Buying the honor of thieves? Performance pay, political patronage, and corruption

https://doi.org/10.1016/j.ijlcj.2020.100439Get rights and content

Abstract

Corruption undermines government effectiveness and performance pay may be used as a "legal bribe" to encourage bureaucrats to act honestly. However, if performance pay is awarded based on loyalty rather than performance, its use may aggravate corruption rather than reduce it. Collusion is facilitated when patronage crowds out merit in the public sector recruitment process, and this study, therefore, tests whether the relationship between performance pay and corruption is conditioned by patronage-based recruitment. A model of corruption is estimated using data from an expert survey of performance pay usage in government in over 100 countries and both a survey-based measure of patronage and a multi-source patronage risk index. A series of cross-country regressions suggest that the effect of performance pay on corruption is negative when patronage is low but either non-significant or positive in countries where patronage pervades the recruitment process. This study demonstrates how a popular anti-corruption reform may interact with competing incentives embedded in the context of implementation to worsen corruption.

Introduction

The uniformly deleterious effects of public sector corruption have made fighting it a priority for policy makers (Persson et al., 2013). Corruption entails leveraging the power and discretion associated with a public office to further the private interests of its holder (Transparency International, 2018). Corruption undermines trust (Chang and Chu, 2006; Hakhverdian and Mayne, 2012; Park and Blenkinsopp, 2011; Serritzlew et al., 2014; Swaleheen, 2011), development (Johnson et al., 2011; Méon and Sekkat, 2005), democracy (Stockemer et al., 2013), and public service provision (Berkovich 2016; Montes and Paschoal, 2016). Developed countries suffer from their share of scandals involving the abuse of power, but corruption is particularly prevalent in developing countries.

The strategy adopted to fight corruption is related to how its underlying causes are conceptualized (Persson et al., 2013). In the classic view, the decision to engage in corruption is a function of its expected payoff (considering the resources necessary to carry out the corrupt act and any moral sufferings associated with it), the severity of the punishment if caught, and the probability of being so (Becker, 1968). Many anti-corruption initiatives are designed to alter this calculus. For example, raising public sector wages should reduce the relative payoff of corruption and increase the relative cost of most punishments, such as termination. However, this solution can be both costly and politically unpopular (An and Kweon, 2017; Brianzoni et al., 2015; Klitgaard, 1988; Miller and Whitford, 2006). An alternative and potentially more cost-effective solution involves making pay contingent on performance (Skladany, 2009; Sundström, 2017). Performance pay can serve as a substitute for monies obtained through corruption, allowing bureaucrats to maintain (or even increase) their income without engaging in corrupt acts. In this sense, performance pay can be thought of as a legal bribe (Skladany, 2009), effective regardless of whether corruption is motivated by need or by greed.

Performance pay in government has been adopted in both developed and developing countries (OECD, 2017; World Bank, 2014). However, the characteristics of the implementation context of a given reform are relevant to its effectiveness (Ho and Im, 2015; Schick, 1998). Importantly, as performance pay will motivate bureaucrats only to the extent that performance and rewards are (perceived to be) in fact correlated (Atkinson et al., 2014), basing the distribution of rewards on honest evaluations of performance is crucial. However, while producing accurate and unbiased performance evaluations is difficult even in developed countries (Park and Berry, 2014), recent qualitative research by Sundström (2017) on South African fisheries inspectors suggests that performance evaluations in developing countries can be compromised due to the absence of honest evaluators. In such contexts, performance pay may be used to reward collusive behavior, inflaming rather than reducing corruption. In contrast to the capture of the state by private interests (Hellman et al., 2003), the capture of incentive programs by corrupt public servants undermines public service performance from the inside.

The criteria used to recruit public servants is important to the overall quality of the public bureaucracy (Dahlström et al., 2012; Rauch and Evans, 2000; Sundell, 2014) and may be relevant to the effect of performance pay on corruption. Bureaucrats hired and fired based on political or personal loyalties are less likely to check abuses of power by those on whom their job depends. In contrast, bureaucrats hired based on their skills and formal qualifications and possessing protections from capricious dismissal have a better chance of maintaining accountability to professional standards independent of their political or bureaucratic superiors (Harris et al., 2020; Maley, 2018; Oliveros and Schuster, 2018; Sundell, 2014; Zeng and Yang, 2017). Not only are civil servants recruited through meritocratic means likely to be of higher quality than patronage appointees, their autonomy can help reduce corruption by creating a collective action problem for the organization of collusion (Dhalström et al., 2012).

I argue below that the distribution of performance rewards is more likely to be decoupled from actual performance when recruitment is driven by patronage, which in turn will undermine the effectiveness of performance pay as an anti-corruption tool. Below, I present theoretical and empirical evidence in support of my position, and then describe the data I use to test my model. The results of a series of cross-national regressions demonstrate the joint effects of performance pay usage and patronage on national levels of corruption. This study contributes to the literature by demonstrating how a popular anti-corruption reform may interact with competing incentives embedded in the context of implementation to worsen corruption. I close by suggesting that improving public sector recruitment standards may be a prerequisite for the effectiveness of a variety of popular anti-corruption reforms.

Section snippets

Wages, performance pay, and corruption

Whether one takes the view that public servants seek to maximize their income beyond a subsistence level or prefer honest effort once this level is reached (Van Rijckeghem and Weder, 2001), wages play a key role in theoretical discussions of public sector corruption (Becker, 1968; Becker and Stigler, 1974; Evans and Rauch 2000; Klitgaard, 1988). Raising the wages of public servants gives them more to lose if they are caught engaging in corruption, and salary increases are understood to be part

Measuring corruption

To measure national levels of corruption, this study uses the 2015 Corruption Perceptions Index (CPI) published by Transparency International (2017),2 a measure that is widely used in cross-national studies of corruption. The CPI is based on data combined from 12 independent international

Methods

Although cross-country time series data is available for the measure of corruption used in this study as well as many of the control variables, the key independent variables do not exist in such a format. Consequently, the analysis is limited to the use of cross-sectional data. To test the hypotheses discussed in this study, ordinary least squares (OLS) regressions are used. Because some of the independent variables come from a single data set (albeit composed of the mean value of numerous

Descriptive statistics

Table 2 shows descriptive statistics for the variables used in this study. The CPI ranges from 8 (the least corrupt country) to 88 (the most corrupt). Performance pay ranges from 1 to 6.1, and has a mean value below the scale midpoint. The mean of patronage is above the scale midpoint, whereas the mean of the patronage risk index is below the midpoint. The direct and indirect measures of patronage are correlated at 0.63 (p < .001, not shown in table).

Fig. 1, Fig. 2 show the correlation between

Discussion and conclusion

Despite the enduring popularity of performance pay, the assumptions underlying it are of questionable validity in the public sector. For performance pay to effectively incentivize effort, performance must be believed to be a sufficiently linear function of it (Park and Berry, 2014; Vroom, 1964). However, public organizations are radically open systems with multiple principals, ambiguous and sometimes conflicting goals, a low tolerance for risk, and mandates to pursue socially valued but distant

Funding

This work was supported by an Incheon National University Research Grant in 2019.

References (85)

  • C. Van Rijckeghem et al.

    Bureaucratic corruption and the rate of temptation: do wages in the civil service affect corruption, and by how much?

    J. Dev. Econ.

    (2001)
  • R. Van Veldhuizen

    The influence of wages on public officials' corruptibility: a laboratory investigation

    J. Econ. Psychol.

    (2013)
  • S. Andersson et al.

    The politics of perception: use and abuse of transparency International's approach to measuring corruption

    Polit. Stud.

    (2009)
  • M.M. Atkinson et al.

    Why do governments use pay for performance? Contrasting theories and interview evidence

    Can. Publ. Adm.: Administration Publique Du Canada

    (2014)
  • M. Bauhr

    Need or greed? Conditions for collective action against corruption

    Governance-An International Journal of Policy Administration and Institutions

    (2017)
  • G.S. Becker

    Crime and punishment: an economic approach

  • G.S. Becker et al.

    Law enforcement, malfeasance, and compensation of enforcers

    J. Leg. Stud.

    (1974)
  • R.D. Behn

    The psychological barriers to performance management: or why isn't everyone jumping on the performance-management bandwagon?

    Publ. Perform. Manag. Rev.

    (2002)
  • N. Belle et al.

    Monetary incentives, motivation, and job effort in the public sector: an experimental study with Italian government executives

    Rev. Publ. Person. Adm.

    (2015)
  • G.A. Boyne

    Public and private management: what's the difference?

    J. Manag. Stud.

    (2002)
  • S. Brianzoni et al.

    Multiple equilibria in a discrete time growth model with corruption in public procurement

    Qual. Quantity

    (2015)
  • S.A. Brierley

    Combining patronage and merit in public sector recruitment

    J. Polit.

    (2020)
  • J.W. Campbell

    A collaboration-based model of work motivation and role ambiguity in public organizations

    Publ. Perform. Manag. Rev.

    (2016)
  • J.W. Campbell

    Efficiency, incentives, and transformational leadership: understanding collaboration preferences in the public sector

    Publ. Perform. Manag. Rev.

    (2018)
  • J.W. Campbell et al.

    Exchange Ideology, Performance Pay, and Pay Satisfaction: Evidence from South Korean Central Government

    (2019)
  • J.W. Campbell et al.

    At the expense of others: altruistic helping behaviour, performance management and transformational leadership

    Publ. Manag. Rev.

    (2016)
  • E. Chang et al.

    Corruption and trust: exceptionalism in asian democracies?

    J. Polit.

    (2006)
  • N. Charron et al.

    Careers, connections, and corruption risks: investigating the impact of bureaucratic meritocracy on public procurement processes

    J. Polit.

    (2016)
  • C.-A. Chen

    “A little is better than zero” or “pay enough or don't pay at all”? Evidence on the size of pay-for-performance across the sectors

    Publ. Person. Manag.

    (2018)
  • S. Choi et al.

    Employee satisfaction in agencies with merit-based pay: differential effects for three measures

    Int. Publ. Manag. J.

    (2017)
  • C. Dahlström et al.

    The merit of meritocratization: politics, bureaucracy, and the institutional deterrents of corruption

    Polit. Res. Q.

    (2012)
  • C. Dahlström et al.

    The QoG Expert Survey II Report

    (2015)
  • C. Dahlström et al.

    The QoG Expert Survey Dataset II [Data Set]

    (2015)
  • D.K. Dalal et al.

    Some common myths about centering predictor variables in moderated multiple regression and polynomial regression

    Organ. Res. Methods

    (2012)
  • B. De Maria

    Neo-colonialism through measurement: a critique of the corruption perception index

    Crit. Perspect. Int. Bus.

    (2008)
  • A. Dhillon et al.

    Corruption, intrinsic motivation, and the love of praise

    J. Publ. Econ. Theor.

    (2017)
  • D. Donchev et al.

    What do corruption indices measure?

    Econ. Polit.

    (2014)
  • E. Duflo et al.

    Incentives work: getting teachers to come to school

    Am. Econ. Rev.

    (2012)
  • K.M. Eisenhardt

    Control: organizational and economic approaches

    Manag. Sci.

    (1985)
  • A. Elias Sarker

    New public management in developing countries: an analysis of success and failure with particular reference to Singapore and Bangladesh

    International Journal of Public Sector

    (2006)
  • M. Gustavson et al.

    Organizing the Audit Society: Does Good Auditing Generate Less Public Sector Corruption?

    (2016)
  • A. Hakhverdian et al.

    Institutional trust, education, and corruption: a micro-macro interactive approach

    J. Polit.

    (2012)
  • View full text