Corporate Political Strategies in Weak Institutional Environments: A Break from Conventions

There is a lack of research about the political strategies used by firms in emerging countries, mainly because the literature often assumes that Western-oriented corporate political activity (CPA) has universal application. Drawing on resource-dependency logics, we explore why and how firms orchestrate CPA in the institutionally challenging context of Nigeria. Our findings show that firms deploy four context-fitting but ethically suspect political strategies: affective, financial, pseudo-attribution and kinship strategies. We leverage this understanding to contribute to CPA in emerging countries by arguing that corporate political strategies are shaped by the reciprocity and duality of dependency relationships between firms and politicians, and also by advancing that these strategies reflect institutional weaknesses and unique industry-level opportunities. Importantly, we shed light on the muttered dark side of CPA. We develop a CPA framework and discuss the research, practical and policy implications of our findings.


Introduction
Africa's business and institutional environments are both uncertain and volatile, largely due to governments' actions and inactions. While some firms adapt to these environments, others proactively engage in corporate political activity (CPA) 1 to shape their competitive space (Mbalyohere et al. 2017;Wocke and Moodley 2015). Within the CPA literature, researchers mainly draw from Hillman and Hitt's (1999) political strategy typology, namely constituency building, financial incentive and information (e.g. Barron 2011; Shirodkar and Mohr 2015b; Wan and Hillman 2006). Constituency building focuses on building coalitions to affect policy issues, financial strategy entails donating funds to politicians or political parties and information strategy, as the name suggests, utilizes information to lobby politicians. These strategies, to a large extent, are formal ways of influencing government policy in strong institutional contexts. In Africa, weak institutions make some of these strategies inapplicable (Liedong and Frynas 2018). Even if they are applicable, their deployment is often fraught with unethical practices (Lawton et al. 2013a). For instance managerial political ties, 2 the most prevalent mode of business-government relations in emerging countries , are largely associated with poor corporate governance, corruption, bribery, cronyism and other social ills (Johnson and Mitton 2003;Liedong and Rajwani 2018;Okhmatovskiy 2010;Sun et al. 2016).
Subsequently, while there is an extensive coverage of managerial political ties in emerging countries, research has 1 3 sparsely explored the full breadth of other political strategies used by firms in these countries, especially those in Africa. The reason underlying this phenomenon is rooted in the widely held belief that because most emerging countries have relationship-based institutional structures owing to their collectivist cultures (Acquaah 2007;Fraser et al. 2006;Sun et al. 2011), informal and personal ties to political actors are the main ways of influencing government policy (Liedong and Frynas 2018;Peng and Luo 2000;White et al. 2015). Hence, there has been little effort to explore beyond political ties, which consequently limits our understanding of both the strategic and ethical implications of firms' political strategies in emerging countries. In this respect, there is a high chance that other institutionally driven political tactics have yet to be uncovered.
Recently, arguments have been advanced to show that emerging market institutions have strengthened to the extent that firms are able to use a wider array of western-oriented political strategies such as constituency building and information (Shirodkar and Mohr 2015a). Indeed, while this may be the case for India, China and other advanced emerging countries, it may not be the same for frontier countries that are still battling with basic institutional voids. In most African countries, there is hardly an interface for deploying an information strategy , neither are the electoral laws effectively enforced to make political campaign financing transparent (Liedong 2017). These lapses limit the scope of political strategization.
Nevertheless, firms in emerging markets, amid severe institutional constraints, may devise innovative and contextfitting strategies to wield political influence. Research has conceptually mooted that CPA is influenced by national government systems and managers' cultural socialization (Barron 2011;Hillman and Keim 1995). Broadly, what this suggests is that CPA is context-dependent, mainly as firms will design, organize and implement their political strategies in ways shaped by the conditions prevailing in their external environments. However, despite this postulation, there is still a lack of research that explore CPA in developing and weak institutional contexts, especially in Africa where various management phenomena remain under-explored (George et al. 2016).
The above gaps in the literature evoke the question: why and how do firms conduct their CPA in weak institutional contexts? We answer this question using interview data from bank executives, which we corroborate with secondary data (i.e. media articles, news blips) about politicians' connections and dealings with banks. The secondary data allowed us to overcome constrained access to politicians, while accounting for both sides of the dyadic bank-politician relationship.
Our study draws upon resource-dependency theory (Pfeffer and Salancik 1978). According to this theory, firms are dependent on the contingencies in their external environments, and their behaviour does not only reflect this dependency, but it also attempts to reduce the uncertainty associated with dependency (Hillman et al. 2009). Extant research recognizes that governments and politicians, due their regulatory power, are sources of dependency for firms (Schuler 1996;Lester et al. 2008;Hillman 2005;Hillman et al. 2000;Keim and Zeithaml 1986;Meznar and Nigh 1995). This is particularly true in developing countries where politicians have a stronger control over access to critical resources and valuable information (Guo et al. 2014;Peng and Luo 2000;Acquaah 2007). However, resource-dependency theory, as used in CPA studies, mainly portrays a one-way relationship between firms and politicians in which the former are dependent on the latter. In other words, the usage of the theory recognizes firms' dependency on external parties, but not external parties' dependency on firms. This fails to show the dynamic relations between firms and politicians whereby the latter can be dependent on the former for resources such as information and money to respectively formulate policy and finance election campaigns (Hillman and Hitt 1999;Bonardi et al. 2005;Lawton et al. 2014). In weak institutional contexts such as Africa where political campaign financing is opaque (Liedong 2017), vote buying is prevalent (Vicente and Wantchekon 2009;Vicente 2014), and electoral success depends on candidates' financial endowment (Collier and Vicente 2012), it is highly likely that CPA will be shaped by an overlooked, under-explored and ethically suspect two-way dependency relationship between firms and politicians.
This study makes significant contributions to the CPA literature in developing countries. First, it reveals that political strategization is highly context-dependent. In this sense, western-oriented CPA is starkly different from CPA in weak institutional environments. Building on De Villa et al. (2018) and advancing the dark side of CPA, we find that Nigerian banks align their CPA with their politico-socio-cultural context, which leads to the deployment of unique strategies that have previously been undocumented. More importantly, we found that banks orchestrate four distinct but ethically suspect strategies-Affective, Financial, Pseudo-Attribution, and Kinship strategies. Using primary and secondary data to capture the bank and politician fronts respectively, we show that institutional weaknesses offer unique opportunities for firms and industries to deploy political strategies that exploit their reciprocal or dual dependency relationships with politicians. In doing so, we help to address the problem of resource-dependency theory being applied in ways that have "largely focused on the dependence of one actor on the other without considering the reciprocal dependency" (Casciaro and Piskorski 2005, p. 169). This problem is very common in the way CPA is conceptualized and operationalized. Our study thus highlights the dyadic nature of resource dependency and demonstrates how the "interdependence" (Pfeffer and Salancik 1978) or "mutual dependence" (Casciaro and Piskorski 2005) dimension of resource-dependency theory manifest in CPA.
Second, this study suggests that political strategies may be industry-specific. The banking industry has a unique opportunity to manage the finances of politicians in return for policy favours. In this respect, banks in corrupt and institutionally voided countries can use financing tactics that are not available to firms in other industries. The use of these financing tactics is underpinned by a reciprocal or a two-way resource-dependency relationship whereby banks depend on politicians for regulatory and operational favours (as revealed in our primary data) while politicians depend on bankers for ethically suspect financial favours (as shown in our secondary data). The former reflects the common one-way portrayal of firms as actors that always depend on politicians (e.g. Guo et al. 2014;Hillman 2005;Hillman et al. 2000;Shirodkar and Mohr 2015a), but the latter demonstrates that firms in industries that are able to offer direct benefits or provide economic rent to politicians have exclusive opportunities to exploit reciprocal dependency in their CPA.
It is worth noting that a small body of research has examined industry-specific CPA. For instance, Schuler (1996) found that between 1976 and 1989 when the U.S steel industry faced strong import pressures, firms used petitions and congressional testimonies to seek protection from the government. Lawton et al. (2013b) also documented how state-owned and non-state-owned airlines in the EU, in the wake of deregulation, leveraged human, network and structural resources to organize their political capabilities. Other studies have examined CPA in the electric utilities (Bonardi et al. 2006) and automobile (Hersch and McDougall 2000) industries. We extend this nascent conversation by not only showing CPA in the banking industry, but also suggesting that some political strategies could be unique or exclusive to certain industries. While our results speak to mainly the banking industry, it is clear in this instance that the opportunity to provide "preferential" financial services to politicians, which is solely available to the financial services industry, provides a conduit for political influence in weak institutional contexts.
Third, this study provides empirical evidence of the complementarity between CPA and social responsibility Hond et al. 2014) and the likelihood of corporate social responsibility (CSR) influencing political access (Hillman et al. 2004;Mellahi et al. 2016;Wang and Qian 2011). It shows that CSR is actually a type of political strategy in Nigeria, just as it is in China (Jia and Zhang 2016;Li et al. 2015;Su and He 2010). We find further evidence showing that when CSR is used as a political strategy versus when it is done voluntarily, social programs are carefully developed and effectively executed. Importantly, we depart from prior research to demonstrate that CSR may serve as a political strategy only when it boosts politicians' re-election chances. We therefore add new knowledge to the bourgeoning CPA-CSR complementarity debate.
Finally, this study makes a methodological contribution by challenging the current conceptualization of political connections. Most studies define politically connected firms as those whose senior managers have relationships with politicians (Peng and Luo 2000;White et al. 2015) or those that have politicians on their corporate boards (Hillman 2005;Hadani and Schuler 2013). As this study reports, there are instances when the connections between firms and politicians are embedded in ordinary employees, not directors, senior managers or large shareholders. Our findings show that in the affection strategy, female employees (mostly account officers, not top managers) are the banks' connection to the polity. Similarly, in the kinship strategy, ordinary employees from politicians' families, communities or constituencies provide the leverage or link for banks to build political capital. While this is the case in Nigeria, we believe it will be similar in other relationship-based emerging countries. On the ethical front, this phenomenon shrouds political embeddedness and makes it harder to unearth shady political dealings. Overall, this study sheds light on the dark side of CPA, and hopes to inspire managerial and policy actions that will strengthen Africa's institutions to curb corruption and other unethical practices.
The rest of the study proceeds as follows. In the next section, we review CPA literature and discuss the resourcedependency theory. We then present the study's methodology and discuss the findings. We conclude by outlining the study's theoretical and practical implications and suggesting avenues for future research.

CPA in Emerging Countries: Concept, Drivers and Strategies
CPA refers to "corporate attempts to shape government policy in ways favourable to the firm" (Hillman et al. 2004, p. 838). It is "any deliberate firm action intended to influence government policy or process" (Getz 1997, pp. 32-33). Due to the impact government decisions have on business (Stigler 1971;Weidenbaum 1980), it is widely believed that firms may struggle to gain sustainable competitive advantage if they do not pay attention to political processes and dynamics within their operating environments (Baron 1995;Schuler 1996).
In emerging countries, one of the foremost drivers of CPA is resource dependency, mainly as political connections serve as substitutes for weak and absent market supporting institutions (Peng and Heath 1996;Xin and Pearce 1996). In these countries, voids in the product, labour and capital markets (Khanna and Palepu 2010) suggest that firms experience severe constraints such as information asymmetry and high transaction costs (Khanna and Rivkin 2001;Meyer 2001;Tong et al. 2008). Additionally, the regulatory environments in these countries are likely to be more restrictive (Ang and Michailova 2008). Further, the governments in these countries play a more prominent role in resource allocation (Acquaah 2007;Park and Luo 2001;Peng and Luo 2000). In the midst of these conditions, firms in emerging countries depend on political actors for competitive resources (Guo et al. 2014) and preferential treatment in capital and policy markets (Dieleman and Sachs 2008;Fisman 2001;Johnson and Mitton 2003;Khwaja and Mian 2005). Essentially, political connections are crucial for obtaining critical resources and facilitating economic exchange (Acquaah 2012), thus the need to establish and nurture them.
Resource-dependency theory posits that even though firms cannot use CPA to reduce their dependency on politicians (Getz 1997), they can use CPA to reduce the uncertainties associated with dependency (Pfeffer and Salancik 1978;Hillman et al. 2009). Indeed, studies using this theory tend to argue that firms do CPA to address their exposure to environmental uncertainty (Mellahi et al. 2016;Hillman et al. 2000;Wei 2006). Emerging countries have risky environments (Puck et al. 2013;Liedong et al. 2017;White et al. 2015), which makes business failures highly likely Nell et al. 2015). Firms, in the quest to mitigate their risk exposure, develop political connections (Liedong and Frynas 2018). These connections help them to obtain superior information and reduce information asymmetry in government policy discourses. They also offer them protection from bankruptcy and failure through bailout programs (Faccio et al. 2006) and lenient regulatory enforcements (Wu et al. 2016). Importantly, political connections enable firms to overcome administrative obstacles such as bureaucracy and red tape-constraints that are replete in emerging markets (Liedong and Frynas 2018).
Building on the drivers of CPA, several scholars have found that CPA leads to improved firm performance (see Hillman et al. 2004;Lux et al. 2011;Mellahi et al. 2016;Rajwani and Liedong 2015). In this respect, it can be argued that competitive advantage is the ultimate latent motivation for CPA. This motivation lies at the centre of every business organization, and has led some firms to establish divisions and offices specially dedicated to government affairs (Brown 2016;Lawton et al. 2014). Some scholars have also advocated for upper echelons restructuring so that firms can better handle policy and social issues ). These scholars contend that the ability to develop political capabilities underscores strategic agility, dynamism and ultimately competitive advantage (Kotabe et al. 2017;Lawton et al. 2013b).
Indeed, CPA may improve firms' bottomlines. However, CPA is also associated with poor governance and unethical practices. In emerging countries, CPA is often fraught with bribery, corruption and cronyism (Domadenik et al. 2016;Lawton et al. 2013a;Johnson and Mitton 2003). The absence of strong institutions to regulate business-government relations in these countries gives rise to clandestine and illegal dealings between public officials and firms. Politically active and connected firms also tend to be associated with poor financial disclosure practices (Chaney et al. 2011;Liedong and Rajwani 2018), financial misrepresentation (Guedhami et al. 2014) and excessive risk-taking (Boubakri et al. 2013;Bliss and Gul 2012). Moreover, CPA may serve as a means for managers to use corporate resources to pursue their personal political interests to the detriment of shareholder value (Okhmatovskiy 2010;Gao et al. 2017), thus creating agency problems (Carretta et al. 2012;Sun et al. 2016). Overall, CPA is not wholly positive. It only benefits a small and privileged minority of firms that are able to deploy political influence to unduly exploit public resources and enhance their competitive advantage (e.g. Capron and Chatain 2008;McWilliams et al. 2002).

Strategies and Antecedents of CPA
To gain competitive advantage, firms "may use political means to alter the condition of the external economic environment" (Pfeffer and Salancik 1978, p. 190). There are various ways of achieving this environmental alteration. In their seminal work, Hillman and Hitt (1999) proposed three political strategies, namely constituency building, financial incentive and information. Constituency building helps to build strong coalitions and public support for policy influence (Baysinger et al. 1985;Lord 2003) while financial incentive provides monetary inducements to political decision makers to sway them towards the interests of donating firms. Examples of financial tactics include political action committee (PAC) contributions and political directorships (Milyo et al. 2000;Stratmann 1991). Information strategy, as the name suggests, entails using information to shape policy decisions. Specific information tactics include lobbying, research reports, and press conferences. Several studies have empirically examined these strategies (e.g. Hadani and Schuler 2013;Hillman and Wan 2005;Marsh 1998;Schuler et al. 2002;McKay and Webb-Yackee 2007), but of course these studies were mostly conducted in developed settings where frameworks and regulations exist to control CPA.
In emerging countries however, some of Hillman and Hitt's strategies are largely inapplicable due to the lack of supporting institutional structures (Liedong and Frynas 2018). In fact, the most prevalent political strategy in these countries is the relational strategy . It involves personal-level connections between managers and politicians (see Acquaah 2007;Li et al. 2008;Peng and Luo 2000;White et al. 2015). Lacking any formal structure, this strategy is less transparent and highly susceptible to corruption, bribery, cronyism, favouritism, and clientelism. As it currently stands, we have sufficient knowledge about personal-level political connections as well as their ethical implications, but we still do not know whether emerging market firms use other political strategies.
More importantly, the theorization and instrumentation of CPA strategies is mainly based on the statement that "firms are dependent upon the government, therefore they will engage in CPA" (Hillman et al. 2009(Hillman et al. , p. 1412. Indeed, studies using resource-dependency theory tend to suggest a one-way dependency relationship between politicians and firms (e.g. Guo et al. 2014;Lester et al. 2008;Hillman 2005;Wei 2006) whereby, according to Baysinger (1984), the latter manages government actors to achieve domain defence (i.e. rebuff threats to organizational goals) and domain maintenance (i.e. rebuff threats to the methods through which goals are achieved). Generally, resource-dependency theory assumes that power and dependency are opposite ends (Getz 1997). In this sense, politicians never depend on firms. However, in reality, government actors depend on firms and other interest groups for information to formulate policy (Kingsley et al. 2012;Bonardi et al. 2005). Similarly, politicians depend on firms for campaign finance and donations (Hillman and Hitt 1999;Getz 1997).
Therefore, it is not always true that the relationship between firms and politicians is unbalanced. Instead, political strategies could reflect and highlight interdependence between firms and the polity, more so in developing countries characterized by collectivist cultures (Barron 2011), relationship-based economic and social systems (Acquaah 2007;Peng and Luo 2000), and weak or non-existent CPA regulations . In Africa, political campaign financing is opaque (Liedong 2017), vote buying is prevalent (Vicente and Wantchekon 2009;Vicente 2014), and electoral success depends on candidates' financial endowment (Collier and Vicente 2012). These conditions create opportunities for firms to provide the resources that politicians need, and by doing so they can exploit interdependency to orchestrate and deploy unique political strategies. However, due to the static nature and application of the resource-dependency theory over the last four decades (Hillman et al. 2009;Sun et al. 2016), such new perspectives have not been explored.
Decades ago when the CPA literature began to grow, there was an interest to explore the antecedents of political intensity and political strategy choice. Several studies have since documented firm-level (e.g. Hillman 2003; Hansen and Mitchell 2000; Mohr 2015a), industry-level (e.g. Gray and Lowery 1997;Grier et al. 1994;Schuler 1996), and nation-level (e.g. Barron 2011;Blumentritt 2003;Hillman and Keim 1995;White et al. 2015) CPA determinants. Noteworthy of these determinants is industry regulation. Research suggests that highly regulated industries tend to be more politically active (Grier et al. 1994;Hadani and Schuler 2013) but beyond this knowledge, we know little about the political strategy choices of highly regulated firms. Bringing this into context, the highly regulated banking industry lacks representation in the CPA literature, which consequently leaves crucial questions unanswered. Considering the strategic role of financial institutions in credit-constrained emerging markets and the importance of money in African politics (Collier and Vicente 2012), do banks have a unique arsenal of political strategies at their disposal? Does the dependency of institutional actors (including politicians) on financial institutions for funds and financial services create opportunities for banks to deploy previously undocumented political strategies? The foregoing leads us to ask the following research question: Why and how do firms conduct their CPA in weak institutional contexts? We answer this pressing research question within the emerging context of Nigeria, and comment on the actual and potential ethical implications of firms' CPA.

Research Context
Nigeria is the world's sixth major oil producing nation. It is Africa's biggest oil exporter and largest economy with an average GDP growth rate of 5.7 percent between 2006 and 2016 (The World Bank 2018). It is home to some of Africa's largest multinational enterprises, including banks. Despite its strong fundamentals, Nigeria's economy has been hobbled by high unemployment, weak infrastructure, security risks, inadequate electricity supply, and untrustworthy dispute resolution mechanisms among other issues. Notably, the high rate of corruption prevalent in the country may have inflamed these institutional problems (PwC 2016). Nigeria records widespread corruption at all levels of government. Besides federal-level graft and political interferences in the work of anti-corruption organizations (Akume and Okoli 2016), there is significant corruption at the State level. For instance, James Ibori, former Governor of Nigeria's Delta State received a 13-year jail sentence in the UK for fraud and money laundering totalling about US$77 million. According to Ihonvbere and Shaw (1998, p. 151), corruption "permeated all aspects of Nigerian society and public affairs", to an extent that Former President Olusegun Obasanjo remarked "there was corruption! Corruption! And Corruption! Everywhere and all the time!" (cited in Agbiboa 2012, p. 330). Despite various efforts, corruption is still persistent in Nigeria (Transparency International 2017), 3 especially in the public and civil services due to a myriad of reasons including institutional weaknesses and culture (Adegbite and Nakajima 2011).
As the level of corruption in Nigeria is very high (Transparency International 2017), it is common for organizations to seek competitive advantage against their rivals through exchanges in the political market. Politicians in Nigeria wield significant power, which gives them an overwhelming control over resource allocation. Their influence permeates the public and private sectors, helping them to provide rent and give operational and policy favours to their associates and cronies. Consequently, when firms are faced with uncertainty or threats, they tend to fall on their political connections to navigate through institutional quagmires. In this sense, business success in Nigeria, to a large extent, depends on how firms utilize their political capabilities and connections. However, little is known about CPA in this context.
Further, strong materialism in Nigeria has led to greed and corruption, not only among the citizens but also among corporations (Agbiboa 2012;Nwabuzor 2005). In the process of wealth accumulation, there is little or no regard for legality, morality and ethics. Making money trumps how money is made, because "the end always justifies the means." In summary, Nigeria's complex institutional environment (Amaeshi et al. 2016) suggests that firms may deploy unique and previously undocumented context-fitting political strategies to facilitate their operations and enhance their commercial prospects. What could these strategies be?
We focus on banks due to well-informed reasons. Research shows that firms in highly regulated industries engage more with politicians (Grier et al. 1994;Mitchell et al. 1997;Hansen and Mitchell 2000). Banking is one of such industries (Hadani and Schuler 2013). Indeed, the banking industry is one of the most highly regulated in contemporary times. Following recent corporate governance failures and the 2007/2008 financial crisis, countries in Africa and beyond have tightened financial rules and increased bank monitoring.
For instance, the Central Bank of Nigeria (CBN) introduced the Public Sector Cash Reserve Requirement (PS-CRR hereafter) in July 2013 to control bank behaviour. By this policy, all banks were subjected to a 50% reserve requirement on all government agency deposits, in addition to a traditional 12% reserve requirement on all other deposits. PS-CRR affected banks' bottom-lines, as most of them relied on public sector deposits to create loans and maintain liquidity. In January 2014, the PS-CRR became more stringent as it was increased to 75% (Central Bank of Nigeria 2018). This further reduced the amount of public deposits available to banks for lending purposes, leading to a drop in interest revenues.
Hence, due to high regulation and strict monitoring, there is a general expectation for banks and other financial institutions to be very active on the political front. Yet, there is a general paucity of knowledge and understanding about how these firms do their political activity. This is because political strategy researchers have largely overlooked the banking and financial services industry. Within the nonmarket strategy 4 realm, studies have explored CSR in Africa's banking sector (e.g. Amaeshi et al. 2006Amaeshi et al. , 2016, but there is little to nothing on CPA. The industry's strategic importance to economic development does not only make it a top priority for regulators (Zingales 2009;Meslier-Crouzille et al. 2012), but it should also be of keen interest to nonmarket strategy researchers.

Design and Data Sources
Studying corporate behaviour is complex. Hence, we used a qualitative research design to allow us to focus on microlevel firm analyses and deepen understanding of CPA strategies in weak institutional and complex political contexts. As Morse and Richards (2002, p. 28) point out, the use of qualitative research is appropriate if 'the purpose is to learn from the participants in a setting or process the way they experience it, the meaning they put on it, and how they interpret what they experience'. We used in-depth semi-structured interviews to collect primary data, which we triangulated using archival and secondary sources to enhance validity and reliability (Eisenhardt 1989). As we explore a dyadic relationship but could only access one-party (i.e. banks) and not the other (i.e. politicians), we used secondary data (i.e. media articles, news blips) about the relationships, connections and dealings between banks and politicians to corroborate the accounts of the interviewed bank executives. We searched Nexis database using search strings such as "(corporate social responsibility OR CSR OR political connections OR politicians) AND (banks OR financial institutions) AND (Nigeria)". We also searched Google to supplement the Nexis data. These secondary data were analysed purely for their content, with a focus on finding evidence or inferences pertaining to politicians' dependency on banks for favours and benefits.
In 2015, there were 21 commercial banks, five merchant banks and one non-interest bank in Nigeria. Of this total, 1 3 24 were local. We sampled all the local banks to ensure a good representation of the banking industry and to gain diverse insights. Email invitations were sent to the banks to introduce the study and to request participation. Due to the sensitive nature of CPA, only two banks responded. We therefore relied on the first interviewees' referrals to executives in other banks, which helped us to reach 13 banks in total. Using referrals, our sampling process was emergent.
We selected interviewees based on their seniority (management level), involvement in managing external dependence and significant years of experience in the organization (at least, 5 years). These criteria enabled us to collect information from executives who were not only involved in their banks' CPA, but were also knowledgeable of it through their longevity at the banks. In the majority of Nigerian banks, senior and general managers are usually members of their banks' management committees, and are therefore involved in making strategic decisions for managing external relations and stakeholders. Longevity is particularly important because in relationship-based systems (e.g. Nigeria, China), trustworthy relations between managers and politicians can take long to develop (Peng and Luo 2000). Therefore, long serving managers may stand a better chance of nurturing connections between their firms and the polity, and may also be better placed to comment on their firms' CPA. Given the volatile nature of political issues addressed in this study, participants and their banks were assured of anonymity and confidentiality to reduce social desirability bias and ensure effective communication (Miller et al. 1997;Yin and Zhang 2012). In keeping with non-disclosure agreements, we do not offer descriptions of the banks in this study, because we believe that even basic information could make them identifiable, especially considering that the entire population of banks in Nigeria is small.
Data collection spanned 14 months (June 2015-August 2016). We conducted in-depth semi-structured interviews with 15 bank executives from different levels and functional areas. To confirm the data, we tried to interview more than one executive in each of the banks and succeeded to do this for two of the 13 cases. The two cases demonstrated consistency, suggesting that the information from single interviews is internally valid. Moreover, previous qualitative studies in the African context successfully used single respondents (e.g. Amaeshi et al. 2016;Luiz and Stewart 2014). Our descriptive statistics show that the informants worked for their respective banks for 8 years on average, suggesting that they were knowledgeable about the banks' political activities. In line with other studies (e.g. Amaeshi et al. 2016), we sent the interview guide (which was piloted and revised) to the respondents to facilitate their preparation. Each interview lasted between 45 min and 1 h, generating over 3000 words of texts. We sent respondent copies of their interview transcripts to check content validity. See Tables 1 and 2 for a summary of data sources and sample characteristics.

Data Analysis
Overall, as there is a paucity of research on political activity in Africa , we used a grounded theory approach to allow theory to emerge from the data rather than fit data to any preconceptions (Strauss and Corbin 1998). Previous CPA studies have also used grounded theory approaches to investigate under-explored issues (e.g. Shaffer and Hillman 2000). As we sought to identify and match patterns within and across the cases, our analysis drew upon previously validated approaches for qualitative theory development (Eisenhardt 1989;Yin 2003). In keeping with grounded theory principles, the analysis was done in three stages, starting from when the first bit of data was collected (Corbin and Strauss 1990). First, we analysed each interview transcript (i.e. within-case) to allow for an in-depth understanding of the study phenomenon (Gioia et al. 2013). This helped us to construct a story about CPA motives, organization and strategies for each of the cases. Subsequently, we did open coding to identify concepts and categories. During this stage, we developed first-order codes from the raw data. Starting from the first case, we developed a coding scheme around emergent patterns. As we analysed subsequent cases, we reflected on prior patterns and looked out for new ones. We merged similar patterns into the same first-order codes until we reached saturation (i.e. no further patterns emerged). After gaining insights from across the cases and inductively developing the first-order codes, we resorted to developing higher themes through axial and selective coding. During axial coding, we started an abductive process whereby we iterated between theory and data (Gioia et al. 2013;Langley 1999). This process helped us to identify linkages among the first-order codes, which led to the development of theoretically oriented second-order codes. In the final selective coding stage, we distilled and integrated the second-order codes into higher aggregated dimensions, which enabled us to understand CPA among Nigerian banks. Figure 1 provides an overview of our data structure.

Findings
This section captures results about "why" and "how" banks do their CPA in the emerging context of Nigeria. Precisely, it explores the drivers and motivations of CPA, and the political strategies used.

CPA Drivers in Weak Institutional Contexts
Our data revealed that the main drivers of CPA are: (1) commercial and business opportunity; (2) regulatory leniency and arbitrary treatment; (3) sense of security and protection; and (4) voice and influence. These drivers, to a large extent, are similar to previously espoused motivations for corporate political embeddedness in emerging countries. It is worth noting that they are also hinged on the strong dependency of private enterprises on governments (Hillman 2005), especially in developing and emerging countries where government overreach is pervasive (Khwaja and Mian 2005;Liedong et al. 2017).
Commercial and business opportunities emerged as the overriding rationale and driver for developing political connections. The banks developed connections in order to win government's underwriting business or contracts for collecting bills and fees on behalf of government agencies and public institutions. Respondents believe that politicians can influence which financial institutions win bids to collect revenues. Indeed, in Nigeria's corrupt context (Agbiboa 2012), political power permeates the entire public sector.
We cannot do without the politicians. They are very powerful people. We have to do their bidding to remain in business. If we do not meet with them in a very personal way, they will not give us business or they can use the government machinery to frustrate us. These politicians ensure that their relations and friends are appointed to head various arms of the public sector, so they have control [Respondent A1].
We found that winning public sector deposits is the most salient commercial driver of CPA. These deposits come from government agencies and they are huge, often running into several billions of Naira. Banks rely on them for loan creation and profitability. It emerged that prior to 2013, much of the loan portfolio of some banks in the country was funded by public sector deposits. Banks still compete to attract public sector accounts and in this competition, political connections are crucial. This commercial driver of CPA is unique and has not been captured in the literature until now, perhaps because previous studies have not explored CPA in the banking industry where deposits are crucial. As respondent H1 noted: Deposit mobilisation is crucial for our profitability, growth and sustainability, and because the public sector deposits is very cheap source of deposits, it is very attractive for us. 95% of these deposits are in non-interest earning current accounts. We use deposits from the public sector to fund around 70-80% of bank's risk assets. As previously argued, the banking sector is exposed to stringent regulation and tight deadlines. The room to manoeuvre is often non-existent, and if it exists, it is usually small and not all banks have equal laxity. Our data shows that the quest for regulatory leniency encourages CPA in the banking industry. Requests for deadline extensions to meet governance, operational or capital requirements are treated favourably if banks have political clout. In other words, political connections help with arbitrary application of the rules (García-Canal and Guillén 2008). In cases of breach, penalties are also less severe for politically connected banks. Even though the Central Bank of Nigeria has the mandate and responsibility for bank supervision, it is not immune from political influence. After all, politicians appoint the Governor of the Bank. Hence as a parastatal, it is susceptible to political capture. According to respondent B2: The president appoints the governor of Central Bank of Nigeria (CBN), who in most cases is loyal to the President and related politicians. Therefore, through politicians, we enjoy some favours Related to leniency, the quest for security and protection equally drive CPA. Operating in a context riddled with weak institutions, poor corporate governance and an ineffective regulatory regime (Amaeshi et al. 2016), Nigerian banks cultivate political connections for protection. Such protection includes shields against drastic penalties and support for credit recovery legal actions. Litigations to recover debt from defaulting parties are successful when there is political backing, a phenomenon that attests to the overreach of political influence in weak institutional environments.
Further, we found that having a voice in policy discourses stimulates the development of political connections. Some bank officials acknowledge the major impact that even small regulations have on their operations and profitability, and therefore want to have a say in how the rules are made. With political connections, they stand better chances of being involved in policy consultations, which not only helps them to advocate for their preferred positions but also helps them to make advance preparations. This, of course, suggests that policy consultation in weak institutional contexts is reserved for privileged "club" members whose voice may override outgroups and the less privileged.
Far-reaching regulatory changes are an enormous challenge to us and we most quickly evolve our strategies. Being connected to politicians gets us into policy consultations that will help us say what we want.

[Respondent C1]
In summary, the foregoing shows that Nigerian banks are motivated to do CPA for a myriad of reasons. In developed markets, profitability and competitive advantage mainly drive CPA Baron 1995). However, in weak institutional environments, firms also do CPA to navigate uncertainty, voids, and other constraints. Table 3 provides summarizes the drivers of CPA in Nigeria's banking industry.

Organization of CPA
We found that the formulation and execution of CPA varies depending on the nature of the regulations or issues confronting banks (Table 4). There are two categories of regulations banks in Nigeria deal with-procedural and transformational regulations. Each category is associated with a different type, target and level of political strategizing. On the one hand, procedural regulations are concerned with governance of the banks. They are common and include dayto-day operational issues and periodic filing of documentation with regulators. Banks are passive about procedural regulations. Their stakes in these regulations are generally lower. Any strategies (mostly persuasive in nature) to influence procedural regulations are done collectively with other banks and are targeted at bureaucrats such as senior officers of the Central Bank (CBN). Banks only resort to political influence when Central Bank officials are not yielding.
For the day-to-day central bank supervision, we really do not bother ourselves on that. If we have to, we do it collectively as an industry through the bakers committee.

[Respondent K1]
On the other hand, transformational regulations are gamechanging. They affect financial performance, raise the stakes, and elicit active political engagement. An example of such regulations was the 2013 PS-CRR, which restricted how much public deposits were available to banks for lending and liquidity. Banks respond to transformational regulations individually, with the hope of receiving exclusive exceptions that will give them superior competitive advantage in the industry. They target politicians at both the federal and State levels, using a top-down filtration approach. Influential and top politicians are prioritized first. All political activity aims to build relationships with political actors.
While the day-to-day CBN guidelines do not bother us, there are reforms and regulations that places enormous challenge to banks, for example, PS-CRR and the treasury single account (TSA) whereby banks were mandated to remit all government ministries' funds to the central bank. We were challenged on every side. We had to talk to top politicians to rescind the decision or give us exceptional treatment [Respondent I1]

Political Strategies
We found four political strategies from the data. We label them as Affective, Financial, Pseudo-Attribution, and Kinship strategies. Some are dark, in the sense that they may not ethical. Others reveal the unique political influence opportunities available to only financial institutions. Below, we explore each of these strategies.
The Affective strategy involves identifying and recruiting bank officials who are willing or capable of being emotionally involved with politicians. It is a subtle approach of gaining political influence by arousing the sexual instincts of politicians. The affective strategy is unconventional and perhaps difficult to believe, but recent events lend credence to its existence and usage. In March 2018, it was revealed that Cambridge Analytica, a British political consulting firm, used dirty tricks to influence election outcomes in many different countries. The firm's executives were recorded boasting about using girls and sex to entrap political opponents (Agerholm 2018;Graham-Harrison et al. 2018). Whereas Cambridge Analytica's tactic was to create sexual scandals to destroy or halt opponents' political careers, Nigerian banks use the tactic to appeal to politicians' soft side so that they can receive preferential treatment through emotionally involved female workers who are mostly employed as relationship or marketing officers. Most politicians in Africa, and perhaps around the world, are male. Due to the prevalence of polygamy in most African cultures (Fenske 2015;Schoellman and Tertilt 2006), male promiscuity is common and this makes the strategy viable. The affection strategy is highly unethical, mainly as it spurs bias in recruitment practices, exploits female employees and exposes them to sexual harassment.
You know that with the high rate of unemployment, girls are more willing to go the extra mile to be employed, retain their work and make extra money. In this industry, most of us (banks) seize the opportunity to make these girls relationship and account managers to these politicians in order to entice them for favours.

[Respondent G1]
The Financial strategy is different from Hillman and Hitt's (1999) campaign financing or political donations. With this strategy, banks offer preferential financial services and treatment to politicians. They relax due diligence in financial transactions involving politicians, which may have implications for money laundering. Banks also help politicians to unduly earn interest on public sector deposits. As Nigeria is a highly corrupt country (Agbiboa 2012), it is common for her politicians to loot public coffers for personal gain. Suggestively, banks help politicians carry out such corrupt practices in return for political access and influence. Further, banks grant loans to the cronies and associates of the politicians in order to gain access to the polity. CPA studies in other emerging countries have documented high financial leverage among politically connected firms (Claessens et al. 2008;Khwaja and Mian 2005;Saeed et al. 2015), which lends credence to this strategy of banks using connected firms to reach politicians (Houston et al. 2014). The extract below illustrates this claim: Do you know that some politicians earn interest on public funds deposited? I can authoritatively tell you that we pay between 5-10% on these deposits. These are paid in cash to their cronies, companies and family accounts to avoid audit trail. I know other banks that grant uncollateralized loans to politicians, their families or friends. Do you see the high volume of non-performing loans in some bank books? [Respondent M1] Financial strategy reveals reciprocity or a two-dependency relationship between politicians and banks, in the sense that the two actors are mutually dependent on each other: the banks help the politicians to gain financial resources; the politicians help the banks to win government business and overcome regulatory challenges. This "I scratch your back, you scratch my back" relationship gives financial institutions greater prominence in the eyes of politicians. Consequently, the financial strategy is the most effective for the banks.
As politicians cannot do without banks, banks also cannot do without politicians. Indeed, I can say that we complement each other. See, I say that this is a win-win situation. They win our secrecy and help while we win their support to achieve the bottom line, and promote our corporate interest. No one will want to stick his or her neck out for a politician if there are no concrete benefits for taking such risk. [Respondent E1] Secondary data corroborate these primary accounts. Financial Action Task Force (FATF), in a report on money laundering, noted how domestic banking systems are used by politicians to embezzle public funds in Nigeria. In other words, banks' support and cooperation is required for politicians to launder money, and this support is based on mutual benefit.
In the case involving former governor of Plateau state in Nigeria, Joshua Dariye, for example, a grant for environmental contracts was made from the federal government to the State, and the money was deposited into a bank account established by the State. Dariye used his influence to cause the bank to issue a bank draft creditable to an account at a different Nigerian bank that Dariye had established under an alias about ten months previously (FATF, 2011:17) In the Pseudo-Attribution strategy, banks use CSR for political ends. However, the way they execute CSR as a political strategy differs from the way they execute it for altruism purposes. When used as a political strategy, CSR is targeted at the constituencies and communities of focal politicians. The politicians are also involved in diverse ways, including commissioning and sod cutting, to make it possible for them to claim ownership or part-ownership of social programs. That way, they are seen to be contributing to their communities, and this bolsters their re-election chances. In Nigeria, elections are fiercely contested; hence, any help towards re-election is considerably reciprocated. Occasionally, banks target their CSR at individuals or groups that are very close to politicians. In this way, they are able to appeal to the politicians. For instance, Stanbic Bank's 2011 sustainability program included "Office of the First Lady's Retreat", a project that: Sponsored the international retreat on etiquette, protocol and administration for wives of all state governors in the Federal Republic of Nigeria. Requested by the CEO as a means of capacity building within individual states of the Federal Republic of Nigeria 5 Earlier suggestions that CSR could be used as a political strategy were mulled over a decade ago (Hillman et al. 2004). Since then, various studies have shown that, indeed, CSR is a political strategy in emerging countries, particularly in China where it helps firms to gain access to the polity (Jia and Zhang 2016;Li et al. 2015;Su and He 2010;Wang and Qian 2011). Within the Nigerian context, the data further shows that when banks use CSR as a political strategy, they work harder to ensure that the social programs are successful. It thus appears that when there is a political imperative for social interventions, CSR is highly needsbased, effectively tailored, more philanthropic, and large scale. Essentially, politically motivated CSR is more intense than altruistic CSR due to high stakes emanating from the involvement of political stakeholders in the former.
Not all our CSR is politically motivated, but we also use CSR as a strategy to get the attention of politicians. Their communities are targeted for development projects. We ensure that the activities are successful, with praise and encomium given to the bank and politician. These things enable us gain access to them. [Respondent A1] Closely linked to the Attribution strategy, the Kinship strategy entails recruiting family members of politicians in order to establish a link with government. While this strategy may seem similar to previous tactics which conceptualize political connections to include friends and family of politicians (e.g. Fisman 2001;Sun et al. 2011), it differs in the sense that it also involves recruiting people from target politicians' constituencies. In emerging countries such as Nigeria, unemployment is high (Ogbeide et al. 2016) and its reduction is a constant promise politicians make on their campaign trails. Additionally, most Nigerian youth have very poor perceptions of blue-collar jobs (Dike 2010); hence, they prefer white-collar positions such as bank work, which is highly regarded and prestigious. Banks exploit these situations, recruit from politicians' communities and actively allow the politicians to take the credit for helping the youth to gain employment. Again, as re-election chances are boosted, so is the chance of reciprocity.
Aside various community development program we embark upon, we also intentionally give employment to people from target communities, most especially relations and family members of politicians.

[Respondent J1]
In Table 5, we provide a summary of the political strategies and their enabling capabilities, supporting resources, and potential risks/ethical issues. Whereas the affective and financial strategies are out rightly unethical (e.g. biassed recruitment in favour of physical looks rather than merit and financial crimes), the case of pseudo-attribution and kinship strategies are not clear-cut.
Collectively, all four strategies fit the politico-sociocultural conditions and weak business-government interface in Nigeria. We acknowledge the similarity of some of these strategies to other strategies used in emerging market contexts. However, we also observe some stark differences. First, to the best of our knowledge, this is the first scholarly study to uncover how emotions and affections are exploited in CPA. Anecdotes suggest that sex is a tactic for political influence. For instance and as we mentioned earlier, senior executives at Cambridge Analytica bragged about how they can gain political leverage by using sex workers to entrap politicians (Agerholm 2018). Our study provides scientific and empirical evidence to corroborate such anecdotes. More importantly, while sexual anecdotes usually entail the blackmail of politicians, the affection strategy we report in this study does not use extortion or intimidation. Instead, it aims to convince politicians, in a non-threatening way, to protect banks' interests.
The exploitation of women to pursue business goals is not exclusive to Nigerian banks. The practice seems to be present in other corporations in the country. For instance, Dutch brewing giant Heineken was found to use "promotion girls" to boost beer sales in Nigeria (van Beemen 2019). Executives justified and framed this practice, which exposed the girls to sexual abuse, as a local custom. While such cases in Nigeria are consistent with globally documented use of sexuality in advertising and sales promotion (e.g. Gould 1994), our affection strategy is unique in the sense that it is about exploiting girls to gain political influence, not to boost sales. Whether this specific use of girls for political access is a widespread practice in Nigeria is not yet clear. Nevertheless, we contend that the affective strategy is facilitated by weak institutional protection of women, which allows the abuse to persist.
Second, numerous studies have examined how firms influence government policy through financial donations and contributions to political parties and candidates (e.g. Claessens et al. 2008;Dean et al. 1998;Hadani and Schuler 2013). However, the financial strategy in this study is different, in the sense that it is not just about firms donating money to politicians. Instead, it captures how banks wield political influence by providing preferential financial services (beyond loans) to politicians.
Third, extant studies have shown that socially responsible firms benefit positive reputations, which give them greater access to the polity (Hond et al. 2014;Liedong et al. 2015). Hence, it is often argued that CSR can be used as a political strategy (Hillman et al. 2004;Li et al. 2015). Indeed, the pseudo-attribution strategy corroborates the use of CSR in CPA, but it goes beyond previously claimed reputational effects to show that political influence in weak institutional contexts may not come from simply doing CSR or merely being socially responsible. Rather, this influence comes from well-orchestrated social programs involving politicians and targeted at politicians' communities. Essentially, until politicians are able to directly use CSR for political (election) purposes, CSR may not be an effective political strategy.
Finally, previous CPA studies measured political connections by examining the relationship between firms on one hand and politicians or their immediate family members on the other hand (e.g. Chaney et al. 2011;Faccio et al. 2006). Hence, a firm could be politically connected when it employs the relation of a politician. The kinship strategy reported herein looks beyond employing only politicians' close relations. This strategy captures the practice of employing members of politicians' wider communities and constituencies. We summarize these differences in Table 6.

Choice of Political Strategy
Our data shows that the choice of strategy depends on two main factors: (1) urgency, which captures the need for speed; and (2) resources, which captures resource availability (e.g. funds or budget). For banks, the affective and financial strategies present the lowest cost. The affective strategy does not use many employees while the financial strategy does not involve additional costs at all. In contrast, pseudo-attribution and kinship strategies require huge financial outlays for CSR programs and mass recruitments. Accordingly, we found that smaller banks (per total assets) tend to mainly use the affective and financial strategies while the larger banks, due to their deeper pockets and stronger reputations, are able to deploy the pseudo-attribution and kinship strategies.
Regarding urgency, banks get quicker results when they deploy financial and kinship strategies. This is because the affective strategy takes time to execute as emotions and trust are nurtured over time while the pseudo-attribution strategy requires careful planning. In terms of scope, we found that banks use more than one strategy at any given time, with the larger banks using three of the reported strategies, on average. Besides urgency and resource considerations, banks retain or drop political strategies depending on their effectiveness. They engage in dynamic portfolio management whereby positive (negative) outcomes reinforce (weaken) reinforce their choice of strategy. In the same way, the outcomes of CPA tend to reinforce or weaken the rationales for doing CPA. For instance, commercial success reinforces the commercial rationale. Figure 2 shows a matrix of the determinants of political strategy choice.

Discussion
Our study addresses the lack of research on corporate political strategization within the African context in general and on financial institutions in particular. Emerging countries are risky (Puck et al. 2013;Liedong et al. 2017), which makes it necessary for firms to develop political strategies for influencing their volatile environments (Nell et al. 2015). However, besides informal political ties and connections (Peng and Luo 2000;Rajwani and Liedong 2015), there is a paucity of research that explores the full breadth of political strategies used in emerging countries, especially in Africa. Drawing on resource-dependency logics, and using interviews with bank executives (for firms' perspective) and secondary data (to capture politicians' dealings with banks), we show that the organization of CPA is dependent on the type of regulations (procedural versus transformational) that firms seek to influence. This finding is a new addition to the literature. Previous studies have only captured the role of national culture, financial resources and product diversification in political strategy formulation (e.g. Barron 2011;Hillman and Hitt 1999). They have barely explored how types of regulations affect how CPA is organized. Firms in Nigeria deploy some "dark" strategies to wield political influence. These strategies are unconventional, unethical and different from what is already captured in extant literature, but they are effective for influencing politicians in the local context. More importantly, these political strategies reflect the nature of the dependency relationship between firms and politicians. This relationship is based on the principle of reciprocal interdependence, in the sense that firms and politicians engage in a social exchange with attendant obligations for each party (Li et al. 2015;Jia and Zhang 2016). Politicians depend on banks to duly or unduly receive the financial benefits of public office. They also depend on firms' CSR and employment practices to enhance their political images and boost their re-election. In return, the politicians offer the banks audience, protection, and commercial opportunities. The challenge of reciprocal interdependence is that in weak institutional contexts, it could be fraught with unethical practices, as our data shows.
Previous scholarship has noted that politicians and firms transact in political markets (Bonardi et al. 2005;Kingsley et al. 2012). In these markets, firms have to offer what politicians want (e.g. money, information) in return for political favours (Hillman and Hitt 1999). In highly corrupt Nigeria (Agbiboa 2012;Smith 2007), politicians need money to finance clientelism, vote buying schemes and re-election. They also need money to support their opulent lifestyles and provide financial security for their families and associates. For banks, this creates an opportunity not to only contribute actual money, but to render financial facilitation services. These services may include preferential access to  Fig. 2 Political strategy matrix loans, overridden due diligence in financial transactions, and collusion to unduly benefit from public financial resources. Generally, financial strategies are ethically suspect (Hillman and Hitt 1999) and this is true in weak institutional contexts such as Nigeria. Based on insights from our data, we argue that not all political strategies are available to all firms. Rather, political strategies may be industry-specific. In other words, different industries may have different ways of appealing to politicians. We found that the banking industry has a unique opportunity to manage the finances of politicians in return for policy favours. This opportunity is not available to firms in other industries. Generally, beyond the understanding that heavily regulated industries are politically active (Grier et al. 1994;Hadani and Schuler 2013), not much is known about what unique political tactics are available to different industries. Our study thus generates valuable insight for CPA, as it suggests that reciprocal interdependence is existent or stronger in some industries than in others depending on the extent to which industries are directly resourceful to politicians. In this sense, political strategies that exploit this dependency may be more effective.
In emerging and developing countries, CSR makes significant contributions to socio-economic development , mainly as socially responsible firms serve political roles (Scherer et al. 2013) and have analogous power of the State (Matten and Crane 2005). Through their activities, they provide a complementary governance structure (Amaeshi et al. 2006), which helps them to gain political access Su and He 2010;Wang and Qian 2011). Subsequently, the extant literature portrays political legitimacy as the main mechanism through which CSR yields political benefits (Lin et al. 2015). Our data provides finer-grained insights about how this legitimacy is gained through reciprocal dependency.
In Nigeria, politics is a "do or die" zero-sum game. Competition for political office is fierce, elections are high-stake events, and incumbent politicians do everything in their power to retain their offices. Nigeria has over 90 registered political parties. However, three parties-Peoples Democratic Party Nigeria (PDP), All Progressives Congress (APC) and Labour Party (LP)-dominate the political arena. Politicians within these parties face fierce competition at two stages. First, they must compete in internal primary elections to represent their parties. Second, they must compete in multi-party State and Federal elections. These elections, whether intra or multi-party, are perilous and expensive. Party nomination forms could cost as much as N45 million (about US$ 125,000), but there is never a shortage of contestants because of the unfettered benefits associated with political power. The attractiveness of political office perhaps explains why 73 candidates competed in the 2019 presidential election. This number of presidential contestants, arguably the highest in the world, demonstrates the competitiveness of Nigerian politics. By orchestrating targeted CSR in ways that allow politicians to take direct credit for bringing social programs and development to their communities, firms improve politicians' re-election chances, which helps them gain legitimacy and political access. Previous studies about CSR-CPA have not captured this deeper process of legitimization because they were mainly conducted in China (e.g. Jia and Zhang 2016), a one-party State where political rivalry and competition is comparatively lower than multiparty democratic Nigeria and where politically targeted CSR may not be as important.
While the role of CSR in politics, largely described as political CSR, has received attention (e.g. Scherer et al. 2013) and may be somewhat related to what we found in this study, we note a significant difference. Political CSR concerns social impacts related to the functioning of a State (Frynas and Stephens 2015), not direct business activity as we see in our data. In this sense, CSR as a political strategy is different from the concept of political CSR. The latter is arguably altruistic and driven by utilitarianism while the former is underpinned by reciprocal interdependence and motivated by idiosyncratic gain. Therefore, we use resourcedependency theory to advance that reciprocity and duality in the dependency relationship between firms and politicians is a necessary foundation on which CSR can be deployed as a political strategy.
Bringing all the findings into perspective, we develop a political strategization framework (Fig. 3). This framework shows that the quest for commercial and business opportunities, the desire for regulatory leniency and protection, and the need to have a voice in policy arenas motivate banks to orchestrate CPA. Banks organize their CPA depending on whether they are addressing procedural or transformational regulations. When they deal with procedural regulations, they are transactional, passive and collective in their political engagements with bureaucrats. In contrast, when they buffer transformational regulations, they actively and individually use a relational approach to target politicians (see detail in Table 4). More importantly, banks use a unique array of political strategies (see detail in Table 5) that exploit their reciprocal dependency relationships with politicians. These strategies are supported by politicians' dependency on funds and goodwill for personal enrichment and reelection. They are also supported by the lack of formal business-government interfaces for stakeholder involvement in political processes, and further by the existence of politico-socio-cultural conditions (particularly weak institutions) that allow unethical practices to prevail. For instance, the financial strategy exploits not only poor bank supervision and monitoring, but it also capitalizes on absent CPA regulations, rampant corruption, and a winner-takes-it all politics. Similarly, the kinship and pseudo-attribution strategies exploit political patronage in relationship-based systems.
Banks choose their political strategies depending on their resource endowment and the urgency of the issues they face. Some strategies are cheap (e.g. affective), others are best suited for quick results (e.g. financial). The interplay of urgency and resource endowment eventually determines which strategies are used. It is worth noting that CPA outcomes also affect strategy choice, mainly as positive (negative) performances reinforce (weaken) what is believed to be the optimal strategy. Outcomes further influence the rationale or motivation for political activity by strengthening or undermining CPA drivers. Overall, the framework illustrates the process of orchestrating CPA (i.e. from drivers to outcomes) and shows the moderating and supporting factors of this process. Importantly, it highlights how weak institutional conditions and politicians' need for funds and goodwill enable firms to exploit reciprocal dependency and deploy ethically suspect political strategies.

Practical Relevance and Policy Implications
Besides theoretical contributions, this study offers lessons and insights to firms, governments, regulators and development partners. First, some of the most egregious cases of corruption in Africa usually involve businesses conniving with public officials and politicians. For instance, our findings offer insights into how some political strategies (e.g. financial strategy) may exploit, aid and abet corruption in Nigeria, suggesting the complicity of the private sector in this menace. However, it is worth noting that most businesses in Africa are often the victims of corruption, mainly as they face unsurmountable pressures to pay bribes (Mishra and Maiko 2017). To find a lasting solution to corruption in the region, firms need to help strengthen State institutions by playing a more active role to instigate institutional change (Liedong 2017). Some firms do not believe they can significantly influence Africa's institutions (Luiz and Stewart 2014); hence, they do not seem to be doing much in that regard (Liedong 2017). Nevertheless, this belief might hold true only if firms act alone. By forming coalitions and industry associations, firms can present a collective front that will allow them to buffer institutions in unison and without fear of selective witch-hunt and discrimination. Previous research has acknowledged the effectiveness of industry groups and unions in institutional change (e.g. Barley 2010;. Therefore, banks in Nigeria could unite to use their CPA not to exploit institutional weaknesses and pursue idiosyncratic benefits, but to address fundamental institutional problems that will help alleviate the biggest bane of Africa's under-development-corruption.
Second, this study has revealed some of the dark and unethical strategies used for political influence. These strategies have negative ethical implications. For instance, the affection strategy could lead to the abuse and exploitation of female workers while the financial strategy could aid money laundering. We hope that regulators, especially Central Banks, can develop industry-wide codes for governing CPA. There is also a need to strengthen the monitoring and supervisory functions of regulatory bodies so they can detect and address some of these ethically suspect political strategies.
Finally, this study calls on international agencies to support or pressure African governments (through aid and development assistance) to enact laws that will regulate and make CPA transparent. Currently across Africa and other developing countries elsewhere, structures and mechanisms for business-government relations are either non-existent or inefficient (Liedong 2017). This leaves the use of informal ties as the main way for businesses to have a voice in   , but these ties are opaque and often fraught in corruption (Lawton et al. 2013a). We hope that the insights herein will provide a catalyst for development agencies to support African countries to institute mandatory and publicized "comment windows" that will enable external stakeholders, especially businesses, to make inputs into public policies and regulations. Additionally, African governments need to impose limits on campaign financing to curb vote buying and corruption. They also need to adopt public lobbying registers that will show who is lobbying who and on what issues. This transparency could curb the excesses of CPA. Perhaps, if legitimate and formal CPA structures existed, banks in Nigeria would not resort to ethically suspect and unconventional strategies.

Conclusion
In conclusion, we note that there is a significant difference in how CPA is formulated and executed in Western countries, emerging Asian countries and emerging African countries. While African and Asian firms may share same motives due to having similar institutional conditions, the orchestration of CPA is starkly nuanced for African firms. In this nuance lies some ethical issues bordering on unfair recruitment practices and corruption. Hence, the main thesis of this paper is that within the weak institutional environment of Africa, firms use context-fitting but ethically suspect strategies that are built on the principle of reciprocal interdependence. The scanty research on CPA in Africa (e.g. Liedong et al. 2017;Liedong and Frynas 2017;Wocke and Moodley 2015;Mbalyohere et al. 2017) does not only understate the prevalence of firm-politician embeddedness in the region, but it also mutes the ethical issues of CPA. Recently, research revealed the negative effects of political connections on corporate governance in Ghana (Liedong and Rajwani 2018). More of such research is needed to identify the negative externalities of business-government relationships in Africa.

Limitations and Future Research
Despite our study making important theoretical and practical contributions, it has some limitations. As seen in many qualitative studies, a small sample size limits generalizability. More importantly, this study focuses on a single industry, which limits the extent to which the findings can be extrapolated. Further, we could not gain direct access to politicians, and hence used secondary data to fill this void. Nevertheless, our study has generated rich insights about CPA. As we found the financial strategy to be unique to the banking industry, it is plausible to think that subindustries within financial services (e.g. insurance, mutual funds, brokerage, etc.) or even other industries may also have their own arsenal of unique strategies for political access and influence. Therefore, we hope research can build on our study to explore other industry-specific political strategies.
We also note that earlier concerns about the negative outcomes of CPA have evolved into the development of ethical frameworks for evaluating the rationales, mechanisms and outcomes of lobbying and political activity (e.g. Keffer and Hill 1997;Oberman 2004;Weber 1997). These frameworks are normative and prescriptive, and have rarely been put to empirical tests. Moreover, the evaluation of ethicality is subjective, just as morality is. Hence, the way CPA is judged to be (un)ethical in one context may differ from the way it is judged in other contexts. In this respect, it is important for research to investigate how managers assess the ethicality of their political activity. Contextualizing this point, we found some of the political strategies in the Nigerian banking industry to be ethically suspect, but that is our distant judgement. Do the managers of these banks agree with our ethical stance about their CPA? If no, how do they determine ethicality? Addressing these questions will enrich both the ethics and CPA literatures.