Political Risk and Deal Completion Likelihood

1. Introduction

Risks related to political systems, especially politics associated with regulation, trade, tax policy, economic policy and budgets, and the enforcement of rules, have a major impact on the business environment. Recent events, such as repeated shutdowns in the US Federal Government and the United Kingdom’s (UK) decision to leave the European Union (EU), have provided valuable insights into the risks emanating from political systems and their economic effects on company investments, employment, and capital expenditures. Many firms are subject to political risks. For example, Google and Amazon spend millions of dollars each year on lobbying to influence the government not to engage any antitrust lawsuits.

Hassan et al. (2019) argue that an increase in the firm-level degree of political risk-taking leads to a significant increase in firm-specific stock return volatility and a significant decrease in the firm’s investments, planned capital expenditures, and hiring. Gad et al. (2019) find that lender-level political risk has a significant effect on the supply of credit and affects loan pricing. Using a novel measure of firm-level political risk, this paper examines the relationship between political risk and deal completion takeover and duration.

M&As are an important form of investment decision, and managers must consider the political environment at the firm level and at the country level before engaging in such decisions. Despite the prevailing importance of political risk in management studies, related theoretical and empirical developments are lacking in the field of research concerning M&As (Sitkin & Pablo, 2005). Therefore, examining whether political risk affects the likelihood of a company completing a deal is likely to make a novel contribution to the literature. A priori, an increase in political risk faced by the bidder decreases the propensity of firms to engage in M&A investments (Cao et al., 2017; Dang et al., 2022; Nguyen & Phan, 2017). Hence, the entities involved in deals tend to withdraw this irreversible investment as they face greater uncertainty and the expected synergy may not be realised. While this argument seems convincing, some empirical studies have found that uncertainty positively affects M&A decisions through possible mechanisms of real option and prospect synergy effect (Shen et al., 2021). Therefore, this debate is still to be settled. We contribute to this debate by empirically examining the effect of political risk on M&A decisions, using a sample of US deals.

Our M&A data were retrieved from the Securities Data Company (SDC) Platinum database. Our sample spans from 2002 to 2019 to match the availability of the political risk measurement data. We used the firm-level political risk measure developed by Hassan et al. (2019). This measure is constructed using the entire conference call held, in conjunction with an earnings release. The overall measure of political risk is the weighted sum of bigrams associated with political risk or uncertainty in a firm’s quarterly earnings conference call. This measure indicates the percentage of the conversation dedicated to politics, using the firm’s quarterly earnings conference call with market participants (usually financial analysts). Intuitively, when market participants ask more questions about political topics or when management disclose more political information in their discussions, the firm is more likely to be exposed to political risk. Indeed, we include several controls in all the regressions to ensure that our political risk measure is not a proxy for other factors that can have an impact on the likelihood of deal completion.

Using a sample of 18,341 firm-year observations of 3,262 unique firms over the 2002 to 2019 period, we find a negative relationship between political risk and the likelihood of deal completion. Our findings also suggest that an increase in the firm-level measure of political risk leads to a significant increase in deal duration. While canceling bids and a longer deal duration can impair the reputation and credibility of the firm, M&A deal completion can also incur a dramatic cost for the company. Indeed, the empirical evidence shows that the acquirer will tend to be prudent with their investment strategy during periods of high political risk.

Our study makes several contributions to the ongoing debate about the factors that affect acquisition decisions made by the bidder. Firstly, our study complements the literature on the pre-completion stage of the acquisition process. While most of the literature focuses on post-acquisition performance to explain M&A success, our study provides valuable insight into the importance and complexity of the intermediary acquisition phase to explain M&A failure. Studying this phase is crucial because of the complexity of the bargaining process and the importance of the learning experience, insight into which can be useful for the bidder. Secondly, our study contributes to the previous literature by showing that acquirers’ political risk is an important factor determining the likelihood of merger completion. Within the M&A literature, most attention has been paid to explaining completion likelihood or outcomes of M&A deals using economic or political factors related to the target’s country. Relatively less attention has been paid to the bidder firm and firm-level factors, and especially political risk, which can influence deal completion likelihood.

Our paper is closely related to the studies of Nguyen and Phan (2017) and Cao et al. (2017), both of which found that political uncertainty affects M&A takeovers. We nevertheless differ from these papers by focusing on US M&As using a firm-level measure of political risk. Prior research in this area explores how variations in the aggregate measure of political risk exposure (such as a firm’s sensitivity to (economic policy uncertainty) EPU) can affect M&A decisions. We contribute to the existing literature by using a new measure of political risk that enables us to explore how the variation in firm-level political risk can affect investment-related decisions (such as deal completion and duration).

Moreover, our study adds to the literature dealing with risk management and uncertainty, providing important initial evidence on how political risk – unlike other types of risk – affects deal completion and duration. A firm’s exposure to political risk creates uncertainty regarding the potential for political and regulatory interference in its operating and investment decisions. From a policymaker’s perspective, we also show that some types of political risk (especially political risks associated with health, environment and economic policy) are more detrimental than others with respect to the likelihood of completing an M&A deal. Finally, our study has implications for corporate decision-makers, given the higher failure rate of M&A deals. This study can be useful for acquirers trying to increase M&A success by enhancing the performance of the acquisition at any stage of the acquisition process. Previous research has shown that learning from prior experiences can be beneficial for executives in managing the integration process more effectively.

This paper is structured as follows. Section 3 describes the data and construction of variables and provides summary statistics for the variables used. The empirical analysis results are discussed in Section 4. In Section 5, we provide the results of the robustness tests. In Section 6, we conclude the discussion and the paper.

2. Empirical Literature and Hypothesis

This section reviews several relevant studies on how uncertainty may affect investment decisions and presents our hypotheses to guide the following empirical analysis.

Previous research has studied the effect of politics on corporate investments. Julio and Yook (2012) found a negative relationship between capital expenditures and political uncertainty. In particular, firms tend to reduce their capital investments by an average of 4.8% during election years. Using national elections as a source of variation in political uncertainty, Cao et al. (2017) argued that political uncertainty affects the volume and outcome of M&As. They also found that markets react favorably in the period prior to a national election in the country of the acquiring company. Bonaime et al. (2018) showed how political and regulatory uncertainty is negatively related to the likelihood of deal completion. Indeed, they argued that the bidder is more likely to engage in vertical mergers and to acquire a foreign firm during periods of political uncertainty.

Given the importance of M&A investments and their effects on the economy, several studies have focused on the post-integration process to explain M&A failure. Despite the relevance of the post-completion phase, focusing on pre-merger activities is crucial, as empirical evidence suggests that more than 25% of bids are abandoned at some point in the pre-completion stage (Dikova et al., 2010). Deal abandonment can severely impact both parties involved in the M&A. Besides incurring huge costs like penalties or termination fees, the reputation and share price of these parties can be affected. For example, the telecommunications corporations of Norway (Telenor) and Sweden (Telia) announced their intention to merge on 20 January 1999. “Despite gaining praise from analysts and the press, it was abandoned in December 1999” (Muehlfeld et al., 2012, p. 4). This termination was considered as “the most extreme case of a merger failure, where none of the potential synergies were realized, and where the merger costs were substantial” (Meyer & Altenborg, 2008, p. 522).

Other factors have been documented in the literature to explain deal completion likelihood. Using a sample of 1,556 completed US mergers, Deng et al. (2013) found that acquirers with high levels of corporate social responsibility (CSR) have higher likelihoods for deal completion, with shorter durations for such completion. Using the similarity in firms’ CSR practices as a proxy for cultural fit, Bereskin et al. (2018) show how firms with a better cultural fit are more likely to complete deals, as they incur lower costs associated with their integration. Similarly, Bena and Li (2014) identified a positive relationship between technological similarities and the likelihood of deal completion. Luo (2005) argued that the stronger the market reaction, the greater the likelihood of deal completion. Acquirers learn from the market during M&As and may cancel the merger at the closing if the deal has a lower market reaction. Likewise, Luypaert and De Maeseneire (2015) found that acquiring larger target firms, funding deals with stock or engaging in hostile deals increases their complexity. However, bidder experience, shareholders’ support for the deal and overpayment accelerate deal completion. Nguyen and Phan (2017) suggested that M&A operations undertaken during periods of high policy uncertainty are more likely to fail and take longer to complete. In this study we add to the literature by explaining that firm-level political uncertainty proxied by political risk can explain deal completion and duration.

M&As are among some of the most important decisions companies may have to make and are risky and irreversible investments. Previous studies have investigated the effect of uncertainty on companies’ investment decisions by distinguishing between reversible and irreversible investments, such as M&As. For instance, Gulen and Ion (2016) pointed out that uncertainty can depress corporate investments as firms with higher levels of investment irreversibility are more likely to delay such investments. Rodrik (1991) argued that firms are more likely to delay irreversible investments during periods of high uncertainty. Im et al. (2017) further reported that uncertainty reduces a firm’s current investments and increases the probability of it holding cash. Risks related to politics, such as risks arising from the events surrounding a national election, can change a firm’s business environment and, therefore, create uncertainty. Hassan et al. (2019) demonstrated how a rise in a firm’s political risk triggers a decrease in its rate of capital investment. Banerjee and Dutta (2022) demonstrated that the higher the political risk, the more likely it is that a firm will postpone its ‘irreversible’ capital investments and increase its reversible short-term working capital investments. The political context and government regulation have a substantial influence on firms’ economic environment and, therefore, on their decisions and policies. Recent literatures have discussed the effect of political uncertainty proxied by political risk on firms’ decisions, such as leverage decisions (Gyimaha et al., 2022), excess cash holdings (Jeon et al., 2022), corporate investment (Choi et al., 2022) and innovation (Ahmed et al., 2022), but few literatures study the influence of firm-level political uncertainty on M&A decisions. For instance, Shen et al. (2021) found that firm-level political uncertainty proxied by geopolitical risk positively affects M&A decisions through possible mechanisms of real option and prospect synergy effect.

Hence, we contribute to this existing literature and we expect a significant relationship between firm-level political risk and M&A decisions (completion, deal likelihood and deal duration). Firstly, based on the real option theory, firms postpone M&As as the real options value increases with uncertainty. In line with this theory, Borthwick et al. (2020) and Nguyen and Phan (2017) found that policy uncertainty negatively affects M&A investments both in terms of the number and the aggregate value of M&A deals. Dang et al. (2022) found that policy uncertainty influences deal completion because the expected short- and mid-term synergistic gains can be affected. Consistent with this theory, political uncertainty might lower firms’ willingness to conduct M&A transactions by cancelling the deal announced. Another strand of the literature, based on agency theory, has argued that managers are more likely to conduct M&A deals when they face higher political uncertainty. Managers use external growth to manage the total volatility. For instance, Garfinkel and Hankins (2011) found that bidders conduct M&As to hedge cash-flow volatility. Furthermore, bidders may engage more in M&A operations and are, therefore, more likely to complete M&A deals to hedge against the volatility that arises from the political risk they are facing.

Consistent with the evidence in the literature, we predict a negative relationship between political risk and completion likelihood because political risk breeds uncertainty and is more likely to affect investments in M&A. We also expect that the greater the acquirer’s political risk, the longer will be the deal completion duration. Faced with political risk and its adverse effects, bidders with higher political risk are expected to be careful with their investment strategy. They are less likely to complete an M&A deal and more likely to delay deals when their firms face higher levels of political risk. We argue that bidders facing higher political risk are more likely to cancel M&A deals before closing.

H1: Bidders that are less exposed to political risk are more likely to complete deals.

H2: The higher the acquirer’s political risk, the longer the deal completion duration.

3. Data and Methodology

3.1 Data

Our initial sample includes all completed and failed M&As retrieved from the SDC Platinum database. The deal value disclosed in SDC was valued at $1 million or more from 2002 through to 2019. We placed no restrictions on the public status of the target or the bidder. We collected several deal characteristics from SDC, including payment method (stock, cash), deal attitude (hostile, friendly), the degree of relatedness between the acquirer and the target (diversifying/non-diversifying deal), relative deal size, offer structure (tender/non-tender offer), and a high-tech indicator. We extracted data on firm characteristics^[1] from Compustat. We then merged our M&A data with political risk data and Compustat data to obtain the full sample for the ‘deal-completion-likelihood’ analysis. This merging produced a final sample of 18,341 failed and completed M&As involving 3,262 unique firms.

3.2 Variable Measures

In this section, we discuss the variables used in our study. Our main dependent variables include takeover completion likelihood and deal duration. The bidder’s political risk is our focal explanatory variable.

3.2.1 Measure of Takeover Completion Likelihood

To estimate takeover completion likelihood, we used a dummy variable that takes the value 1 if the announced acquisition is completed, and 0 otherwise. The speed of deal completion is computed as the difference between the completion date and announcement of an acquisition.

3.2.2 Measure of a Firm’s Political Risk and Non-political Risk

We used the HHLA measure developed by Hassan et al. (2019) to proxy for political risk. Hassan et al. (2019) developed firm-level measures of political and non-political risk faced by firms listed in the US using textual analysis of quarterly earnings conference-call transcripts. They relied on a simple pattern-based sequence-classification method developed in computational linguistics to distinguish between patterns of language dealing with political versus non-political issues. Firstly, they defined a training library of political texts^[2] and another training library of non-political texts,^[3] identified either through the textbook-based approach or the newspaper-based approach. Each training library (political and non-political) was the set of two-word combinations (‘bigrams’)^[4] frequently used in political and non-political texts. For the topic-specific measure of political risk, they followed the same approach. They defined a training library of eight political topics (e.g. economic policy and budget, environment, trade, institutions and political process, health care, security and defense, tax policy, and technology and infrastructure) to identify the patterns of language commonly used when discussing a particular political issue.

They then computed the measures associated with political risk, non-political risk as well as the measures of political risk associated with the eight political topics. To construct the measure of overall exposure to political risk, they calculated the frequency of the bigrams used in a conference call in conjunction with synonyms of ‘risk’ or ‘uncertainty’ and divided this frequency by the total length of the conference call to obtain a measure of political risk.^[5]

Prior studies have used an aggregate measure of political risk exposure (such as the aggregate economic policy uncertainty (EPU)) which prevents them from exploring the rich variation in a firm’s political risk that exists within the firm (the ‘within-firm’ variation). In addition, several recent studies have used pre-defined dictionaries of significant words to construct their concept of interest. This approach lacks the ability to identify the distinct issues related to politics. Indeed, prior studies have used newspaper archives and corporate disclosures as source texts, whereas the approach adopted by Hassan et al. (2019) relies on conference calls, which provide a natural context to assess a firm’s overall exposure to a certain risk.

3.2.3 Control Variables

Following Nguyen and Phan (2017), we include bidder size, market-to-book ratio, book leverage, the previous 12-month stock returns, firm age, stock dummy, cash dummy (to avoid perfect collinearity, mixed stock-cash payment is not included), high-tech dummy, diversifying dummy, hostile dummy, and tender-offer dummy.

4. Baseline Empirical Results

4.1 Descriptive Statistics

Table 1 presents the distribution of M&A deals in our study according to acquirer industry. As shown in this table, most of the acquirers are in manufacturing (36.628%), finance, insurance and real estate (22.234%), and services (21.312%). We also present the distribution of our sample across years in Appendix 2. The number of M&A operations in each year increased gradually until the global financial crisis of 2007–2008. It dropped thereafter, from 1,215 deals with an aggregated value of US$ 651,663.4 million in 2006 to 697 deals with an aggregated value of US$ 371,012.8 million in 2009. It increased again in 2010 with 953 deals.

Table 2 provides descriptive statistics of our sample of M&As. The whole sample consists of 18,341 firm-year observations of 3,262 unique firms.

For the political risk variable, the mean and median are equal to 0.111 and 0.058 respectively. In our sample, 65.2% of the deals were successfully completed, while 34.8% failed to close. As reported in Table 2, bidders show an average market-to-book ratio of 1.912. The average leverage is 0.261, with a median of 0.231. Approximately half (54.9%) of the deals are all-cash offers. Less than 50% of the bidders are in the same industry (30%) or are high-tech firms (25.5%). Table 2 also gives additional characteristics about the M&A deals related to tender offer (the mean is approximately 0.971) and deal attitude (1% of deals are hostile). Moreover, the examination of matrix correlation shows that there is no correlation problem between variables.^[6]

4.2 Cross-sectional Regression Analysis

In this section, we provide results for the takeover completion likelihood and deal duration.

4.2.1 Political Risk and Takeover Completion Likelihood

Our first empirical test concerned the relationship between the probability of completing the M&A deal and the acquirer’s political risk. To investigate this relationship, we estimated the probability that the deal would be successfully completed using the following probit model:

$$\begin{aligned} & P_{i}\left(takeover\ Likelihood = 1|X_{i} \right)\\ &\quad = F\left( \alpha + B_{1}*Political\ risk + B_{2} \ * \ Controls_{i} \right) \end{aligned} \tag{1}$$

Where:

the dependent variable in the probit regression is a dummy variable that takes the value of 1 if the acquirer is more likely to complete the deal, and 0 otherwise.

X_i represents the acquirer’s political risk measure developed by Hassan et al. (2019)^[7] and F is the standard normal distribution.

We also controlled for several deal and firm characteristics, all of which have been shown in previous research to be associated with takeover completion likelihood. These controls include bidder size, market-to-book ratio, book leverage, the previous 12-month stock returns, firm age, high-tech dummy, diversifying dummy, stock dummy, cash dummy, hostile dummy, tender offer dummy and relative deal size. The definitions of these variables are presented in Appendix 1.

Table 3 reports the estimation results of the M&A probit regression (Models 1 and 2) using year and industry-fixed effects. To account for heteroskedasticity, t-statistics are calculated using standard errors adjusted for heteroskedasticity and acquirer clustering (similar to Deng et al. (2013)). Clustering the standard errors by years provides similar results. In Model 2, we include only independent and control variables without year and industry fixed effects and we estimate the effect of these variables on the completion likelihood. Among the firm characteristics, we find that the coefficients associated with bidder size, buy-and-hold return and firm age are significant in Model 1. Larger-sized bidders are more likely to complete M&A deals.

Additionally, the results suggest that the failure rate is higher for bidders acquiring a high-tech target. Across all columns, we argue that deal completion likelihood decreases when there is a hostile offer or a diversifying deal because they require much higher administrative costs.

We then include a measure of the acquirer’s political risk and all associated controls in the second model. In column 2, we find evidence that completion likelihood is negatively related to political risk, corroborating Hypothesis 1. As shown in Table 3, the coefficient associated with political risk is negative (β=-0.148) and statistically significant at the 5% level.

In another analysis, we used additional measures of political risks developed by Hassan et al. (2019). These authors constructed some topic-specific measures of political risks. These measures are associated with eight specific political topics: economic policy and budget, environment, trade, institutions and political process, health care, security and defense, tax policy, and technology and infrastructure. Political risks attributable to the institutions and political process include risks related to government reform, constitutional reform, campaign finance, immigration, and federal elections. Given that all these risks are related to elections and their financing, they are more likely to impact a firm’s decisions or policies such as funding policy, rather than deal completion likelihood. Political risks attributable to security and defense are the risks related to “discussions of how government budget cuts and the winding down of activities in Iraq and Afghanistan affect the demand for the firm’s products” (Hassan et al. (2019), p. 28). As specified in this definition, these risks are more likely to affect the demand for the firm’s products. Therefore, we do not anticipate a significant relationship between deal completion likelihood and these types of risks. Political risks associated with technology and infrastructure are inherent to risks dealing with weapons programmes, aerospace, and computers and information technology. Thus, we expect a non-significant relationship between deal completion likelihood and these specific political topics. Political risks associated with trade are essentially the risks that pertain to free-trade agreements. Political risks linked to tax policy are the risks are attributable to tax reform. These risks could have an impact on the completion of cross-border-deals. As only approximately 10% of announced deals are cross-border, the resulting sample would likely be so small as to be almost meaningless. However, we except a significant relationship between political risks related to environment^[8], health^[9] or economic policy and deal completion likelihood, as these types of risks breed uncertainty. Regarding political risk associated with economic policy, Nguyen and Phan (2017) find that uncertainty arising from economic policy affects M&A investment. In the same vein, Bose et al. (2021) argued that environmental issues matter in M&A decisions.

Columns 1 through 8 of Table 4 provide empirical support for results presented in Table 3. Consistent with this prediction, the coefficients associated with specific political topics such as trade, institutions and political process, security and defense, technology and infrastructure, and tax policy are not significant. This means that political matters related to these specific topics do not directly affect deal completion. Moreover, the results presented in Table 4 indicate a negative and significant relationship between political risk associated with environment (1), healthcare (4) and economic policy (7), and likelihood of completion. This negative relationship is highly statistically significant for political risks associated with economic policy. Our results suggest that the effect of deal completion likelihood is more pronounced with the political risk associated with economic policy. Our findings are in line with Hassan et al. (2019) who suggested that uncertainty about economic policy is a major component of the aggregate variation in political risks. Our results show that firms should integrate political risk related to economic policy when investing in M&As. Hassan et al. (2019) stipulated that political risk generated by economic policy results in an increase in a firm’s propensity to lobby government on that topic of ‘economic policy and budget’ in the following quarter (political risk measures are quarterly measures).

4.2.2 Political Risk and Deal Duration

In addition to the takeover completion likelihood, we also investigated whether political risk affects the speed of deal completion. Empirical evidence advanced several factors that affect deal duration, such as information asymmetry, deal complexity, and profitability of the deal for both parties involved (Deng et al., 2013; Hussain & Shams, 2022; Luypaert & De Maeseneire, 2015; Nguyen & Phan, 2017). For instance, Nguyen and Phan (2017) found that policy uncertainty lengthens the time of deal completion. In contrast, Hussain and Shams (2022) argued the higher the bidder’s CSR scores relative to the target’s, the less time it takes to complete the deal. Drawing from the previous literature, we expected a negative relationship between firm-level political risk and the time taken to complete the deal. As M&As are irreversible and risky investments, firms are more likely to delay them during periods of high political uncertainty.

To examine the effect of political risk on deal duration, we used the Cox proportional hazard model (CPHM) (Bereskin et al., 2018). The CPHM is the most used statistical model for survival analysis in prior studies and particularly in M&A studies (Bereskin et al., 2018; Deng et al., 2013; Nadolska & Barkema, 2007). In the context of M&As, CPHM determines the factors that affect the likelihood of deal completion. The concepts of hazard rate and survival function are both bases for the CPHM. Similar to Nadolska and Barkema (2007), we chose the CPHM to obtain estimates of the hazard rate due to its flexibility. Unlike discrete time models, such as lognormal models, the CPHM does not require the identification of a specific, distinct hazard function (Nadolska & Barkema, 2007). Another advantage of the CPHM is that it considers both censored and uncensored observations; consequently, it offers a suitable estimation of the equation parameters. The dependent variable is the number of days between the announcement date and the effective date of deal completion. Controls include bidder size, market-to-book ratio, book leverage, the previous 12-month stock returns, firm age, high-tech dummy, diversifying dummy, stock dummy, cash dummy, hostile dummy, and tender offer. The regression specification is as follows:

$$Deal\ Duration = F\left( \alpha + B_{1}*Political\ risk + B_{2}{*Controls}_{i} \right)\tag{2}$$

Columns 1 and 2 of Table 5 report the regression results from a CPHM. This table shows the hazard ratios obtained from this model. The coefficients associated with political risk indicate that, as an acquirer’s political risk increases, the speed of deal completion decreases (longer duration to complete the deal). The hazard ratio for the political risk is significantly less than 1, indicating lower hazard rates and that M&As undertaken by acquirers engaging in high political risk take more time to complete. The result suggests that for a one-unit increase in the political risk of the bidder, the rate of deal completion decreases by 10.2% (1-0.898). Hence, this supports Hypothesis 2 and, more specifically, implies that bidders exposed to lower political risk are more likely to complete M&A transactions more rapidly. Additionally, we estimated the relationship between political risk and deal completion likelihood using multivariate ordinary least squares (OLS) regressions. We found similar results and they are reported in column 3 of Table 5.

As such, the estimates imply that firms delay M&As until political risk is resolved. It suggests that even if the bidders engage in M&A deals, it would take them more time to complete the deal when they face higher uncertainty.

In summary, our empirical evidence shows that bidders with higher political risk are less likely to complete bids. Indeed, completion speed is longer when bidders are faced with higher political risk. Given that completion speed and cancellation rates are considered a critical component of post-M&A success, firms with lower political risk are more able to integrate effectively and are more likely to generate synergies. They can also be more valuable to the target firm.

4.2.3 Channel Analysis: Financial Constraints

In this section we examine whether the negative relationship between firm-level political risk and deal completion likelihood is mainly driven by a firm’s financial constraints. Given that uncertainty increases the cost of capital (Xu, 2020) and, thereby, reduces the propensity of the firm to engage in M&A investments, we expect a more pronounced negative relationship between firm-level political risk and deal completion likelihood when the bidder is under financial constraints. Based on the previous literature, small firms experience more financial constraints (Frank & Goyal, 2003). Therefore, we used acquirer size as a proxy of financially constrained bidders (Alshwer et al., 2011; Hadlock & Pierce, 2010). It takes the value of 1 if the acquirer size has a below-median size, and 0 otherwise. Then, we ran the cross-sectional regressions for financially constrained and unconstrained bidder sub-groups; the results are presented in Table 6. Consistent with the hypothesis that firms tend to withdraw deals when they are under financial constraints, we found that the coefficient estimate of the bidder’s political risk is negative and statistically significant at the 10% level in column 2, while losing its significance in column 3. The negative effect of firm-level political risk on deal completion likelihood is consistent with the idea that financial constraint is a possible channel by which firm-level political risk influences deal completion likelihood.

5. Robustness Tests

Bidders may fail to complete the deal in response to other types of risk or uncertainty. Therefore, we checked whether the decision to complete an M&A deal is essentially associated with political risk or rather with other measures of risk by conducting falsification and placebo tests using non-political risk. The results are presented in columns 1 and 2 of Table 7. Interestingly, while the acquirer’s political risk decreased the likelihood of deal completion, non-political risk did not affect deal completion. As shown in Table 7, the coefficient associated with non-political risk is non-significant, indicating that takeover completion is mainly driven by political risk.

Despite using several control variables to reduce omitted variables bias in our main equation, our results may still comprise endogeneity issues derived from unobserved omitted variables. To address these endogeneity concerns, we performed an instrumental variable (IV) probit model using one instrument variable for political risk: the Partisan Conflict Index (PCI).^[10] Chatjuthamard et al. (2021) found that the degree of political risk is significantly correlated with the PSI. Therefore, the first condition satisfies the requirement for instrumental variables to be relevant. Moreover, the PCI is unlikely to be correlated with takeover completion except via its effect through political risk. There is no reason to believe that political risk could have a significant direct effect on takeover completion. Hence, the exclusion condition of instrumental variables is satisfied. Like Nguyen and Phan (2017), we performed two tests to confirm that our selected instrument is relevant: the weak identification test (the Cragg–Donald test) and the under-identification test (the Kleibergen–Paap test).

In Table 8, we report the first-stage results obtained from the IV probit model. Like Deng et al. (2013), we included bidder characteristics, and industry and year fixed effects in the first stage as our independent variables. The political risk taken by the acquirer is our dependent variable.

The results presented in Table 8 show that the correlation between the IV and the acquirer’s political risk is significant at the 1% level, which confirms the relevance of our instrument. Moreover, both the under-identification test and the weak identification test give more support to the relevance of our instrument. In Table 8, we also show in column 3 the second-stage results drawn from the IV probit model. Deal completion likelihood is used as the dependent variable, and the predicted variable of political risk, acquirer and deal characteristics, and industry and year fixed effects are the independent variables. We found that the coefficient associated with the predicted variable of political risk is negative and significant at the 1% level.

The results from the IV probit model are similar to those obtained from the main model, which means that our results are robust in controlling for endogeneity concerns.

6. Conclusion

This study demonstrates the importance of the intermediary phase of the M&A process. Some M&A deals fail to complete due to both the complexity of the bargaining process and new information revealed to the acquirer during this phase, which can affect the decision to engage in or complete an M&A deal. Acquirers, especially, tend to cancel bids before closing for regulatory reasons, or factors relating to policy or political uncertainty.

This research provides evidence of other factors of deal failure associated with the political environment of the bidder. Using a firm-level measure of political risk, we found that the political risk taken by the acquirer has a significant effect on both takeover completion and deal duration. M&As are one of the most salient of corporate investments; we conclude that the political risk undertaken by the acquirer is a critical factor which firms should consider when engaging in such investment decisions. Firms should also engage in activities that can reduce the negative impact of political risk on corporate investments and improve investors’ perception of firms’ valuation and trustworthiness. For example, bidders can manage their political risks by engaging in political lobbying, as political risk affects firms’ decisions to complete or to delay the deals (Choi et al., 2022). CSR is also an effective hedge against stock return volatility arising from political risk (Andriosopoulos & Deepty, 2022). Our results also suggest a positive relationship between deal duration and political risk, indicating that lower political risk eases merger integration. However, our study presents certain limitations. Decisions to cancel M&A deals are also related to the political risk of the target firm. Our study can be also extended by examining whether bidders can mitigate their political risk by engaging in political lobbying or by using CSR initiatives. An important issue can also be developed in future studies, offering further insight, by examining the effect of the political risk distance between bidders and targets on the likelihood of takeover completion.

Conflict of Interest Statement

The present contribution is free from any conflicts of interest, including all financial and non-financial interests and relationships.