Politically connected firms and corporate social responsibility implementation expenditure in sub-Saharan Africa: Evidence from Ghana

While previous research has emphasized the role of stakeholder pressures, firm-specific factors, as well as CEO characteristics as important drivers of corporate social responsibility (CSR) implementation, our understanding of how political connections impact small and medium-sized enterprises' (SMEs') CSR implementation expenditure is quite limited. In this study, we contribute to filling this gap by investigating the effects of political connections and CSR expenditure and explain the conditions that impact this relationship. Using data from 473 SMEs in Ghana, we find that political connections negatively influence CSR implementation expenditure. However, the negative effect is weakened when a firms' reputation and competitive CSR implementation pressures are high. Implications for theory and practice are discussed.


| INTRODUCTION
As corporate social responsibility (CSR) issues continue to dominate the academic literature and the popular business press, policymakers are taking steps to rain in on firms' social performance.Indeed, business organizations now consider their social performance as part of the overall business strategy.Accordingly, firms are using their strong CSR performance as a marketing tool to enhance social legitimacy and reputation (Brammer & Millington, 2008;Julian & Ofori-Dankwa, 2013).One important research stream within this perspective, undertaken primarily in transition economies, has explored the relationship between political connections and CSR activities (Huang & Zhao, 2016;Lin, Tan, Zhao, & Karim, 2015;Luo, 2006).These studies have used the social capital theory (Adler & Kwon, 2002;Bourdieu, 1986) to explain that, firms increase their CSR activities when they are politically connected.However, the existing literature is inconclusive concerning how political connections predict increases in firms' CSR expenditure.
Whiles the existing literature has focused on the large multinational corporations when it comes to discussion on CSR issues, recent debates are turning attention to small and medium-sized enterprises (SMEs) in institutionally uncertain environments (e.g., Luken & Stares, 2005;Lund-Thomsen, Lindgreen, & Vanhamme, 2016;Sethi, Shah, & Jan, 2018).Consequently, scholars have started to pay attention to CSR issues in institutionally challenging environments, such as sub-Saharan Africa (e.g., Julian & Ofori-Dankwa, 2013;Matten & Moon, 2008).For example, Julian and Ofori-Dankwa (2013) found that financial resource availability negatively impacts on CSR activities in Ghana, suggesting that the previous conclusion of most studies examining this relationship is inconclusive (Orlitzky, Schmidt, & Rynes, 2003).Thus, the drivers of CSR implementation may differ depending on contextual factors (Campbell, 2007;Matten & Moon, 2008) as different institutional environments influence norms and practices of CSR activities (Gjølberg, 2009).
Whiles previous studies have established that political connections positively influence firms' CSR performance (see Huang & Zhao, 2016;Luo, 2006;Zhang, 2017), we argue that politically connected firms situated in institutionally challenging environments may not positively increase their CSR activities.We argue that the universally agreed position regarding a positive effect of political connections on CSR performance may be premature as institutional environments in most developing countries are characterized by arbitrary enforcement of the law, bureaucratic inconsistency, insecurity of property rights.Given this situation, CSR may get a very different twist in these countries (Dobers & Halme, 2009;Kuznetsov, Kuznetsova, & Warren, 2009;Zou, Xie, Meng, & Yang, 2019).Besides, scholarly effort to delineate the boundary conditions of the relationship between political connections and CSR spending is lacking.This paucity of research is surprising given that many studies have highlighted the influence of political connections on firm behavior (Li & Atuahene-Gima, 2001;Zhang, 2017;Zhou, 2013).We test corporate reputation and CSR implementation pressure as potential boundary conditions of this relationship.As firms develop a reputation over time, they are likely to ensure the quality of CSR activities that they directly or indirectly pursue.This is likely to facilitate the promotion of firms' products and services.Related to CSR implementation pressure, firms tend to enhance their legitimacy by imitating successful competitors (Campbell, 2006).Therefore, these factors could pressure firms to pursue CSR activities to exploit new opportunities and create a competitive advantage in the marketplace to obtain superior performance.
This study contributes to the literature in three main ways.First, our study contributes to the social capital literature by showing how politically connected SMEs in a developing country pursue CSR activities.Specifically, we integrate the impact of social capital developed from social networking relationships and ties with government officials to explain a firm's CSR expenditure.This is an important extension of the literature given that previous studies (e.g., Dobers & Halme, 2009;Julian & Ofori-Dankwa, 2013;Lin et al., 2015) have overlooked this line of thought in developing countries.Second, examining CSR in a resource-poor context enables us to provide a nuanced understanding of impending factors of CSR spending.Third, by testing the role of corporate reputation and competitive CSR implementation pressure on the political connections-CSR expenditure relationship, we show the conditions under which social capital developed from the social networking relationships and ties with government officials effectively drive CSR expenditure in a resource-constrained environment.In doing so, we add to the debate on whether and, if so, how political connections promote or inhibit CSR expenditure within an institutional setting, namely, Ghana.This context enables us to provide a channel through which political networking affects CSR expenditure.
The rest of the paper is organized as follows.First, we present the theoretical framework and derive the hypotheses.Next, we describe the sample and data collection procedure.We then present the analyses and results of the study.We conclude the study with the discussion of findings as well as the theoretical and practical implications of the study.

| Social capital theory
Political network and ties are not uncommon in sub-Saharan Africa due to the collectivistic nature of cultures found in this region (Acquaah, 2007).Scholarly studies in organizational theory and industrial sociology tend to direct our attention to the role of managerial networking and interpersonal relationships in facilitating the exchange of valuable information, resources, and knowledge among entrepreneurs (e.g., Adler & Kwon, 2002;Laumann, Galaskiewicz, & Marsden, 1978).Our study draws on social capital theory (Bourdieu, 1986;Granovetter, 1985) to offer a better understanding of the effects of political networking arguing that political connections are important informal governance mechanisms that lessen the effect of institutional voids (Acquaah, 2007).Previous research shows that network ties are a crucial form of a managerial tool when legal and regulatory institutions are weak in enforcing economic exchanges (Granovetter, 2005;Sheng, Zhou, & Li, 2011).This study contends that because formal entrepreneurial-supporting institutions in emerging country settings are less effective (London & Hart, 2004), political connections (defined as entrepreneurs' social ties and connections with regulatory and governmental institutions) may be crucial for explaining CSR expenditure in a developing country setting.
While the value of inter-organizational networking activities may be less important in industrialized societies, they are critical in explaining variations in firm activities and performance creation in transition and developing economies (Guo, Xu, & Jacobs, 2014

| Political connections and CSR expenditure
Based on social capital theory, we posit that the effect of political connections on CSR performance is negative.The social exchange theory suggests that the relationships firms develop with political authorities (including ties with governmental officials and public officeholders) have the potential to determine the competitive advantage of firms (Getz, 1997;Hillman & Hitt, 1999).While some scholars have argued that many firms remain politically disengaged and tend to pursue an arms-length relationship with political leaders to avoid the accusation of political patronage (e.g., Baron, 1995;Yoffie, 1987), recent observations indicate that political actions of corporate entities are globally pervasive because of the obvious benefits it provides firms: (a) it enables firms to influence public policy and regulations (Oliver & Holzinger, 2008); and (b) it is a channel through which firms immerse themselves in their target markets as it signals a firm's legitimacy (Welsh & Young, 2011).
Accordingly, we argue that a firm's increased nonmarket activities can be leveraged to achieve strategic objectives (Schuler, Rehbein, & Cramer, 2002).Because public policy decisions have great potential to influence firms' current and future strategic directions, firms view their efforts to influence the political process as a matter of exercising their fundamental right to protect their shareholders' investments (Hillman, Keim, & Schuler, 2004).Based on the assumption that firms leverage opportunities and challenges in the nonmarket environment to their economic advantage (Baron, 1995) and given the potential opportunities associated with such nonmarket strategies with respects to expenditure on the sustainability courses (Julian & Ofori-Dankwa, 2013), we assert that the impact of political connections on firms' CSR expenditure is negative.There are several reasons why we expect the relationship to be negative.First, we reason that firms may rely on their relationships with political leaders to build and protect the value of their investments, such that not adhering to CSR concerns of the population may become less damaging to the firm when the firms are well connected in public policymaking corridors.For example, a firm can legally obtain exemptions from regulatory restrictions on employee rights and environmental emissions by lobbying political and public officeholders to enact policies that favor the firm's actions.In this direction, the political strategy management literature indicates that political networking is a viable option for firms to influence regulations, and even change the way the general public perceive CSR issues (Hillman, Zardkoohi, & Bierman, 1999).In other words, when firms are well connected to political and public policy officeholders, they can still enjoy a superior competitive advantage by spending less of their retained profits on social activities.Thus, we argue that: Hypothesis H1 For firms operating in a developing economy, political connections will negatively relate to CSR expenditure.

| Moderating role of firm reputation
An important aim of the present study was to explain the conditions under which social capital derived from government officials may be effective in driving a firm's CSR expenditure.Evidence shows that firms are getting ahead of competition and can substantially influence local, national, and international environmental regulations by hiring former politicians and diplomats to lobby policymakers to enact legislation that is supportive of firms' strategic goals (Lipton & Hakim, 2013).In China, knowing someone close to political leadership is seen as a major strategic asset for firms seeking favorable treatment from Chinese industry regulators (Sheng et al., 2011).This suggests that firms may be able to influence policies related to CSR and environmental practices and thereby refuse to abide by these regulations when they forge strainer ties with politicians.However, we suggest that because firms with a good reputation tend to prevent issues that will erode their good standing in society, they are likely to abide by the rules related to CSR and environmental practices.Accordingly, we introduce a firm reputation as a moderating variable of the relationship between political connections and CSR spending.A firm's reputation reflects the overall estimation of how the firm is held by its stakeholders based on its past actions to determine how the firm is likely to behave in the future (Walsh & Beatty, 2007).Arguably, a firm can show its concern about the environment by using an environmentally friendly strategy such as product offerings, equal opportunity employment, philanthropy, and other similar endeavors (Nguyen, Wu, Evangelista, & Nguyen, 2020).By employing such a strategy, the firm's reputation can be boosted as well as generating reciprocal goodwill in the long run (Brammer & Millington, 2005;Nguyen et al., 2020).
In this, we argue that the negative influence of political connections on CSR expenditure will be attenuated when the firm's reputation is greater.First, given that a good reputation influences positive responses from customers, firms will continue to boost their reputation through environmentally friendly product offerings, equal-opportunity employment, contributions to charitable causes, and other similar endeavors.This is because customers tend to measure the underlying quality and value of the firm's product offerings (Hansen, Samuelsen, & Silseth, 2008).Second, a good reputation tends to attract customers as most of these customers value their association with the reputation of the firm (Roberts & Dowling, 2002).Third, a good corporate reputation is likely to attract superior human resources which in turn could enhance the firm's competitive advantage (Wright, Ferris, Hiller, & Kroll, 1995).Given that firms with a good reputation are highly interested to attract candidates with skills, they are likely to reduce the political activities to main their reputation.Thus, a good reputation elicits good behavior by responding to stakeholders' demands including customers, employees, and suppliers.
The notion here is that firms with a good reputation cannot afford to ignore their primary responsibility toward CSR.Therefore, we propose that: Hypothesis H2 For firms operating in a developing economy, the negative effect of political connections on CSR expenditure will become weaker when the firm's reputation is greater.

| Moderating role of competitive CSR implementation pressure
The tendency for competitors to vie for resources is a major characteristic of market economies (Schwartz & Carroll, 2008).This competition results in pressures for firms to offer innovative products and services.
Competitive CSR implementation pressures are defined as the pressures emanating from competitors to embark on CSR activities (Helmig, Spraul, & Ingenhoff, 2016).We argue that since political connections tend to attenuate CSR implementation spending, pressures from competitors will stimulate firms to implement CSR activities.Given the pressures from competitors, it could be the case that CSR would change from a social or stakeholder obligation to a business obligation, which alters the entire competitive environment (Kowalczyk & Kucharska, 2020).In this regard, firms are likely to reduce their political networking activities and increase their effort to boost their CSR standing to stay competitive (Du, Bhattacharya, & Sen, 2007).Besides, when CSR implementation represents a competitive advantage, firms are likely to receive direct pressures to implement CSR activities (Dupire and M'Zali, 2018;Maignan, Ferrell, & Hult, 1999).This notion has been established in the CSR literature arguing that industry pressures for environmental responsibility can also result from competitors' actions (Boiral, Heras-Saizarbitoria, & Testa, 2017;Christmann, 2004, p. 749).
A major rationale is that firms tend to imitate their successful competitors to gain legitimacy.Thus, we argue that: Hypothesis H3 For firms operating in a developing economy, the negative effect of political connections on CSR expenditure will become weaker when competitive CSR implementation pressure is greater.

| Study setting: Ghana
The choice of Ghana as the study's setting was informed by two reasons.First, Ghana, in some respects, is considered a representative of sub-Saharan African emerging economies that also shares the challenges of doing business in less developed societies (Hoskisson, Eden, & Wright, 2000;Julian & Ofori-Dankwa, 2013).Second, Ghana is characterized by high information asymmetry due to institutional and infrastructural underdevelopment (Abor & Biekpe, 2006), resulting in a requirement for firms to go beyond formal codes of business transaction to incorporate informal interpersonal relations to be successful (Peng & Luo, 2000).Hence, Ghana is a useful case to study firm behavior in a more challenging circumstance.

| Sample and data
The sampling frame for this study was developed from Ghana's company register database (Acquaah, 2007).We selected a sample of 1,000 small and medium-sized enterprises (SMEs) from this register.
The sample met the following: (a) firms with complete contact information on the CEO, (b) firms that are manufacturers of physical products or service providers that engage in productive business activities (c) firms employing a maximum of 250 employees to SMEs (Ghana Statistical Service, 2000).To obtain a high response rate, we promised the CEOs that information about respondents and companies would be kept in strictest confidence.The respondents were also promised a summary of the research findings.
Data were collected in two waves.In the first wave, the CEOs of the 1,000 SMEs were contacted with a questionnaire delivered inperson to collect data on political connections, firm reputation, CSR implementation pressure, and the control variables.Out of the 1,000 questionnaires, we received 493 responses and 9 questionnaires were discarded due to missing values.Thus, 484 complete responses; representing 48.40% were valid in the first phase.
Next, in wave 2, we contacted finance managers/chief accountants of the 484 SMEs using the door-to-door approach to collect data on CSR implementation expenditure.We found that 11 of the firms had no finance managers or the CEO was also the finance managers.Thus, we relied on 473 matched responses (47.30%) for the analysis.
The average number of full-time employees was 15 and the average annual turnover was US$841,129.The average age of the firms was 7 years.The participating firms represented two industries: manufacturing (65%) and service (35%).To assess nonresponse bias, early and late respondents were compared in terms of firm age, firm size, CEO tenure, and education.The results show no significant difference in the study constructs.This suggests that nonresponse bias does not significantly affect the data (Armstrong & Overton, 1977).

| Measures
All the construct measures were derived construct measures from previous studies and measured on 7-point scale.Table 1 presents specific items used and their respective factor loadings and t-values.
Political connections.We measured political connections using a four-item scale from Acquaah (2007).This scale measures network relationships with government officials and politicians.We asked the CEOs to indicate how extensive (or how little) their ties were to different types and levels of bureaucratic and government officials.
Firm reputation.We used six items from previous studies to capture firm reputation (Fombrun et al., 2000;Rettab et al., 2009).This scale has been recently found to be robust (Nguyen et al., 2020).
Competitive CSR implementation pressure.We adopted five items from Helmig et al. (2016) to measure competitive CSR implementation pressure on a Likert scale ranging from 1 = strongly disagree and 7 = strongly agree.
CSR implementation expenditure.This construct was measured as the expenditure of the firm's spending on CSR activities as a percentage of sales, return on investment, and profit; thus, providing a more reliable measure of the actual extent of the commitment of companies with CSR activities (Julian & Ofori-Dankwa, 2013).Overall, we used three items to capture CSR expenditure.Finance managers/chief accountants were asked to state the percentage of return on investment, return on sales, and profit margin spent on CSR activities.

| Common method bias, validity, and reliability
To assess common method variance in the data, we followed Cote and Buckley (1987) and assumed a conservative single-method bias using confirmatory factor analysis (CFA) to test three alternative hierarchically nested measurement models.Accordingly, we estimated three competing models.In Model 1, we estimated the method-only model in which all indicators loaded on a single latent factor.Next, Model 2 estimated a trait-only model where each indicator loaded on its respective latent factor.Finally, Model 3 was a method and trait model involving the inclusion of a common factor linking all the indicators in Model 2. When the three models were compared, we found that Model 2 and Model 3 are better than Model 1 and that Model 3 is not substantially superior to Model 2. Table 2 presents the common method bias nested models.Collectively, we assumed that the variance in responses can be explained by the simultaneous effect of traits, method, and random error.Thus, common method bias does not sufficiently describe the data.
Next, we established the validity and reliability of all the multiitem constructs.Accordingly, we tested the measurement model in CFA using LISREL 8.71 with the maximum likelihood estimation procedure.All the items loaded significantly on the expected constructs, indicating convergent validity and discriminant validity of the measures (Montoya-Weiss, Massey, & Song, 2001).Table 1 displays specific items used to measure the constructs and their respective factor loadings and t-values.All values for composite reliabilities (CR) were significantly larger than 0.60, the level considered as evidence for convergent validity (Bagozzi & Yi, 2012).Besides, we received Cronbach's α values that were higher than the suggested cut-off value .70 for all constructs (Fornell & Larcker, 1981).Overall, the fit indexes showed that the model fit the data reasonably well.The fit indices were adequate: χ 2 /df = 1.33,RMSEA = 0.06; NNFI = 0.96; CFI = 0.96.Besides, all the constructs had accepted levels of reliability, average variance extracted (AVE), and discriminant validity (Table 1).(Cohen & Cohen, 1983).To attenuate the treat of multicollinearity, the independent and moderating variables were mean-centered before the interaction terms were created (Aiken & West, 1991).The largest variance inflation factors (VIF) was 2.11, which is lower than the critical value of 10, suggesting that multicollinearity is not a concern in the data (Neter, Kutner, Nachtsheim, & Wasserman, 1996).
Table 4 contains the results of the regression models.Model 1 contains the results of the control variables.Model 2 added the direct effects variables.Hypothesis 1 proposed that, political connections will negatively relate to CSR expenditure.We find support for this hypothesis as the coefficient of CSR expenditure is negative and significant (β = −0.18,p < 0.01).
Hypothesis 2 states that the negative influence on political connections on CSR expenditure is weaker when the firm's reputation is greater.The finding in Model 3 shows that the interaction term for political connections and firm reputation is positive and significant (β = −0.37,p < 0.01).To gain further insights into the direction of the moderation impact, we followed the guidelines advanced by (Dawson & Richter, 2006) and created an interaction plot using mean-centered values.The graph of this interaction (Figure 2) shows that the negative relationship between political connections and CSR expenditure is weaker when the firm's reputation is high but not when it is low.
Thus, we received support for 2.
Hypothesis 3 proposes that the negative effect of political connections on CSR expenditure is weaker when competitive CSR Findings of regression analyses (dependent variable: CSR expenditure) (N = 473)  F I G U R E 2 Interaction effect of political connections with a firm reputation on corporate social responsibility (CSR) expenditure implementation pressure is high.The result in Model 3 indicates that the interaction of political connections with competitive CSR implementation pressure prevention focus is significant and negative for CSR expenditure (β = −0.34,p < 0.01).The graph of this interaction (Figure 3) demonstrates that the relationship between political connections and CSR expenditure is weaker when competitive CSR implementation pressure is high but not when it is low.Therefore, the result in Model 3 supports Hypothesis 3.

| Robustness tests
We substantial the robustness of the findings, we performed additional analyses.First, we tested a three-way interaction between political connections, firm reputation, and competitive CSR implementation pressure in Model.As shown in Model 4, this interaction is significant and negative (β = −0.53;p < 0.01).Furthermore, as shown in Figure 4, the negative relationship between political connections and CSR expenditure is weaker when both firm reputation and competitive CSR implementation pressure are high.Data points for plotting  Dimov, & Belausteguigoitia, 2016).However, the interaction terms in the main model and interaction term models show consistency.This provides support for the robustness of our regression models (Covin, Green, & Slevin, 2006).

| DISCUSSION AND CONCLUSION
Using insights from the social capital theory (Adler & Kwon, 2002;Bourdieu, 1986), we examined the influence of political connections on CSR expenditure of SMEs operating in a developing country-Ghana.Besides, we tested the conditions under which political connections may effectively influence SMEs' CSR implementation expenditure.In particular, the first finding (political connections negatively relate to CSR expenditure) indicates the role of firms' political ties in CSR activities in a developing country.The second finding of this study (i.e., firm reputation moderates the relationship between political connections and CSR expenditure) shows the importance of the prior neglected role of firm reputation in explaining when political connections could be beneficial for a firm's CSR activities.By integrating insights from the social capital theory and corporate reputation literature (Janney & Gove, 2011;Roberts & Dowling, 2002), this study argues that corporate reputation is a valuable resource for converting political networking relationships into greater CSR spending in a developing country.Third, the role of competitive CSR pressure is hypothesized and tested in Hypothesis 2 as a moderator of the relationship between political connections and CSR expenditure.Given that competitors tend to vie for resources, the resultant effect puts pressure firms to innovate to enhance their product outlook in the market (Helmig et al., 2016).Therefore, given the CSR activities in the firm's competitive environment, it may be the case that firms with stronger political connections would pursue CSR activities when competitive pressure to implement CSR is greater especially if implementing CSR represents a competitive advantage (Du et al., 2007).Put together, these findings have several critical contributions for both theory and practice.
First, our findings extend our understanding of the role of political connections in impeding CSR expenditure in developing country contexts.The political ties literature has conventionally argued that when firms cultivate stronger relationships with government officials and bureaucrats, it enables access to resources for firms (Acquaah, 2007) and this translates into greater CSR spending (Huang & Zhao, 2016;Luo, 2006;Zhang, 2017).In contrast, we find that when firms Second, given that our sample comes from SMEs, our results contribute to the political networking literature by demonstrating that forming closer political ties with politicians, judges, and governmental officials impedes SMEs' CSR implementation spending.Political connections have been traditionally investigated in the contexts of large listed multinationals (Luo, 2006;Zhang, 2017).Extant knowledge of the roles of political connections in SMEs operating a developing country is very limited.Our study suggests that political connections are counterproductive in driving SMEs' CSR expenditure in developing countries.
Third, we advance scholarly understanding of the boundary conditions of the effects of political connections.Specifically, we test two conditions (firm reputation and competitive CSR implementation pressure) under which the negative effect of political connections on CSR spending is attenuated.To the best of our knowledge, our study is among the first to test empirically the moderating role of these two conditions.Particularly, the findings in Hypothesis 2 and Hypothesis 3 suggest that firm reputation and competitive CSR implementation pressure weaken the negative effect of political connections.
Beyond the theoretical contributions, our study has some practical and policy implications.First, the findings can guide developing country managers to improve their firms' CSR implementation spending.In particular, the finding that political connections impede CSR spending can help SME managers to moderate excessive political ties in developing countries.Second, findings from our study have implications for understanding how firm reputation and CSR implementation pressure attenuate the negative impact of political connections on CSR implementation spending.This can guide companies to devise effective CSR strategies in developing economies.Overall, the importance of the research topic and context suggests that this study can extend our theoretical understanding and to guide managerial implications as well across other transforming economies.

| LIMITATIONS AND FUTURE RESEARCH
Our study has some limitations that open avenues for future research.
The conceptual model of the study [Colour figure can be viewed at wileyonlinelibrary.com] et al., 2011).Therefore, integrating insights from organizational social capital theories and CSR, we examine the relationship between political connections and CSR expenditure under differential levels of firm reputation and competitive CSR implementation.Our conceptual model is presented in Figure 1.
firm age, firm size, financial resource slack, industry, CEO tenure, and education.Firm size was measured with the number of full-time employees, and firm age was captured as the number of years the business has operated since its first sales.Financial resource slack was captured by four items following Atuahene-Gima et al. (2005).The industry was measured with a dummy variable with "0" indicating the manufacturing industry and "1" indicating otherwise.CEO tenure was

Figures 2
Figures 2, 3, and 4 were computed using +/− 1SD for political connections, firm reputation, and competitive CSR implementation pressure.Second, we extended our analysis beyond the usually examined mean centering where variance inflation factor (VIF) values are used as proxies for detecting multicollinearity by including all the interaction terms concurrently (Model 5) in the regression equation.The results are in the direction of hypothesized specifications.Prior research suggests that when interaction terms with common variables are concurrently included in a model, they can conceal the detection of the true moderating variables as a result of multicollinearity (e.g., De Clercq, High firm reputation, High competitive CSR pressure (2) High firm reputation, Low competitive CSR pressure (3) Low firm reputation, High competitive CSR pressure (4) Low firm reputation, Low competitive CSR pressureF I G U R E 4 Interaction of political connections, firm reputation, and competitive corporate social responsibility (CSR) pressure on CSR expenditure cultivate greater relationships with political authorities, firms' expenditure on CSR activities reduces significantly.This is because firms that maintain high levels of institutional ties (e.g., connection to politicians, judges, governmental officials, country-level resource allocators) tend to be counterproductive as it has the potential to lead to the accusation of nepotism, corruption, and favoritism, which may also negatively impact on CSR activities.We provide a more nuanced understanding of how political ties influence firms' CSR activities in institutionally precarious environments such as those in developing countries.Thus, we add to the "Institution Difference Hypothesis" (IDH)(Julian & Ofori-Dankwa, 2013), suggesting that the effects of political connections on CSR expenditure are different depending on contextual conditions.

First
, the cross-sectional nature of our sample precludes us from making causal claims.Future research could use longitudinal data, collected over time to help alleviate causality concerns.Second, this study was conducted in the context of SMEs in Ghana.Consequently, the generalizability of the study's findings is limited to and extrapolating the results reported here to other countries should be done with caution.We encourage future researchers to examine our conceptual model in a developed country where SMEs ae more resourceful(Engelen, Kube, Schmidt, & Flatten, 2014).Third, beyond competitive CSR implementation pressure and firm reputation examined in this study, further research could consider other moderators of the political connections CSR expenditure relationship.For example, future studies could examine potential environmental and firm-level moderators such as environmental dynamism and uncertainty.Fourth, future research could address gender differences in political connections-CSR expenditure in SMEs.Finally, future studies could also seek to identify mediators of the effects of political connections on CSR.One of these might be inter-organizational cooperation.Firms with high networking relationships may corporate with other firms to undertake CSR activities.The present research suggests several potentially valuable avenues for future studies, which must be tackled to advance current knowledge on the subject.In conclusion, we investigated the influence of political connections on CSR implementation expenditure.Besides, we examined two conditions under which greater political connections translate in greater CSR expenditure in SMEs.The findings from the study show that the negative relationship between political connections and CSR expenditure is attenuated when both firm reputation and competitive CSR implementation pressure are greater.Our conceptual model, which links political connections, firm reputation, competitive CSR implementation pressure, and CSR expenditure, was empirically validated using data from 473 SMEs operating in Ghana.The findings theoretically add to the CSR literature by clarifying when SMEs' nonmarket strategies could help enhance CSR activities.

Table 3
presents the correlations and descriptive statistics.Hierarchical regression was used to test the hypotheses