationality and consider that investors’ cognitive biases can influence the bid premium. Investor sentiment is expected to influence investor perception of potential synergies and risks involved in the acquisition, leading to a non-rational component of the bid premium (Danbolt et al., 2015). Almost no scholarly attention has been devoted to a direct accurate proxy of individual investor sentiment in the context of M&A. This article contributes to the existing literature by empirically examining the effect of investor sentiment, proxied by the non-rational part of the consumer confidence index (CCI), on bid premium using an international sample of M&A data.

The offer price is often affected by optimistic views and beliefs about the expected synergies of the deal. Individuals are indeed prone to sentiment and make decisions based on non-rational inputs. Sentiment can be viewed as non-rational beliefs that are not consistent with a fundamental approach. Evidence of the influence of sentiment on the financial markets has been gathered (see for instance, in the context of the stock market: Baker and Wurgler, 2007; for bond markets: Piñeiro-Chousa et al., 2021; for crypto-assets: López-Cabarcos et al., 2021). Thus, sentiment has something to do with the pricing of securities and the valuation of companies. Danso et al. (2019) find that sentiment influences firms’ investment decisions in the United States. They underline that there is a large literature providing evidence of the effect of behavioral biases on investing and financing decisions made by CEOs. They show, using a survey-based measure of sentiment, namely the Michigan Consumer Confidence Index, that sentiment has a positive effect on firms’ capital expenditures.

This is of particular importance in the context of M&A, as the premium paid may be influenced by investor sentiment. Some form of mispricing the initial bid price may impede the acquirer’s future performance. Some theoretical papers study the issue of investor sentiment in the context of M&A (Shleifer and Vishny, 2003; Rhodes-Kropf and Viswanathan, 2004). The studies generally use market valuation as a proxy for investor sentiment and do not explicitly address the direct influence of sentiment on the premium paid by the bidder. Bouwman et al. (2009) examine the performance of acquirers in high-valuation markets and low-valuation markets. They find that acquisitions announced in a booming market generate high short-run performance but underperform in the long run. Baker et al. (2009) find that foreign direct investment, which often consists of cross-border M&A, is positively associated with the current aggregate of market-to book ratio of the home country’s stock market (i.e. high investor sentiment) and is inversely related to subsequent market returns.

Otherwise, some studies suggest that media news contains information relevant to M&A performance (Yang et al., 2019; Liao et al., 2021). These studies examine the effect of financial media sentiment on deal premium paid to target firm and find conflicting results. Liao et al. (2021) show that acquirers with high media optimism tend to pay a higher premium to target firms. They argue that those acquirers either have better growth opportunities or can pay for the target with more overvalued shares. However, Yang et al. (2019) find the opposite result showing that a more pessimistic media attitude leads to higher bid premiums. They argue that when the shareholders of target firms are influenced by pessimistic news, acquirers have to boost their bid prices to compensate them. We further develop those findings by relying on a robust; longstanding and established measure of investor sentiment based on the consumer confidence index (Lemmon and Portniaguina, 2006; Danso et al., 2019; Wang et al., 2021). This index has the advantages of being common to many countries allowing for international comparison and, having a long track record that allows testing our sentiment hypothesis over a long period of time.

The heterogeneity and diversity of countries of our sample are particularly propitious to an extension of the previous analyses and allow for investigation of the influence of institutional factors and national culture on the relationship between investor sentiment and bid premium. Culture and institutions tend to moderate the impact of investor sentiment on financial behaviors (La porta et al., 1998. Schmeling, 2009; Wang et al., 2021). We provide evidence of the moderating effects of national culture and of market integrity on the association between investor sentiment and the bid premium.

Our results show that investor sentiment exerts a strong and positive effect on M&A premium, i.e., the higher the sentiment surrounding the acquirer, the higher the premium paid by the bidder. In line with the literature, we find that the effect of the sentiment is higher in countries with a culture prone to overreaction and herding. Moreover, we show that the effect of sentiment is exacerbated in countries with low market integrity and weak institutions. To extend our behavioral story, we examine serial bidding. If CEOs behave in the same way as retail or non-financially educated investors, we should observe no learning effect, i.e., premiums paid by serial bidders inflate with both investor sentiment and transactions. If executives are learning from their experience, the impact of sentiment should be smaller over time and transactions. Our results show that managers benefit from their experience and are less influenced by investor sentiment as they conduct M&A transactions.

Our paper contributes to the literature in several ways. First, our paper contributes to the emerging literature on the determinants of bid premiums (Eckbo, 2009; Jost et al., 2022). We supplement these studies by identifying a new determinant of the bid premium in M&A transactions consistent with a behavioral approach to financing decisions. This adds to the behavioral finance literature that examines the impact of investor sentiment on firms’ decisions (Bergman and Roychowdhury, 2008; Liao et al., 2021). Second, the introduction of culture and institutions as moderators of the impact of the investor sentiment on premiums provides a better understanding of corporate decisions. Third, our paper contributes to the literature on CEO behavior (Huang et al., 2022; Aktas et al., 2011) by showing that while CEOs are subject to investor sentiment, a learning behavior counteracts this tendency to be led by instincts. Last, our study expands the growing literature examining the effect of country-level factors, such as trust (Ahern et al., 2015), culture (Stahl and Voigt, 2008), leadership style (Rouine, 2018) and economic policy uncertainty (Nguyen and Phan, 2017), on bidder’s decisions making.

Our paper is structured as follows. Section 2 develops our hypotheses. Section 3 details our data and our variables. Section 4 presents our empirical strategy. Section 5 details and discusses our results. Section 6 provides some robustness checks. The last section concludes.

Hypotheses development

Bid premium and investor sentiment

M&A are one of the most important decisions companies have to make. As Hayward and Hambrick (1997) point out, a large part of the corporate finance literature offers rational explanations for these decisions (poor target company management, empire building, and synergies). Another strand of the literature, based on the concept of hubris (exaggerated pride), has emerged by Roll (1986). He argues that firms prone to hubris pay too much for targets because managers overestimate their own ability to realize value from non-organic growth operations.

Along similar lines, but from a behavioral perspective, Malmendier and Tate (2005, 2008) show that CEOs’ behavioral characteristics influence investment decisions in the context of M&A. Baker and Wurgler (2007) define investor sentiment as “a belief about future cash flows and investment risks that is not justified by the facts at hand.” Jiang et al. (2019) argue that CEOs’ decisions may be guided by their unrealistic beliefs about fundamentals, i.e., their sentiment. Dicks and Fulghieri (2021) develop a theoretical model that predicts a positive relationship between investor sentiment and firm valuations. Thus, CEO can be influenced by the general investor sentiment surrounding their decisions. Those authors apprehend sentiment as the “investor attitude toward investment in risky assets.” Conducting an M&A transaction is an investment in a risky asset. This line of reasoning suggests that investor sentiment can play a role in understanding M&A activity.

Huang et al. (2022) show that investor sentiment influences the decision to exercise stock options and the decision to engage in M&A activity. CEO exercise options more quickly during periods of high sentiment. They find that firms engage more often in M&A transactions when sentiment is high. Danbolt et al. (2015) highlight a positive (negative) market reaction to M&A announcements during periods characterized by high optimism (pessimism). The authors argue that investor sentiment influences their perception of the potential synergies and risks of M&A. During periods of high enthusiasm, investors tend to overestimate the positive benefits of M&A and underestimate the risks associated with this type of transaction. We can reasonably expect CEOs to propose higher bids during periods of high sentiment. This leads us to formulate the following hypothesis:

H1: The bid premium is positively associated with investor sentiment

National culture

Academic research did not deeply explore the role of national culture in determining the financial behavior of individuals. The pioneer model of the link between national culture and financial markets is that of Hofstede (1980). This author defines culture as “the collective programming of the mind distinguishing the members of one group or category of people from others”. Hofstede (2001) argues that culture can be captured by six dimensions: power distance, individualism, uncertainty avoidance, masculinity, long-term orientation and indulgence. While there are critiques of Hofstede’s framework (McSweeney, 2002; Ailon, 2008), it is the most comprehensive approach with national culture, as evidenced by its cumulative impact on research generally and in finance specifically. Reuter (2011), in his review of the literature on culture and finance, finds that over 80% (24/29) of the articles considering a dimensional approach to culture in finance rely on the framework developed by Hofstede. Similarly, Beugelsdijk et al. (2015) point out that Hofstede’s approach dominates quantitative research in strategic management and that Hofstede’s individual dimension scores are not outdated and are rather stable over time. Furthermore, Kaasa (2021) finds high correlations between Hofstede’s cultural items and other items developed by other authors (Schwartz,1994; Inglehart, 1997). This leads us to use Hofstede’s approach in this work. In the financial literature, three dimensions are particularly retained in empirical studies working on the link between culture and financial markets: individualism, uncertainty avoidance and masculinity (Chui et al., 2010; Schmeling, 2009; Lucey and Zhang, 2010). Data on cultural dimensions are collected from Hofstede et al. (2001)[1].

Individualism determines the extent to which the identity is defined by personal choices or by the collective. In countries where the degree of individualism is high, people care more about themselves than others. On the contrary, countries with a high level of collectivism relate to societies where individuals are well integrated into a strong, cohesive group and in which consensus is the rule. According to Hofstede (1980), a high level of collectivism is synonymous with a tendency to herd behavior. This herding behavior can be interpreted as the result of correlated behavioral tendencies between individuals. The actions of noise traders are correlated because they replicate the actions of other traders based on overly optimistic or pessimistic expectations. This propensity to invest with the group is exactly what is supposed to guide the relationship between investor sentiment and stock return. As a result, we can expect that the impact of investor sentiment on bid premium will be weaker in countries where the degree of individualism is high.

The uncertainty avoidance represents the most appropriate cultural dimension to justify a behavioral explanation of stock returns (Lucey and Zhang, 2010). According to Hofstede (1980), uncertainty avoidance can be defined by the degree to which members of a culture can react to equivocal, unknown and unexpected situations. In countries with higher uncertainty avoidance index, individuals prefer predictable situations, are wary of risky situations and are more emotional than in countries with lower uncertainty avoidance index. The uncertainty avoidance is thus used in the financial literature as a proxy to the tendency of individuals to overreact. As a result, we can expect that the impact of investor sentiment on bid premium will be stronger in countries with high degree of uncertainty avoidance.

Masculinity refers to a preference for assertiveness, heroism and material rewards for success. Psychological studies show that in different fields, men tend to overestimate their real abilities and have unwarranted certainty in the accuracy of their beliefs compared to women. In finance field, overconfident men overreact when investing in stock markets (Daniel et al., 2001). Eilnaz et al. (2020) point out masculine culture and masculinity are linked to overconfidence. Picone et al. (2014) notice that hubris and overconfidence are used as synonymous in the management literature. A masculine culture could increase both overconfidence and hubris of CEO and thus enhance the effect of investor sentiment on their decision. Thus, we can except that the impact of investor sentiment on bid premium is stronger in countries where the degree of masculinity is high.

In this context, several studies find strong influence of investor sentiment on stock returns in countries that are more likely to demonstrate herding behavior or overreaction (Schmeling, 2009; Wang et al., 2021). Similar to these studies, we use individualism, uncertainty avoidance, and masculinity as proxies for herding behavior and for the tendency of investors to overreact across countries. We assume that behavioral biases are proxied by low levels of individualism, high uncertainty avoidance and masculinity. These findings lead to the following hypothesis:

H2: The impact of investor sentiment on deal premium is stronger when the acquirer is based in countries which are culturally prone to behavioral biases.

Market integrity

Market integrity means that financial markets with a higher level of institutional sophistication are characterized by a better flow of information and are consequently more efficient. The efficiency of an economy and capital markets is based on how well the legal system protects outside investors. The market integrity indicators used in our study can be found in La porta et al. (1998) and include: outside investor rights, legal enforcement and accounting standards index. Outside investors’ rights are proxied by the anti-director rights index which captures how strongly the legal system favors minority shareholders over dominant shareholders. Legal enforcement is measured by two legal variables: the efficiency of the judicial system and the corruption index. The accounting standards index measures the quality of the financial reporting across countries. Our last market integrity indicator is similar to Hung’s (2001) index. A high index value in a particular country indicates that higher use of accruals accounting is permitted. This indicator assesses the extent of accruals accounting in various countries by evaluating the extent to which the accounting systems depart from a cash method. Hung (2001) show that higher use of accruals accounting decreases the quality of accounting information and the efficiency of capital markets.

McLean et al. (2012) find that stronger investor protection leads to accurate stock prices and efficient investment. In line with this finding, Wang et al. (2021) expect that high-quality institutions will help improve information flow and make stock markets more efficient. Thus, investor sentiment should have a lower impact on stock prices in countries with highly efficient institutions and higher market integrity. The reason behind this is that a high market integrity helps rational investors to offset the effects of noise trader’s sentiment. The empirical results of Wang et al. (2021) and Schmeling (2009) support this view.

In our study, we use a cross-section of countries to determine if there is evidence that impact of investor sentiment on bid premium is related to the level of development of financial institutions and the level of sophistication of stock markets. We argue that high market integrity is identified by strong outside shareholder rights, a high legal enforcement, a high quality of accounting standards and low allowance of accruals. Thus, we formulate the following hypothesis:

H3: The impact of investor sentiment on deal premium is weaker when the acquirer is based in countries with high market integrity.

Serial deals

Serial bidding is frequent in M&A transactions. Through their internationalization strategy, bidders mostly adopt serial corporate acquisition programs to seize new markets, to obtain new skillsets and competitive technologies, and to keep and strength their market position in the world. However, the imperative to select appropriate targets and then execute the relevant transactions successfully is far greater in a serial acquisition strategy context (Smit and Moraitis, 2010). Execution of a serial acquisition strategy is vulnerable to the biases of managers’ judgment manifested in their strategy and how they perceive risks and losses. Moreover, serial deals lead to empire building where overconfident managers overpay for targets (Malmendier and Tate, 2008). Following hypothesis H1, we expect that the higher the investor sentiment surrounding the serial acquirer, the higher the premium paid. However, it can be argued that serial auctions are special events that conduct to the emergence of a learning effect. Aktas et al. (2011) develop this view and propose a hypothesis that serial acquirers benefit from their experience and tend to adjust their bids according to market expectations without being influenced by investor sentiment. Aktas et al. (2013) further develop their model and provide evidence that the time elapsed between successive deal announcements demonstrates a learning behavior. They argue that serial acquisitions allow even hubristic CEOs to learn by experience and regulate their valuation process. Thus far, the ability of investors to learn from their past experience is still a matter of debate in the literature. Chang et al. (2015), in the context of Taiwan’s options and futures markets, find that investors do not learn by doing and that investor sentiment exacerbates mistakes. Chiang et al. (2011), in the context of IPOs, document that investors’ abilities decline with experience. However, in the M&A context, firms develop knowledge and expertise regarding target selection, valuation assessment, negotiation strategy, and integration (Collins et al., 2009). Using cross-border mergers, Pandey et al. (2021) provide evidence of acquirer learning in acquisitions of private and public targets from less competitive takeover markets. These elements lead us to the following nondirectionally hypothesis:

H4: The impact of investor sentiment on deal premium is associated with serial acquirers.

Figure 1 in Appendix 1 summarizes the hypotheses developed in this article.

Data and variables

M&A data

Our sample consists of all announced M&A between 2000 and 2021. Our initial sample is obtained from Thomson SDC. Following previous studies, we include only significant deals (valued at U.S. $1 million or more). We place no restrictions on the public status of the target or the bidder. We identify 369,727 deals during the period 2000–2021. Investor sentiment data come from various sources (national banks, international organizations, etc.)[2] while the M&A premiums are obtained from SDC. Firm-year observations with missing data are excluded from our sample. Merging our M&A data with investor sentiment data yields a final sample of 44,697 M&A transactions released by 25,752 firms in 54 countries for which M&A premium data are available in SDC.

Bid premiums

The main dependent variable is the M&A premium which is defined as the percentage increase (decrease if negative) of the bid price over the closing market price four weeks prior to the announcement, [(offer price - previous 4 weeks’ closing market price)/previous 4 weeks’ closing market price] × 100% (Surendranath et al., 2016). 28 days before the announcement date is a relevant window to avoid the informational leakage, the influence of run-up effect on the stock price of the target prior to the announcement.

Investor sentiment

Investor sentiment can be defined as a belief about future cash flows and investment risks that is not warranted by fundamentals (Baker and Wurgler, 2006; Lemmon and Portniaguina, 2006). In the literature, several studies have attempted to quantify investor sentiment[3]. Nowadays, a universally accepted measure of investor sentiment has not yet been identified. Given our international setting, we choose the consumer confidence index that offers significant advantages over others sentiment indexes. First, the financial literature highlights a positive and significant relationship between consumer confidence and investor sentiment. Qiu and Welch (2006), for example, show that the alternative candidate, consumer confidence index, is a reasonable proxy of investor sentiment. They find that consumer confidence changes correlate strongly with changes in the direct UBS/Gallup investor sentiment survey data. They conclude that (p.26) “Consumer Confidence can be validated as a proxy for investor sentiment.” Second, the consumer confidence index is a direct accurate proxy of individual investor sentiment because it is based on a monthly survey of a large number of households about their current and expected financial situations and their beliefs about the economy. Because the consumer confidence index captures individual beliefs, it reflects the philosophy of behavioral finance by including the opinions of imperfect people who have social, cognitive and emotional biases (Shleifer, 2000). Third, our selection is the result of the well documented established relationship between the consumer confidence index and the equity market. Many studies use the consumer confidence index as the main proxy of investor sentiment and show that this sentiment indicator seizes stock market aspects not already contained in rational macroeconomic indicators (Lemmon and Portniaguina, 2006; Antoniou et al., 2013; Wang et al., 2021). Fourth, the data on consumer confidence index are available for several international markets for long and regular periods of time, including the emerging market, giving an international dimension to the validity of our results. Finally, and most importantly, the nature of this paper examining international M&A transactions including both developed and emerging markets requires consistency across all sample markets. Thus, the investor sentiment measure used in an international study should be applied in all countries of our sample. The consumer confidence index offers such wide availability in all 54 countries of our study. Form this view, several studies have used the consumer confidence index in international setting as a relevant proxy for investor sentiment (Schmeling, 2009; Wang et al. 2021).

The raw consumer confidence index encompasses a psychological component related to sentiment and a rational component related to economic fundamentals. As noted by several studies, consumer confidence index varies in part for entirely rational reasons related to the macroeconomic conditions. Isolating sentiment consists precisely of identifying investors’ optimism (pessimism) although there is not a good (bad) valid economic reason for being so. In order to properly measure sentiment, we estimate equation (1) and take the residuals (ε_j,t) from it as a measure of investor sentiment.

CCI_j,t is the consumer confidence index for each country j at time t, α_j is the constant, and are the parameters to be estimated. is the set of fundamental variables representing rational expectations based on risk factors of every country. These variables include growth of industrial production, inflation, term spread, and growth in durable, nondurable, and services consumption. The fitted values of equation (1) capture the rational component, and the residual (ε_j,t) (captures the psychological component (Sent_j,t = ε_j,t). To ensure comparability across countries, CCI and fundamental variables were standardized in each country with zero expectation and unit variance.

Bidder and deal characteristics

The vector of control variables includes the following bidder and deal characteristics. Acquirer_size measures the natural logarithm of the acquirer’s total assets the fiscal year before the deal announcement. Managers of large firms might be more prone to overconfidence; thus, larger acquirers pay larger premiums (Moeller et al., 2004). Relsize is the relative size of the target. This variable is measured by the ratio of deal value divided by the acquirer’s total assets the fiscal year before the deal announcement. The smaller the target, the higher the offer premium (Betton et al., 2008). Acquirer_profitability is the acquirer’s return on assets over the last 12 months. The performance of acquirers and targets does not seem to have a significant effect on the premium (Mpasinas, 2007). Acquirer_leverage is the ratio of the acquirer’s total debt to total assets. Highly leveraged bidders are less likely to engage in negative net present value projects and are less likely to overpay in acquisitions (Black, 1989). Acquirer _public is a dummy variable that equals 1 if the acquirer firm is a public company and 0 otherwise. Public target shareholders receive higher premiums when the acquirer is a public firm rather than a private equity firm (Bargeron et al., 2008). Toehold corresponds to the percentage of equity of the target held by the bidder firm before the announcement. Simonyan (2014) shows that holding a substantial proportion of shares in the target prior to the deal provides the bidder with substantial control. Thereby, the bidder firm may induce target firm shareholders to accept a lower premium. Relatedness shows the effect of activity relatedness of the acquirer and the target on premium. According to Bae et al. (2002), unrelated deals are sometimes motivated by private benefits enhancing the amount of premiums. However, premiums are more associated with related acquisitions due to higher synergies. Cash is used to control the method of payment. A dummy variable equals 1 if the acquisition is entirely paid in cash and 0 otherwise. Simonyan (2014) reported that cash-offers are associated with large premiums. Number_bidders is the number of entities (including the acquirer) bidding for the target. Competition in the takeover market can increase the amount of premiums. Therefore, this variable can control for the effect of the winner’s curse on premiums. The winning bid premium overstates the value of the expected takeover gain (Varaiya, 1988). Cross-border is a dummy variable that is equal to one whether target’s and bidder’s nations differ. According to the hypothesis that idiosyncratic synergies for foreign acquirers exist, cross-border transactions offer higher premiums (Harris and Ravenscraft, 1991). Hostile is a dummy variable that equals 1 if the bid is classified as unsolicited and 0 otherwise. Control contests are likely to drive up premiums, thus, hostile targets have the highest premiums (Eckbo, 2009). Tender_offer is a dummy variable that equals 1 if the acquisition technique is a tender offer and 0 otherwise. Tender offers are more costly than mergers or other types of acquisitions. Premiums will then be higher in tender offers than in mergers (Offenberg et al., 2015). M&A waves is a dummy variable that equals 1 if the bid is announced during the periods 1985–1989, 1993–2000, 2002–2007 and 0 otherwise. During M&A waves, managers displayed less over-optimism and offer significantly lower premiums (Alexandridis et al., 2012). Appendix 2 presents the definition of the variables used in this study and their expected effect on bid premium.

Empirical strategy

There is wide literature (e.g., Eckbo, 2009; Aktas et al., 2009) discussing the relation of acquirer- and deal-specific characteristics with deal premium. To assess the effect of investor sentiment on premium, we run regression of the bid premium on the investor sentiment measure. Specifically, we employ the following econometric specification:

Where is the premium_i,t for deal i at time t. Bidder and deal characteristics’ control variables were discussed in the previous subsection. We also include, as a control variable, the annual real GDP rate in the regression. We use lagged explanatory variables to mitigate the endogeneity problem. A country’s M&A market is impacted by the legal business environment for investors (Ciobanu, 2015). Therefore, we also controlled for country, year and industry fixed effects. υ_C, υ_I, and υ_t are respectively the country fixed effect, industry fixed effect and the time fixed effect. Finally, we computed standard errors clustered by firm because standard errors may be underestimated in panel data sets.

Summary statistics and empirical results

Descriptive statistics

Table 1 reports the descriptive statistics for all the variables used in our study. The bid premium averages 18.432%. The average investor sentiment is equal to 0.187 with a standard deviation of 4.098, where the range is from -9.072 to 3.442. This suggests that the investor sentiment for each bidder’s country varies from one to another and some countries may have lower consumer confidence. Table 1 also provides additional information about M&A deals (37.2% of the firms are in the same industry, 22.6% of the deals are cross-border, more than half of the deals (58.5%) are all-cash offers, 30.9% of acquirers are serial bidders and 5% of the deals are hostile). The average bidder size is 7.711, which is about 2232.7 million USD. The average leverage ratio of sample firms is 0.544, with a minimum ratio of 0.0811 and a maximum ratio of 0.945 during the sample period.

Table 2 presents the distribution of the average deal value and average premium during the sample period. This table shows that global M&A activity has increased substantially around the world and firms expand internationally through non-organic growth operations. As reported in this table, the aggregate deal value decreased dramatically after the global financial crisis of 2008 from an aggregated dollar value of 1,005.83 million in 2006 to an aggregated dollar value of 234.85 million in 2009, before increasing again in 2014. M&A activity reached its highest number of deals (3,112 M&As) in 2009 with the lowest aggregate volume of $234.85 million. The financial crisis has significantly altered the global M&A market, where a significant number of M&As involving financially distressed firms are carried out. The highest average premium is about 30.21% observed during the internet bubble of 2000.

Table 3 illustrates the distribution of M&A premiums by acquirer macro industry. Among them, the healthcare and materials industries have the highest bid premiums. Finally, table 4 provides the distribution of bid premiums across 54 countries. These countries are classified by geographic zone and from the highest level to the lowest level of investor sentiment. Amongst these countries, Colombia, Argentina, Ireland, and Canada offer the highest premiums.

Table 3

**Summary statistics by acquirer macro industry**

Table 4

**Summary statistics by acquirer nation**

Multivariate tests

Baseline regression analysis

To examine the effect of investor sentiment on deal premiums in an international setting, we run an OLS panel regression by including the bid premium as our dependent variable and the investor sentiment measure as our independent variable. The results of this first regression are reported in Table 5.

The results in column 1 are consistent with our first hypothesis and show that investor sentiment has a positive and statistically significant impact on bid premiums. As suggested by Dicks and Fulghieri (2021), market sentiment influences the firm’s valuation and investment decisions. During high sentiment periods, CEOs incorporate the expectation of future profitability into their strategic decisions (Danso et al., 2019). Following this reasoning, CEOs are prone to hubris, and therefore they tend to overpay for a potential target because they overestimate the long-term growth of this investment (Roll, 1986). In terms of economic significance, a one standard deviation increase in sentiment is associated with an increase of about 2% in deal premiums.

In model 2, we include only control variables. The coefficients associated with Acquirer_size, Acquirer_leverage, Hostile, Cross-border, Tender_offer and GDP_growth are significant. According to the previous studies, large acquirers use to pay high premiums (Moeller et al., 2004); high levels of debt discourage acquirers to pay excessive amount of premiums (Black, 1989); and Cross border deals (Harris and Ravenscraft, 1991), hostile takeovers (Eckbo, 2009) and tender offers (Offenberg et al., 2015) enhance the bid premium.

In model 3, we also include the Sent variable, and we note that the significance of the control variables is not affected by the inclusion of our variable of interest. Similarly, we find that Sent variable remains highly significant and that the explanatory power of the model increases to 24%, representing a 0.178-point improvement in the R² adjusted. This last model shows that, after controlling for several variables, the acquirer country’s investor sentiment still has a significant and positive effect on the deal premium. In unreported tests, we use other measures of the bid premium (1-day premium; 1-week premium) and the results remain the same.

Overall, our evidence supports the hubris hypothesis (H1). This effect is economically significant. Influenced by investor sentiment in their country, bidders are more likely to make non-rational decisions by overpaying for the target firm. Estimates show that during periods of high investor sentiment, CEOs of bidder firms are more optimistic about creating synergies and, therefore, tend to overvalue the target firm by paying a high premium. They are willing to pay high premiums to complete deals in periods characterized by excessive investor euphoria. Our results are in line with previous studies that argue that investor sentiment proxied by the non-rational part of CCI affects the firm’s strategic decisions and policies (Bergman and Roychowdhury, 2008). We confirm that investor sentiment is a real determinant of target selection process which can play a crucial role in M&A synergies and risks valuation (Bouwman et al., 2009; Danbolt et al., 2015).

Table 5

**Effect of Investor Sentiment on the Deal Premium**

Cultural and institutional factors

To test the moderating effects of culture or market integrity on the association between investor sentiment and bid premium, a new model is estimated including the variable investor sentiment, the moderating variable (culture or market integrity), the control variables and the interaction variable (culture variable × Sent variable or market integrity variable × Sent variable).

Table 6

**The moderating role of institutional and cultural factors**

Table 6 presents the regression results incorporating the cultural and institutional dimensions. Model 4 focuses on cultural factors such as collectivism (vs. individualism), uncertainty, and masculinity. These three dimensions significantly accentuate the positive effect of the Sent variable on premiums (see coefficients of interaction terms). This confirms H2. The effect of investor sentiment on the bid premium is enhanced when the acquirer is based in a country culturally prone to herd behavior. According to Schmeling (2009) and Wang et al. (2021), national cultures that promote herding enhance the effect of investor sentiment. In a such environment, a CEO of a bidder firm is more likely to be overconfident about her expectations and tend to overpay. She can also just replicate an excessive behavior of other bidders based on overly optimistic takeover market. This finding supports the idea that bidders’ CEO in different cultures have different biases. Furthermore, our result confirms the fundamental role of the three cultural dimensions (individualism, uncertainty avoidance and masculinity) to justify a behavioral explanation of international stock markets[4].

To capture the incremental effect of investor sentiment resulting from the institutional context, we introduce interaction terms between the Sent variable and institutional variables such as the anti-director rights index, corruption index, efficiency of the judicial system index, accounting standards index, and accruals index. Model 5 shows a negative and significant impact of three interaction terms on bid premiums. The corruption index, efficiency of judicial system index, and accounting standards index reduce the positive impact of investor sentiment on bid premiums. High-quality institutions like judicial system and accounting standards system improve information flow and reduce herding behavior associated to market sentiment.

The interaction with the accruals index shows a positive and significant impact on bid premiums. The impact of investor sentiment on bid premiums is stronger in countries with allowance for accrual accounting. Market institutions influence the effect of investor sentiment as advanced institutions improve information circulation and thus make stock markets more efficient. A high accruals index value indicates that extensive use of accruals accounting is permitted in the country, making this index a poor indicator of an efficient institutional system. These results are in line with H3 and with those of Schmeling (2009) and Wang et al., (2021) find that investor sentiment has less influence in countries with highly efficient institutions and greater market integrity.

Serial deals

Table 7 examines the effect of serial deals on the relationship between investor sentiment and bid premiums. The first column of the table focuses on bidders announcing more than one deal in the past 12 months while model 8 deals with bidders announcing only one deal. Investor sentiment has a positive effect on bid premiums in both models, but the coefficient for serial acquirers is lower (0.177 versus 0.501).

To test the moderating role of serial deals, we use an interaction term between the Sent variable and the Serial variable. The coefficient of the interaction term in model 9 is negative and statistically significant. The relationship between investor sentiment and the bid premium is attenuated as follow-up deals occur. This result supports the learning effect hypothesis (Aktas et al., 2011). As bidders develop knowledge and expertise across their serial M&A, learning can lead acquirers to develop additional skills in M&A activity. They can select appropriate targets, negotiate better the integration strategy and assess expected synergies more accurately (Collins et al., 2009). Thus, CEOs of serial acquirers are less influenced by market investor sentiment.

Table 7

**The moderating role of the serial deals factor**

Robustness Tests

In the following, we perform a battery of robustness checks. The results are presented in Table 8 below.

U.S is considered one of the most individualistic cultures in the world. According to Hofstede (1980), individualistic countries tend to exhibit less herding behavior and the impact of the sentiment is lower in these countries. As a result, the effect of investor sentiment on the bid premium is less pronounced when the acquirer belongs to an individualistic country. As a relatively large proportion of M&A deals took place in the US market, our results can be driven by U.S effects. For this robustness test, we exclude from our sample all U.S bidders during the period 2000-2021 and we re-estimate our model as previously specified. We display the results in column 1 of Table 8. The coefficient associated to investor sentiment measure is still significant and the effect of investor sentiment on the bid premium becomes more pronounced.

Further, we re-estimate our model by removing the withdrawn deals from our sample. According to Jacobsen (2014), CEOs exhibit restraint and cancel a deal when it becomes too expensive. Therefore, the deal status can drive our results. The results tabulated in column 2 of Table 8 confirm that deal status is unlikely to influence the results on the relationship between investor sentiment and bid premium.

We check whether our results are affected by the 2008 financial crisis and the COVID-19 crisis. Crises often trigger M&A waves. For instance, the 2008 financial crisis results into the third wave of bank mergers, when several financial banks acquired failing firms because of their inability to secure funding. During the Covid-19 pandemic, some other merger waves occurred at the aggregate and industry level. As a result, the takeover premiums paid during these crises can drive our results. The results in model 12 reveal that our findings persist. The effect of investor sentiment on bid premium is positive and significant and such effect appears to be even stronger when we exclude deals during the 2008 financial crisis and the COVID-19 crisis from our sample. The reason is that the bidder undervalues a target with high information uncertainty (Li and Tong, 2018) and, therefore, doesn’t tend to overpay for target firms. Excluding announced deals during the crises accentuates the effect of the investor sentiment on bid premium.

Lastly, we also carry out an additional robustness test using the widely-used Baker and Wurgler (2006) sentiment index[5] as an alternative measure of investor sentiment. We limit our analysis to the U.S sub-sample as this sentiment index is only available for the US market. We find that the U.S. consumer confidence index is strongly correlated with Baker and Wurgler sentiment indicator over our study period 2000–2021. The Pearson correlation coefficient is 0.421 and statistically significant at 1%. This finding explains why the consumer confidence index has acquired a solid reputation as a measure of investor sentiment (Lemmon and Portniaguina, 2006; Schmeling, 2009; Antoniou et al., 2013). We also re-estimate the model (2) using the Baker and Wurgler (2006) sentiment index on the U.S sub-sample. The results in model 13 show that our conclusion remain unchanged across this robustness test.

Conclusion

This paper provides international confirmation that investor sentiment is an important determinant of bid premiums and plays a role in understanding bidding behavior. A large part of literature offers rational explanation of overpayment phenomenon such as private benefits of control, poor target management and overvaluation of synergies. We adopt a behavioral perspective, and we use a longstanding and established measure of investor sentiment based on the consumer confidence index available for several international markets for long and regular periods of time. Our results indicate that investor sentiment affects significantly and positively bid premiums. High investor sentiment induces CEOs to be more optimistic about creating synergies and tend to overvalue the target expectations and pay high premiums. Thus, this paper joins other studies revealing the effects of behavioral market biases on investing and financing decisions made by CEOs. This is of particular importance in the context of M&A because mispricing the initial bid price may impede the acquirer’s future performance.

The international dimension of our sample allows for investigation of the influence of institutional factors and national cultures on the relationship between investor sentiment and bid premium. Cultural factors such as collectivism, uncertainty and masculinity accentuate the effect of investor sentiment on premiums. Thus, bidders in countries culturally prone to herd-like behavior use to replicate an overpayment behavior when the takeover market is overly optimistic. Furthermore, we find that the investor sentiment has less influence in markets with stronger market institutions than in those with relatively weaker market institutions. Market integrity leads to high level of institutional sophistication characterized by more efficient flows of information. Lastly, we find that managers benefit from their previous experiences and are less influenced by investor sentiment as they conduct M&A transactions. Our results are robust to several checks (alternative measures of bid premium and investor sentiment, respective exclusion of US deals, withdrawal deals, and deals announced during crises).

This paper offers a set of recommendation and important implications for academics and practitioners. Managers and analysts should take into consideration the risk of investor sentiment during target synergies valuation and negotiation process. They should bear in mind that periods of high optimism are followed by excessive bid premiums. Firms undertaking serial corporate acquisition programs may hire experienced managers in M&A field who can be less influenced by market investor sentiment. Lastly, regulators are invited to improve the institutional system (e.g., reducing corruption, enhancing minority shareholder rights, improving judicial system integrity and accounting standards system quality). This can help to avoid target overpayment behavior. Future research might further develop the effects of the investor sentiment on other characteristics of M&A activity.

Does investor sentiment impact premiums in M&A transactions?Impact du sentiment des investisseurs sur les primes d’acquisitionsImpacto del sentimiento de los inversores en las primas de adquisición

Abstract

Résumé

Resumen