With other people’s money: Campaign financing as an agency problem

Abstract

Agency theory explanations for corporate political activity assume that managers distort resource allocation to invest in political connections to pursue personal benefits. While distorted resource allocations yield poor earning quality, we expect that companies with efficient governance may curb this opportunistic behavior. We used matching procedures to identify the effects of financing political campaigns on the earning quality of the firm. We assembled an original panel of listed firms in Brazil from 1998 to 2013. We found that firms that donated to electoral campaigns had a lower earning quality than nondonor firms. Firms with superior corporate governance instruments were able to reduce the harmful effects on earning quality. These results support the tenets of agency theory in explaining why firms engage in politics.

Keywords

accruals corporate governance corporate political activity corporate political behavior donation to election campaign earning quality

Introduction

Agency theory framing research into corporate political activity (CPA) argues that managers invest in political activity to pursue private benefits at the expense of shareholders (Hadani et al., 2015; Mellahi et al., 2016). These private benefits include exchanging favors (Claessens et al., 2008), increasing compensation (without performance pressure), improving the manager’s reputation (Hadani et al., 2015), increasing personal social capital (Faccio, 2006), and possibly pursuing an interest in a political career (Coates, 2012). The resulting distorted allocation of resources (Chaney et al., 2011) causes agency conflicts because shareholders are not guaranteed a return on their investments and because insiders utilize these investments for their own benefit (Aggarwal et al., 2012; Jensen and Meckling, 1976).

Politically connected firms tend to exhibit opaque behavior in that they are not expected to report earnings and expenses from the connection, which creates informational asymmetry between shareholders and managers that exacerbates the agency problems (Dahan et al., 2013). We used earning quality to measure reported versus real earnings and took the total accruals as a proxy for earning quality (Chan et al., 2006; Solan, 1996). Accruals are defined as the difference between a company’s accounting profit and its cash flow and are considered important indicators of the quality of a firm’s profit (Chan et al., 2006). Empirical research has shown that political connections affect earning quality (Chaney et al., 2011) because connected firms derive gains from their connections and because managers may hide these gains with the purpose of intentionally misleading investors, leading to less transparent behavior (Chaney et al., 2011).

Agency theory assumes that managers distort resource allocation for personal reasons, or that they are not following the orientation of shareholders. Good governance practices are intended to reduce agency conflicts (La Porta et al., 1999; La Porta et al., 2000). Therefore, we argue that if political donations are driven by self-interest managers, then good governance practices will reduce the agency-related conflicts caused by the political behavior of firms.

We tested this agency conflict mechanism using a sample of firms that nurtured political connections. There are some mechanisms underlying the political connections of firms, such as campaign donations (Claessens et al., 2008), a former politician on the board of directors (Faccio, 2002), lobbying (Brasher and Lowery, 2006), and personal ties (Hillman et al., 1999). However, here, we used the political contributions to electoral campaigns (Claessens et al., 2008). We investigated whether good corporate governance could reduce the negative effects of political connections on earning quality. We tested our hypotheses using a sample of firms listed on the Brazilian stock exchange from 1998 to 2013. Brazil is an excellent natural laboratory for this analysis for several reasons. First, Brazil is well suited for studies designed to reveal the mechanisms by which political ties impact company performance. Second, data regarding firms’ donations to campaigns are easily accessible, and electoral regulations are permissive regarding firm donations; this makes it easier for firms to manage political influence. Third, the existence of relationships between firms and the government is a phenomenon rooted in the national culture (Aldrighi and Postali, 2010), and the context demands many policy benefits (Boas et al., 2014). Fourth, candidates in Brazil rely on corporate money, and donors may be able to extract greater policy concessions (Boas et al., 2014). Fifth, weak institutions are a problem in the country. Finally, Brazilian firms, as in many emerging markets, are highly dependent on government resources such as financing, licenses, and preferential treatment. We used the propensity score matching method to improve causal identification (Rosenbaum and Rubin, 1983).

Our findings make the following contributions to the literature on agency theory corporate political action and general manager practices. First, we provide supporting evidence of agency-driven political activity (Liedong and Rajwani, 2018). Although the potential use by self-interested managers of corporate money to nurture political connections for personal reasons is an obvious feature of corporate life, empirical research into CPA has overlooked this theoretical explanation (Mellahi et al., 2016) Studies from the institutional- and resource-based perspectives have limited ability to explain the distortion that is introduced by political connections (Liedong and Rajwani, 2018). Additionally, agency theory may indicate that political connections can lead to agency problems since the connected manager can use the gains with political connections for his or her benefit (Hadani et al., 2018). Second, the research context allows for a more robust claim about the causal identification. We apply matching techniques to provide better evidence about the detrimental effects of political donations on earning quality. Third, the moderating effect of corporate governance on the relationship between political connections and earning quality adds new knowledge to the research stream about the accounting practices of politically engaged companies (Chaney et al., 2011). Moreover, we provide implications for general managers in the sense that the power of corporate governance mechanisms leads to more transparent behavior, even in instances of connected managers who exhibit less clear behavior in earning and expense reports from the connection. For the general managers capturing resources, we also propose that it is necessary to be aware of the accounting practices of politically engaged companies.

Theoretical reference and hypotheses

Harmful effect on firms’ allocation of resources

Previous studies of corporate political strategies have outlined methods for firms to gain access to policy makers and influence the political process to obtain economic benefits (Hillman et al., 2004). Scholars have assumed that firms engage in corporate political strategy to improve the performance (Bonardi et al., 2006; Hillman et al., 1999; Lux et al., 2011). Firms assume that political strategy is a type of investment and allocate resources to political activities to obtain better returns than investments in alternative activities, such as R&D (Lux et al., 2011). Furthermore, the use of political strategies becomes attractive when the benefits outweigh the costs (Baron, 1995).

The political strategies identified in previous studies include the following: (i) donations to election campaigns (Claessens et al., 2008); (ii) politicians who have greater knowledge of legislation, as well as access to key politicians serving on boards of directors or in executive roles (Faccio, 2006); (iii) businessmen entering politics and/or befriending politicians who may influence legislation, obtain privileged information, and influence the government in their firms’ favor (Luechinger and Moser, 2014); and (iv) direct or indirect government participation in firms as shareholders or owners (Faccio, 2006; Hillman and Hitt, 1999; Lux et al., 2011; Ridge et al., 2016). Firms use these political connections to obtain access to politicians, information, legislative processes, preferential treatment, and protection from nonfacilitating governments (Bonardi et al., 2006; Choi et al., 2014; Lazzarini et al., 2015; Pearce, 2001; Schuler et al., 2002).

In this study, we analyze political connections by donations to election campaigns; some previous studies have used this metric. For example, Claessens et al. (2008) indicated that firms use donations to election campaigns to “buy” political favors. Some evidence suggests that, in Brazil, donations to successful candidates for the federal deputy (members of the lower legislative chamber) are associated with positive returns on the shares of donating firms (Claessens et al., 2008). Bandeira-de-Mello et al. (2012) argued that political connections made by way of donations to election campaigns were associated with improved performance. According to Lazzarini et al. (2015), political connections created through donations to election campaigns are employed in exchange for direct or indirect benefits to private firms. Samuels (2001) stated that when a firm donates to a political campaign, the politician receives money, and the company obtains a favor. One political favor that may be “bought” with donations to election campaigns is preferential access to bank financing, primarily from state-owned banks (Claessens et al., 2008).

There is some evidence that firms with political connections gain advantages. For example, Khwaja and Mian (2005) determined that firms with political connections in Pakistan (i.e. firms with at least one director taking part in elections) received twice the volume of resources from state-owned banks and had higher rates of default. Charumilind et al. (2006) provided evidence of a similar phenomenon in Thailand. They found that politically connected firms gained easier access to capital and were required to provide fewer guarantees to obtain long-term financing. Faccio et al. (2006) argued that banks decide to offer credit to politically connected firms because they presume that these firms will be bailed out by the government in the event of financial difficulties. Bandeira-de-Mello and Marcon (2011) suggested that preferential access to credit by politically connected firms is particularly important in countries where the government controls significant resources, or where the institutional apparatus is weak and ineffective at reducing information asymmetry or is unable to guarantee ownership rights. According to Boubakri et al. (2008), the use of political connections is one option that may reduce the activity financing costs of firms operating in such countries.

However, the advantages gained by politically connected firms may be mitigated by some factors, a phenomenon that we confirm in this study. Firms that donate to campaigns can distort their investments since politically connected firms allocate considerable resources to rent-seeking activities (Fisman, 2001), which mitigates financial gains (Chaney et al., 2011). It is not easy to track rent-seeking activities, but there is some evidence that this practice might eliminate any advantage gained from political connections (Faccio, 2010). For example, in Peru, bribes made to purchase favors were equivalent to the taxes that were not paid by the companies (De Soto, 1989). There is additional evidence showing that political connections may be costly for connected firms because the accounting performance of firms managed by connected CEOs is lower than that of nonconnected firms (Bertrand et al., 2007). Fan et al. (2008) and Shleifer and Vishny (2002) argue that politically connected firms extract resources to achieve objectives that are not consistent with maximizing the company value.

Based on a literature review, we argue that firms that donate to campaign elections distort investments and, as a consequence, have poor earning quality; this occurs even though they have a lower cost of capital given their preferential access to bank financing. In these firms, resources are extracted to achieve objectives that are not consistent with maximizing the value of the firm; rather, they are a means of “exchanging” favors with politicians with whom the firm has connections. The allocation of resources is therefore based on the expectation of gains on the part of the majority interest and/or the manager’s expectation of obtaining personal benefits. We propose the following hypothesis:

H1: A firm’s donation to election campaigns negatively affects the earning quality of the firm.

Improving the quality of earnings

Corporate governance is defined as the means by which executives ensure the rights of the company, shareholders, and stakeholders, as well as ensuring that stakeholders act responsibly to generate, protect, and distribute the wealth invested in the company (Aguilera and Cuervo-Cazurra, 2004; Aguilera et al., 2015). The theory that best supports corporate governance is agency theory (Jensen and Meckling, 1976). The essence of agency theory lies in the separation of roles between an agent and principal, where the relationship is contract based and the agent must perform a task for the principal. However, the agent does not always act in the best interests of the principal, which leads to agency problems. Conflict is controlled or reduced by corporate governance mechanisms that are designed to control agency problems in various forms.

Corporate governance mechanisms may be either internal or external (Walsh and Seward, 1990). Both types are used by shareholders to safeguard their returns on equity, provide transparent information disclosure, and protect stakeholder rights by monitoring executives and holding them accountable (Aguilera et al., 2015; Denis and McConnell, 2003; Heyden et al., 2014). Internal mechanisms include monitoring by the board of directors, incentive packages, and ownership concentration. However, despite such internal mechanisms, governance problems continue to exist (Jensen, 1993). Therefore, research has promoted external mechanisms (Walsh and Seward, 1990) that originate outside the company and that help to ensure that executives respect the rights and interests of the company and act in a transparent manner. Some external mechanisms include the legal system, the market for control, external auditors, stakeholder activists, rating organizations, and the media (Aguilera et al., 2015; Denis and McConnell, 2003; Heyden et al., 2014).

Agency theory states that managers may use gains from political connections for their own benefit and extract gains with limited interference, thereby exacerbating agency problems (Sun, 2019). Thus, the political capital might benefit a group of managers who will appropriate rents, which directly impacts the conflicts between the principal and agents. The political ties introduce more challenges to the firms to address the agency problems since the ties make it more difficult to track the manager’s misappropriation (Sun et al., 2016). Therefore, agency theory reveals the dark side of political connections (Liedong and Rajwani, 2018). Some researchers have noted that the relationship between managers and corporate political activities is associated with an increase in agency costs, a deviation in the audit process, and a decline in shareholder earnings and interests (Bebchuk et al., 2002). This opportunistic behavior occurs due to the opaque characteristics of political connections because, given the characteristics of CPA, shareholder monitoring will become more complex and more difficult to achieve (Hadani, 2012). This difficulty leads to information asymmetry between managers and shareholders that can be associated with an increase in agency costs (Dahan et al., 2013; Hadani and Schuler, 2013). However, effective corporate governance is expected to reduce the agency costs of politically connected firms, leading to an improvement in earning quality.

It is known that results are mixed regarding the impact of CPA on firm performance. Some aspects of governance may moderate the relationship between CPA and firm performance. For example, Hadani et al. (2015) have shown that CEO duality can negatively moderate the relationship between CPA and firm performance. Additionally, Ding et al. (2018) found that the management of real earnings mediated the effect of political affiliation on firm performance among privately held firms. In addition, Fan et al. (2007) reported that compared with firms that were not politically connected, those with politically connected CEOs underperformed by nearly 18%. In this study, we aim to extend understanding of how corporate governance moderates the relationship between CPA and firm performance.

When corporate governance mechanisms are better developed, investors tend to be better protected, and there is less erosion of the company’s future operational performance (Bertrand and Mullainathan, 2003; Filatotchev et al., 2005). This protection is a source of corporate governance-related regulations that improve transparency and discourage illegal or irregular practices (Christiansen and Koldertsova, 2009). Thus, once a firm is listed, it must comply with solid corporate governance practices even in developing markets with a weak enforcement of corporate governance rules. However, the agent does not always act in the best interests of the principal, and the use by managers of political connections gains for personal benefits (Liedong and Rajwani, 2018) can lead to a distortion of the firm’s earning quality. Therefore, we argue that better corporate governance mechanisms may reduce the distorted allocation of resources in firms and, as a consequence, reduce the harmful effects on the earning quality. We propose the following hypothesis:

H2: Corporate governance mechanisms positively moderate the relationship between a firm’s donations to election campaigns and the earning quality of the firm.

Empirical strategy

Research setting

Chaney et al. (2011) provided evidence that politically connected firms have poor earning quality. The authors used politically connected boards of directors as a proxy for political connections. In contrast, in this article, we test the agency problem by using donations to campaigns as a proxy for political connections. Legislation allows corporations to make direct contributions to candidates of up to 2% of their gross revenue. Companies represent the vast majority of donors. The high costs of election campaigns and the direct link between donors and recipients favor particularistic and quid pro quo relationships between firms and politicians.

Causal identification

It is not easy to track the direct gains from political connections, as there are usually subtler rent dynamics (Sun, 2019). Thus, the direct benefits of political connections are difficult to observe. However, most studies of political connections have attempted to use methodologies to provide findings that indicate that firm-specific gains may result from political strategies (Hillman et al., 1999). We used matching techniques to improve the identification of the causal relationship between political connection and earnings quality. In this design, observable features are used to match similar firms belonging to both the treatment and control groups. The treatment (control) group is composed of firms having (no) political connections. Using a set of covariates, we computed the propensity scores, or the propensity of “being treated,” of each firm. The resulting treatment and control groups include a subsample of matched firms that are similar in propensity score. The causal effect of political connections is the average outcome of the observed earning quality between the treatment and control groups (Rosenbaum and Rubin, 1983).

To quantify both the magnitude and sign of the average effect of political connections on earning quality, it is necessary to isolate the causal relationship between these effects from other influential factors. In this context, it would be ideal to simultaneously observe the same firm with and without political connections. However, since it is impossible to conduct such an observation, a feasible alternative is to compare two statistically identical groups during a period in which one group is subjected to a “treatment,” that is, the acquisition of a political connection through campaign donations, while the remaining group is not subjected to the identical “treatment.”

Another issue that should be clarified is related to the random selection of groups. If it were possible to select the members of both groups (treatment and control) at random, then no other firm characteristic would be responsible for determining the allocation of each company to either the treatment or control group; to clarify, there would be no selection bias. However, it is necessary in this study to utilize a method that minimizes the selection bias because the variable representing the treatment (donating to political campaigns) is correlated with the stochastic error term; therefore, the parameter that measures the sign and magnitude of the average effect of the treatment is subject to self-selection bias.

In view of this issue, the propensity score matching analysis method was selected because it reduces bias in estimates of treatment effects from a set of observed data (Rosenbaum and Rubin, 1983). Rosenbaum and Rubin (1983) state that bias may be eliminated from estimates of treatment effects if exposure to that treatment may be distributed entirely at random among individuals with the identical propensity score. As such, this method minimizes bias regarding self-selection and its influence on the parameter that measures the sign and magnitude of the average treatment effect.

To improve the robustness and validate our measurements, we tested the overlap and common support in the model, and thus we tested the balancing hypothesis. According to Lee (2013), the balancing hypothesis is satisfied when $D ⊥ X | P (X)$ , that is, when each value for the propensity score X (observable variables) has a similar distribution between the groups of politically connected firms (treatment) and the groups of unconnected firms (control). The sample was subdivided into five intervals, and each interval was tested to assess the absence of differences in either the mean propensity score or in the means for each observable variable for firms belonging to the treatment and control groups. Figure 1 shows that the distribution of propensity scores between the treatment and control groups shows good overlap and common support. We used four different matching methods to produce balanced samples: nearest-neighbor matching, radius matching, kernel matching, and stratification matching (Becker and Ichino, 2002). We computed the estimates of the average treatment effect of the treated (ATT) for every matched sample.

Figure 1.

Common support and overlap between treatment and control groups.

Sample and data

The population is composed of publicly listed firms in Brazil. This choice mitigates the problems that stem from finding reliable data when researching the strategies of firms operating in emerging economies (Hoskisson et al., 2000). We excluded observations for firms that reported a negative net equity or near-zero total assets since these firms could be in the process of corporate restructuring (operations involving transformations, takeovers, mergers and spin-offs) (Bandeira-de-Mello and Marcon, 2011). The resulting sample contained 4819 observations between 1998 and 2013.

We used official data for political contributions to electoral campaigns to identify firms with political connections. In Brazil, candidates are obliged to disclose this information, which is available on the Brazilian Superior Electoral Court’s website (Tribunal Superior Eleitoral). Based on the timeframe of our sample (1998–2013), we gathered data regarding donations for the 1998, 2002, 2006, and 2010 major general elections.

Observed outcome

Each company’s earnings quality is measured as the ratio of total accruals to total assets. Accruals are the difference between accounting profit and cash flow. Increases in accruals (and equivalent reductions in cash flow) can serve as an advance indicator of future deterioration in a company’s operational performance (Chan et al., 2006; Solan, 1996). The use of accounting data can help assess current and future performance. Management judgment regarding earnings is usually associated with discretionary accruals. Thus, managers can use discretionary accruals to increase their profits or to improve the informational value of informing shareholders regarding long-term firm earnings (Chaney et al., 2011).

Chaney et al. (2011) used accruals as a proxy for the profit quality published by politically connected firms. Following Solan (1996), Chan et al. (2006), and Chaney et al. (2011), total accruals were calculated as follows:

Total accruals (TCA) = Δ (current assets) - Δ (current liabilities)

where $Δ (current assets)$ is the variation in noncash current assets and $Δ (current liabilities)$ is the variation in current liabilities, excluding short-term debt and taxes payable.

Treatment variable

In this study, we considered political connection to be a binary variable that we coded “1” for firms that donated to the campaigns of any candidate running for president, state governor, senator, or a federal- or state-level representative. According to Samuels (2001), making donations to election campaigns is one method of “buying” a political connection. Once established, the relationship is clear: the politician “takes” the money, and the company is granted a favor. Donations to campaigns have been used as a proxy for political connections in previous studies (Claessens et al., 2008). Donations during the electoral cycle imply a commitment or a bond between a candidate and firm throughout the term during which the elected candidate is in office.

Covariates

We used a set of observable measures to compute the propensity scores. These variables have theoretical justification and statistical advantages for achieving comparable groups. The first are performance metrics that are measured by the ratio of net earnings to total assets and by the ratio of earnings before interest, taxes, depreciation, and amortization to total assets. CPA is motivated by a desire to ensure favorable political results, gain tax benefits, and subsidies and to win government contracts (Stigler, 1971; Shleifer and Vishny, 1994). We also used the level of a firm’s access to debt, or the debt-to-equity ratio, as measured by the total value of loans divided by the total liabilities and net equity. It is possible to access these data in the firms’ balance sheets. Politically connected firms possess greater leverage, and several prior studies have used leverage as a proxy for access to debt (Cooper et al., 2010; Faccio et al., 2006); it is also a widely used measure in finance. To measure the entirety of the obligations to financial creditors and shareholders, payments to banks were added to dividends paid and then divided by total assets. Boubakri et al. (2008), Charumilind et al. (2006), Claessens et al. (2008), Faccio (2006), and Khwaja and Mian (2005) have argued that politically connected firms appear to have lower risk, which would facilitate access to loans and make it possible to reduce the cost of obtaining financial resources, in turn reducing their obligations to financial creditors.

Another covariate is investment in capital goods, as measured by the value of a firm’s investments in fixed assets divided by its total assets. Political connections are the most important determining factor in access to long-term bank financing. Politically connected firms require fewer guarantees (i.e. fixed assets) to obtain long-term loans than unconnected firms (Charumilind et al., 2006). The allocation of resources obtained through bank financing is measured using two variables: the total resources obtained through bank financing used for the acquisition of capital goods divided by the sum total of bank financing and the total resources obtained through bank financing used to pay off obligations to creditors and shareholders divided by the sum of bank financing. According to both Fan et al. (2008) and Shleifer and Vishny (2002), politically connected firms extract resources to achieve objectives that are not consistent with maximizing the value of the firm. Faccio (2010) noted that connected firms tend to dedicate substantial resources to rent-seeking activities, which could eliminate the advantages gained from their political connections.

Finally, we conditioned the treatment effects for corporate governance instruments. We measured these instruments using dummy variables according to five levels of corporate governance, as classified by the Brazilian stock exchange. The five levels of corporate governance are bound by rules that go beyond companies’ obligations. According to the Brazilian Corporations Law, these rules ensure shareholder rights and guarantee the dissemination of information about the companies, mitigating risks related to informational asymmetry. The levels have the following characteristics: Bovespa Mais and Novo Mercado are the two upper levels of corporate governance, indicating commitment to a high standard of transparency with the market. These levels involve enforcement, which guarantees compliance with corporate governance mechanisms. Nivel 2, Nivel 1, and Tradicional are the levels with less enforcement, in which there is a relaxation of corporate governance mechanisms. For the two upper levels, 1142 observations were coded “1,” and the remaining lower levels of governance were coded “0.”

Results

Table 1 presents descriptive statistics and correlations for the dependent, independent, and control variables. Table 2 displays descriptive statistics for the covariates of the treatment and control groups, as well as the significant difference between them. We observed significant differences between the two groups. The first step in the matching technique is to estimate the propensity scores. We included all covariates to estimate the propensity of “being treated,” as well as the lagging variables of the observed outcome for three prior financial years; this step was an attempt to capture historical factors that may have caused differing trends in total accruals.

Table 1.

Means, standard deviations, and correlations.

Variables	Mean	Standard deviation	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
Dependent
(1) Quality of profit in terms of total accruals	1.68	1.78	1.00
Observable characteristics
(2) Quality of profit in terms of total accruals (time delayed, n−1)	2.19	1.29	−0.04	1.00
(3) Quality of profit in terms of total accruals (time delayed, n−2)	2.42	1.31	0.04	−0.00	1.00
(4) Quality of profit in terms of total accruals (time delayed, n−3)	2.61	1.33	−0.00	−0.00	−0.02	1.00
(5) Return on assets	2.76	1.28	0.03	0.08	0.05	0.00	1.00
(6) Earnings before interest, taxes, depreciation, and amortization	0.10	0.14	0.33	0.06	0.03	−0.03	0.73	1.00
(7) Access to debt	6.35	1.28	0.04	0.02	0.03	0.02	−0.03	0.00	1.00
(8) Obligations to financial creditors and shareholders	0.09	0.18	−0.60	0.03	−0.03	0.00	0.15	−0.07	0.48	1.00
(9) Investment in capital goods	7.45	1.11	0.00	0.00	−0.01	−0.01	0.02	0.12	0.16	0.02	1.00
(10) Allocation of debt to pay financial creditors and shareholders	22.12	3.65	−0.04	−0.00	−0.04	−0.03	−0.04	0.01	0.42	0.28	−0.04	1.00
(11) Allocation of debt for acquisition of capital goods	15.60	3.02	0.00	−0.04	−0.02	−0.02	0.06	0.07	0.13	−0.03	0.31	−0.09	1.00

Table 2.

Descriptive statistics and t-tests on observable characteristics.

Variables	Treatment		Control		Diff.
Variables	Mean	Standard deviation	Mean	Standard deviation	Diff.
Quality of profit in terms of total accruals (time delayed, n−1)	2.951	1.098	1.696	1.404	−3.106***
Quality of profit in terms of total accruals (time delayed, n−2)	3.352	1.096	1.784	1.440	−3.595***
Quality of profit in terms of total accruals (time delayed, n−3)	3.495	1.103	1.995	1.470	−3.177***
Return on assets	3.448	0.848	2.326	1.486	−2.971***
Earnings before interest, taxes, depreciation, and amortization	0.113	0.091	0.102	0.167	−2.583***
Access to debt	7.423	1.483	5.673	1.137	−4.614***
Obligations to financial creditors and shareholders	0.099	0.013	0.095	0.208	−0.624*
Investment in capital goods	6.290	3.475	8.196	4.292	5.791***
Allocation of debt to pay financial creditors and shareholders	23.446	3.712	21.282	3.622	−2.002**
Allocation of debt for acquisition of capital goods	13.948	2.790	16.650	3.151	3.030***

*Significant at 10%.

**Significant at 5%.

***Significant at 1%.

The results (Table 3) show that, irrespective of the matching method used, political connections fostered by way of donations to electoral campaigns increased accruals. The magnitude of the treatment effect was considerable. On average, politically connected firms had 36% higher total accruals than unconnected firms. This result lends support to hypothesis 1.To test hypothesis 2, we split the matched samples into groups composed of high- and low-level governance firms. We computed the ATT of each subsample and tested whether these estimates were significantly different for all four matched samples. The results (Table 4) showed an average decrease of 17% in the deleterious effect of political connections on the earning quality for firms with stronger governance. In addition, the estimated impact was statistically significant for all matching methods. These results lend support to hypothesis 2.

Table 3.

Average effect of political connections on quality of profit.

	Propensity score matching
	Nearest neighbor	Radius	Kernel	Stratification
Effect	1.312	1.066	0.759	0.599
Standard error	0.643	0.438	0.274	0.266
t statistic	2.041**	2.436**	2.772***	2.249**
N (treatment group)	1311	1311	1311	1311
N (control group)	804	1822	1822	1822
N (total)	2115	3133	3133	3133

*Significant at 10%.

**Significant at 5%.

***Significant at 1%.

Table 4.

Heterogeneity of treatment effects by governance level.

	Propensity score matching
	Nearest neighbor	Radius	Kernel	Stratification
Strong governance
Effect	1.045	1.163	0.887	0.890
Standard error	1.065	0.808	0.713	0.692
t statistic	2.921***	2.438**	2.244**	2.286**
N (treatment group)	356	356	356	356
N (control group)	191	390	390	390
N (total)	547	746	746	746
Weak governance
Effect	1.761	1.299	0.962	0.962
Standard error	1.245	0.788	0.676	0.929
t statistic	2.414**	2.650***	2.424**	2.035**
N (treatment group)	225	225	225	225
N (control group)	109	229	229	229
N (total)	334	454	454	454
Difference (strong–weak)	−3.227***

*Significant at 10%.

**Significant at 5%.

***Significant at 1%.

Finally, we conducted a robustness test to confirm our results. We divided our sample by accrual volume and by quartile; we used only the superior quartile to estimate the ATT. Table 5 shows the results, which confirm that connections derived from donations to political campaigns had a deleterious effect on earning quality.

Table 5.

Average effect of political connections on third quartile (Q3) quality of profit.

	Propensity score matching
	Nearest neighbor	Radius	Kernel	Stratification
Effect	0.663	0.278	0.357	0.724
Standard error	0.325	0.114	0.129	0.322
t statistic	2.340**	2.265**	2.294**	2.736*
N (treatment group)	155	155	155	155
N (control group)	68	104	104	104
N (total)	223	259	259	259

*Significant at 10%.

**Significant at 5%.

***Significant at 1%.

Discussion and conclusion

The results of our study show that politically connected firms with better corporate governance mechanisms may have superior earning quality than politically connected firms with reduced mechanisms of corporate governance. This finding addresses the agency theory view suggesting that managers pursue benefits at the expense of shareholders; thus, managers may use resources for corporate political activities that are not beneficial to the firm. This research builds on previous studies of CPA, firm performance, and corporate governance (Ding et al., 2018; Hadani et al., 2015; Hadani et al., 2017). Additionally, we contribute to the literature on agency theory by shedding light on agency-driven political activity. Studies on the institutional- and resource-based views have a limited ability to explain the distortion introduced by political connections (Liedong and Rajwani, 2018). Furthermore, agency theory is able to demonstrate that political connections can lead to agency problems since the connected manager can use the gains with political connections for his or her benefit (Hadani et al., 2018). Finally, the improvement of corporate governance generally curbs opportunistic behavior. We also show that it helps to curb, in particular, political opportunistic behavior.

Our findings have important implications for shareholders and boards of directors because they indicate that a better level of corporate governance will decrease the deleterious effect of political connections (through campaign donations) on earning quality. Thus, shareholders will incur fewer losses because they will be supported by more accurate levels of corporate governance that consider different internal and external mechanisms. Thus, regarding a donation to campaigns, it is essential that decision makers be aware of the dark side of the political connections (Liedong and Rajwani, 2018), require more explicit behavior with respect to campaign donations, and identify the impact on the firm’s earnings. There are also important implications for general managers regarding the less clear reports on earnings and expenses from the connection. Moreover, resource managers might be aware of misbehavior from the connected managers.

Previous research has shown that donations to successful candidates are associated with a positive return on the shares of donating firms (Claessens et al., 2008) and that bank financing is more accessible to firms that possess political connections (Charumilind et al., 2006; Claessens et al., 2008; Faccio et al., 2006; Khwaja and Mian, 2005). Moreover, there is evidence that the quality of earnings decreases in firms that are politically connected by way of their board of directors (Chaney et al., 2011). We address the results shown by Chaney et al. (2011) by demonstrating that firms that are politically connected by means of campaign donations also experience a deleterious effect on the quality of their earnings. Therefore, both types of political connections (via the board of directors and campaign donations) lead to deleterious effects on a firm’s earning quality. This result may be linked to rent-seeking activities since politically connected firms might have to devote some of their resources to activities such as lobbying to guarantee benefits, including subsides, grants, or tariff protection (De Soto, 1989; Fisman, 2001; Faccio, 2010).

Accruals are negatively related to future returns on shares, and an increase in accruals can be an early indicator of the deterioration of a firm’s operational performance (Chan et al., 2006; Solan, 1996). These results are consistent with evidence that political connections may be costly and that the advantages gained by politically connected firms may be mitigated by important disadvantages (Bertrand et al., 2007; Chaney et al., 2011; Fan et al., 2008; Shleifer and Vishny, 2002).

The present study has some limitations based on the data, specifically the access of the firms to long-term loans from public banks and private banks. Firms do not need to disclose the distinction between public and private banks in their financial reports, but such data would permit an understanding of whether the debt-to-equity ratio of firms derives more from public or private banks, and it might supply fine details regarding the study. An opportunity for future research is to test the heterogeneity in corporate governance levels that moderate political connections and accruals, such as CEO duality and shareholder concentration ownership. For example, compared to firms in which the CEO does not hold the chairman position, firms with CEO duality experience the same effect in the moderation between political connections and accruals. Another opportunity is to test the moderation of the external mechanism of corporate governance, for example, stakeholder activism or the media moderating political connections and accruals (Aguilera et al., 2015).

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors received support from the GV Pesquisa Foundadion (FGV-EAESP) for this article.

ORCID iD

Marina Amado Bahia Gama

References

Aggarwal

Meschke

Wang

(2012) Corporate political donations: investment or agency. Business and Politics 14(1): 1–38.

Aguilera

Cuervo-Cazurra

(2004) Codes of good governance worldwide: What is the trigger? Organization Studies 25: 415–443.

Aguilera

Desender

Bednar

, et al. (2015) Connecting the dots – bringing external corporate governance into the corporate governance puzzle. The Academy of Management Annals 9(1): 483–573.

Aldrighi

Postali

(2010) Business group in Brazil. In: The Oxford Handbook of Business Groups. New York: The Oxford University Press, pp. 353–386.

Bandeira-de-Mello

Marcon

(2011) The value of business group affiliation for political connections: preferential lending in Brazil. In: Annual Meeting of the Academy of Management. Montreal, Canada: AOM.

Bandeira-de-Mello

Marcon

Goldszmidt

, et al. (2012) Firm performance effects of nurturing political connections through campaign contributions. African Journal of Business Management 6(9): 3327–3332.

Baron

(1995) The nonmarket strategy system. Sloan Management Review 37(1): 73–85.

Bebchuk

Fried

Walker

(2002) Managerial power and rent extraction in the design of executive compensation. University of Chicago Law Review 69: 751–846.

Becker

Ichino

(2002) Estimation of average treatment effects based on propensity scores. The Stata Journal 2(4): 358–377.

10.

Bertrand

Mullainathan

(2003) Enjoying the quiet life? Corporate governance and managerial preferences. Journal of Political Economy 111(5): 1043–1075.

11.

Bertrand

Kramarz

Schoar

, et al. (2007) Politicians, Firms and the Political Business Cycle: Evidence from France.

12.

Boas

Hidalgo

Richardson

(2014) The spoils of victory: campaign donations and government contracts in Brazil. The Journal of Politics 76(2): 415–429.

13.

Bonardi

J-P

Holburn

GLF

Vanden Bergh

(2006) Nonmarket strategy performance: evidence from U.S. electric utilities. Academy of Management Journal 49(6): 1209–1228.

14.

Boubakri

Cosset

J-C

Saffar

(2008) Political connections of newly privatized firms. Journal of Corporate Finance 14(5): 654–673.

15.

Brasher

Lowery

(2006) The corporate context of lobbying activity. Business and Politics 8(01): 1–23.

16.

Chan

LKC

Jegadeesh

, et al. (2006) Earnings quality and stock returns *. The Journal of Business 79(3): 1041–1082.

17.

Chaney

Faccio

Parsley

(2011) The quality of accounting information in politically connected firms. Journal of Accounting and Economics 51(1–2): 58–76. Elsevier.

18.

Charumilind

Kali

Wiwattanakantang

(2006) Connected lending: Thailand before the Financial Crisis. The Journal of Business 70(1): 181–218.

19.

Choi

S-J

Jia

(2014) The structure of political institutions and effectiveness of corporate political lobbying. Organizational Science 26(1).

20.

Christiansen

Koldertsova

(2009) The Role of Stock Exchange in Corporate Governance. OECD Journal: Financial Market Trends 209–238. Available at: https://www.oecd.org/finance/financial-markets/43169104.pdf (accessed January 2014).

21.

Claessens

Feijen

Laeven

(2008) Political connections and preferential access to finance: the role of campaign contributions. Journal of Financial Economics, 166/2007. Available at: http://www.sciencedirect.com/science/article/pii/S0304405X07001353 (accessed 22 April 2013).

22.

Coates

(2012) Corporate politics, governance, and value before and after citizens united. Journal of Empirical Legal Studies 9: 657–696.

23.

Cooper

Gulen

Ovtchinnikov

(2010) Corporate political contributions and stock returns Main question. Journal of Finance 65(2): 687–724.

24.

Dahan

Hadani

Schuler

(2013) The governance challenges of corporate political activity. Business & Society 52(3): 365–387.

25.

De Soto

(1989) The Other Path: The Invisible Revolution in the Third Worlds. New York: Harper & Row.

26.

Denis

McConnell

(2003) International corporate governance. The Journal of Financial and Quantitative Analysis 38(1): 1–36.

27.

Ding

(2018) Government affiliation, real earnings management, and firm performance: the case of privately held firms. Journal of Business Research 83: 138–150.

28.

Faccio

(2002) Politically–connected firms: Can they squeeze the State? In: AFA 2003 Washington, DC Meetings, 23 March 2002, p. 56. DOI: 10.2139/ssrn.305099.

29.

Faccio

(2006) Politically connected firms. The American Economic Review 96(1): 369–386.

30.

Faccio

(2010) Differences between politically connected and nonconnected firms: a cross-country analysis. Financial Management 39(3): 905–928.

31.

Faccio

Masulis

McConnell

(2006) Political connections and corporate bailouts. The Journal of Finance 61(6): 2597–2636. Available at: http://onlinelibrary.wiley.com/doi/10.1111/j.1540-6261.2006.01000.x/full (accessed 19 April 2013).

32.

Fan

Huang

Oberholzer-Gee

, et al. (2008) Corporate diversification in China: Causes and consequences, CEI Working Paper Series from Center for Economic Institutions, Institute of Economic Research, Hitotsubashi University. Available at: http://apps.olin.wustl.edu/FIRS/PDF/2008/213.pdf (accessed 21 April 2013).

33.

Fan

Wong

Zhang

(2007) Politically connected CEOs, corporate governance, and Post-IPO performance of China’s newly partially privatized firms. Journal of Financial Economics 84(2): 330–357.

34.

Filatotchev

Lien

Y-C

Piesse

(2005) Corporate governance and performance in publicly listed, family-controlled firms: Evidence from Taiwan. Asia Pacific Journal of Management 22(3): 257–283.

35.

Fisman

(2001) Estimating the value of political connections. The American Economic Review 91(4): 1095–1102. Available at: http://www.jstor.org/stable/10.2307/2677829 (accessed 21 April 2013).

36.

Hadani

(2012) Institutional ownership monitoring and corporate political activity: governance implications. Journal of Business Research 65(7): 944–950. Elsevier Inc.

37.

Hadani

Doh

Schneider

(2018) Corporate political activity and regulatory capture: how some companies blunt the knife of socially oriented investor activism. Journal of Management 44(5): 2064–2093.

38.

Hadani

Munshi

Clark

(2017) The more you give, the more you get? The impact of corporate political activity on the value of government contracts. Business and Society Review 122(3): 421–448.

39.

Hadani

Nicolas

Jonathan

(2015) The CEO as chief political officer: managerial discretion and corporate political activity. Journal of Business Research 68: 2330–2337.

40.

Hadani

Schuler

(2013) In Search of El Dorado: the elusive financial returns on corporate political investments. Strategic Management Journal 181: 165–181.

41.

Heyden

MLM

Kavadis

Neuman

(2014) External corporate governance and strategic investment behaviors of target CEOs. Journal of Management 43(7): 2065–2089.

42.

Hillman

Hitt

(1999) Corporate political strategy formulation: a model of approach, participation, and strategy decisions. Academy of Management Review 24(4): 825–842. Available at: http://www.jstor.org/stable/10.2307/259357 (accessed 19 April 2013).

43.

Hillman

Keim

Schuler

(2004) Corporate political activity: a review and research agenda. Journal of Management 30(6): 837–857.

44.

Hillman

Zardkoohi

Bierman

(1999) Corporate political strategies and firm performance- indications of firm-specific benefits from personal service in the U.S. Government. Strategic Management Journal 20(1): 67–81. John Wiley & Sons, Inc.

45.

Hoskisson

Eden

Lau

, et al. (2000) Strategy in emerging economies. Academy of Management Journal 43(3): 249–267. Available at: http://www.jstor.org/stable/10.2307/1556394 (accessed 21 April 2013).

46.

Jensen

(1993) The modern industrial revolution, exit, and the failure of internal control systems. Journal of Finance 48: 831–880.

47.

Jensen

Meckling

WHH

(1976) Theory of the firm: managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3(4): 305–360.

48.

Khwaja

Mian

(2005) Do lenders favor politically connected firms? Rent provision in an emerging financial market. The Quarterly Journal of Economics 120(4): 1371–1411. Available at: http://qje.oxfordjournals.org/content/120/4/1371.short (accessed 21 April 2013).

49.

La Porta

Lopez-de-silanes

Shleifer

(1999) Corporate ownership around the world. Journal of Finance 54(2): 471–517.

50.

La Porta

Lopez-de-Silanes

Shleifer

, et al. (2000) Investor protection and corporate governance. Journal of Financial Economics 58: 3–27.

51.

Lazzarini

Musacchio

Bandeira-de-Mello

, et al. (2015) What do state-owned development banks do? Evidence from BNDES, 2002–09. World Development 66: 237–253.

52.

Lee

W-C

(2013) Propensity score matching and variations on the balancing test. Empirical Economics 44(1): 47–80.

53.

Liedong

Rajwani

(2018) The impact of managerial political ties on corporate governance and debt financing: Evidence from Ghana. Long Range Planning 51: 666–679.

54.

Luechinger

Moser

(2014) The value of the revolving door: Political appointees and the stock market. Journal of Public Economics 119: 93–107.

55.

Lux

Crook

Woehr

(2011) Mixing business with politics: a meta-analysis of the antecedents and outcomes of corporate political activity. Journal of Management 37(1): 223–247.

56.

Mellahi

Frynas

Sun

, et al. (2016) A review of the nonmarket strategy literature: toward a multi-theoretical integration. Journal of Management 42: 143–173.

57.

Pearce

(2001) How we can learn how governments matter to management and organization. Journal of Management Inquiry 10(2): 103–112.

58.

Ridge

Hill

Ingram

(2016) The signaling role of politician stock ownership: effects on lobbying intensity. Journal of Management, 44(5): 2116–2141.

59.

Rosenbaum

Rubin

(1983) The central role of the propensity score in observational studies for causal effects. Biometrika 70(1): 41–55.

60.

Samuels

(2001) Does money matter? Credible commitments and campaign finance in new democracies: theory and evidence from Brazil. Comparative Politics 34: 23–42.

61.

Schuler

Rehbein

Cramer

(2002) Pursuing strategic advantage through political means: a multivariate approach. Academy of Management Journal 45(4): 659–672.

62.

Shleifer

Vishny

(1994) Politicians and firms. Quarterly Journal of Economics 109(4): 995–1025.

63.

Shleifer

Vishny

(2002) The Grabbing Hand: Government Pathologies and their Cures, 09/30/2002. Cambridge: Harvard University Press.

64.

Solan

(1996) Do stock prices fully reflect information in accruals and cash flows about future earnings? The Accounting Review 71(3): 298–315.

65.

Stigler

(1971) The theory of economic regulation. The Bell Journal of Economics and Management Science. The RAND Corporation 2(1): 3–21.

66.

Sun

(2019) Corporate political ties in emerging markets. In: Grosse

Meyer

(eds.) The Oxford Handbook of Management in Emerging Markets. Oxford: Oxford University Press, p. 888.

67.

Sun

Hillman

(2016) The dark side of board political capital: enabling blockholder rent appropriation. Academy of Management Journal 59(5): 1801–1822.

68.

Walsh

Seward

(1990) On the efficiency of internal and external corporate control mechanisms. Academy of Management Review 15(3): 421–458.