Corruption lowers public sector productivity, crowds out private investment, reduces foreign direct investment, and negatively impacts a host of social outcomes. These outcomes suggest that the optimal level of corruption is zero, however, reducing corruption requires the commitment of scarce resources. A non-zero level of corruption may thus be acceptable in the face of tight budget constraints. The question that remains is whether there is an optimal level of corruption within public sector organizations and, if so, what necessary level of oversight achieves the optimum. This paper develops a dynamic model of a public sector organization to illustrate the benefits and costs of using managerial oversight to address corruption. We find that if public sector managers desire a reduction of corruption to the economic optimum, they must utilize proactive management with a high level of initial oversight. Without an initial 'shock' in terms of increased oversight, managerial attempts to combat corruption may spur its increase.
Keywords: corruption, oversight, optimal control
JEL Codes: H8
Corruption, commonly defined as the abuse of public power for private gain (Amundsen, 2000; Johnston, 1996, 2005; Kurer, 2005; McPherson, 2014; Tanzi, 1998; Theobald, 1990) has become the cause celebre of problems related to economic development. Corruption appears to lower public sector productivity, crowd out private investment, reduce foreign direct investment, and negatively impact a host of social outcomes. Others have argued that, in the presence of excessive regulation or weak institutions, corruption may not be deleterious and, in extreme cases, improve economic efficiency (Huntington, 1968; Meon & Weill, 2010). Many analysts, however, discuss the impact of corruption and offer scant guidance on how to reduce corruption. Given the emergence of anti-corruption efforts as a key policy question over the past two decades, the question remains whether there is an optimal level of corruption within public sector organizations, and, if so, what is the level of oversight necessary to achieve this optimal level?
This paper develops a dynamic model of a public sector organization to illustrate the benefits and costs of using managerial oversight to address corruption. We explicitly move beyond the discussion of macroeconomic outcomes to an examination of corruption within an organization. If one accepts the assumptions that controlling corruption has a cost, is subject to diminishing returns, and that institutional structure is fixed beyond changes to monitoring and oversight, then the optimal level of corruption within an organization may not be zero. A non-zero level of corruption suggests that managers may need to balance the costs of corruption with the expense of oversight. The resulting level of corruption and oversight within an organization, under these assumptions, may be efficient relative to the level of oversight necessary to eliminate corruption. On the other hand, if we find that corruption's impact is so egregious that its elimination is paramount, then citizens may have to accept a lower level of output from public sector organizations in return for the elimination of corruption.
This paper contrasts political rhetoric calling for corruption's elimination with a proposition that considers the marginal benefits of reducing corruption relative to the marginal costs of enforcement. We explicitly model corruption within an organization and question whether the optimal level of corruption is zero. Assuming public sector managers desire to reduce corruption to the economic optimum, we argue that these managers must utilize proactive management with a high level of initial oversight. Without an initial shock in terms of increased oversight, organizational corruption may increase in response to managerial attempts to reduce it.
The remainder of the paper is structured as follows. In the next section, we briefly review the literature. The third section develops a dynamic model of corruption. The fourth section of the paper explores the model's findings. The last part concludes and offers suggestions regarding the control of corruption in public sector organizations.
A BRIEF REVIEW OF THE LITERATURE
Setting aside issues related to its definition and measurement, why should we care about the impact of corruption? A predictable economic environment is important for private investors (Johnson, McMillan, & Woodruff, 2002). When the environment assures investors that the returns on enterprise and investment accrue to the entrepreneur and investor, investment is more likely to ensue. Conversely, increases in uncertainty are likely to lead to increased volatility in investment flows, lowering economic growth (Lensink & Morrissey, 2006). Corruption, by increasing the uncertainty of investment decisions, thus has two primary effects on private investment decisions: one, investors expect lower returns due to increased costs, and two, the dispersion of outcomes is larger (Johnson, Kaufmann, McMillan, & Woodruff, 2000). Both effects serve to limit investment, which is critical to long run, sustainable economic development (Mauro, 1995, 1998).
Investigating the impact of corruption is not a new concept, yet quantifying the impact remains elusive (Foster, Horowitz, & Mendez, 2012; Jain, 2001; Kurer, 2005; Rose-Ackerman, 1999; Theobald, 1990). More recently, investigators have moved beyond broad measures of corruption at the national level to survey based approaches attempting to quantify the micro-economic impact of corruption (Reinikka & Svensson, 2006). Informal payments consumed, for example, almost 40 percent of expenses of respondents of new enterprises in transitional economies (International Monetary Fund, 2003). Only 13 percent of central government education transfers for non-wage expenditures in Uganda reached local governments during 1991-1995; local governments either appropriated the remaining funds for non-education purposes or corrupt activities (Reinikka & Svensson, 2005). While we must express a note of caution with respect to surveys that ask potential participants and victims of corrupt practices to report on the magnitude and frequency of such practices, there appears to be sufficient consensus in the literature that less corruption might translate into more resources available for investment. Moreover, strengthened public revenues, because of less "leakage" due to corruption, could translate into more public services or reduced taxes.
Although one might expect a broad consensus to exist concluding that corruption is bad, some authors have argued that under the proper circumstances, corruption may facilitate faster growth by serving as an "efficient grease" (Braguinsky, 1996; Kaufmann & Wei, 1999; Petrou & Thanos, 2014). Firms may also use corruption to reverse the government's errors in judgment (Leff, 1964; Lui, 1986). Bribery may allow "better" firms to bypass red tape and thus reward market performance (Bardhan, 1997; Lui, 1985). Corruption incorporates otherwise alienated groups, integrates them, and provides them with an alternative to violence (Arezki & Gylfason, 2013; Huntington, 1968). Acceptance of corruption may be a cultural norm but this norm may also be transient, that is, individual values are mutable depending upon the culture in which they reside (Barr & Serra, 2010; Harrison & Huntington, 2000).
Despite these interesting perspectives on corruption, the economics literature generally disapproves of such practices. Firms that devote more management time to bureaucrats spend more money on bribes and also pay a higher cost of capital (Gupta, Davoodi, & Alonso-Terme, 2002; Kaufmann & Wei, 1999). Corruption may strain the linkages between taxes and public sector goods and services, thus promoting tax evasion and growth of the unofficial economy (Friedman, Johnson, Kaufmann, & Zoido-Lobaton, 2000; Johnston, 2005). Higher levels of corruption and bureaucratic inefficiency appear to positively influence the unofficial economy's share of gross domestic product (GDP) (Alexeev & Pyle, 2003; Frye & Zhuravskaya, 2000; Hellman, Jones, & Kaufmann, 2003; Slinko, Yakovlev, & Zhuravskaya, 2005).
The literature has found that corruption limits economic development by inhibiting growth in per capita income, child mortality, and literacy (Chuang, Sung, Chao, Bai, & Chang, 2013; Kaufmann, Kraay, & Mastruzzi, 2011; Mo, 2001). Corruption also appears to adversely affect public and private investment (Del Monte & Papagni, 2001, 2007; Habib & Zurawicki, 2002; Mauro, 1995, 1998; van de Walle, 2006). Corruption may also affect economic policy by distorting the judgment of policymakers (Everhart, Vazquez, & McNab, 2009). One cannot assume that corruption is exogenous from the distortions it creates in the allocation of resources; distortions can generate incentives for increased corruption.
If one accepts the current consensus in the literature, that corruption negatively influences private and public sector outcomes, improving governance may be one way of combating corruption. Corruption thrives where states are too weak to control their own bureaucrats, to protect property and contract rights, and to provide the institutions that underpin an effective rule of law (Brunetti & Weder, 2003; Chowdhury, 2004; de Vaal & Ebben, 2011; Eigen, 2002; Shen & Williamson, 2005). Improving political accountability appears, as this line of reasoning goes, to improve governance and reduce corruption. Accountability allows for the punishment of politicians who adopt bad policies and the limitation of bureaucratic monopoly power, thereby more closely aligning politicians' and bureaucrats' preferences with those of the populace (Djankov, Porta, Lopez-de-Silanes, & Shleifer, 2002; Estache & Wren-Lewis, 2009; Laffont, 2003; Laffont & Meleu, 2001). Democratic elections, parliamentary systems, political...