Corruption and the quality of transportation infrastructure: evidence from the US states

Abstract

Few studies have linked public corruption to the quality of public infrastructure, particularly in developed countries. This article examines how public corruption affects the quality of transportation infrastructure in the context of the US states. Using state panel data for the period from 2002 to 2008, we found that public corruption had a negative impact on the quality of state roads, as captured by the International Roughness Index and overall road condition scores. This study concludes that the prevention of corruption is crucial to improving infrastructure quality and suggests preventive policy tools.

Points for practitioners

Transportation is one of the more corruption-prone sectors. This sector allows public officials discretion, attracts rent-seeking activities, and conceals malfeasance through secretive transactions. Our study finds that public corruption diminishes the quality of transportation infrastructure. Strengthening good governance is a critical way to improve public infrastructure performance. A variety of key anti-corruption strategies and actions are worth pursuing in the context of infrastructure development.

Keywords

corruption quality of transportation infrastructure US states

Introduction

This article examines how public corruption affects the quality of transportation infrastructure in a sub-national context. Transportation was selected as the focus of this study for a combination of theoretical and methodological reasons. To begin with, transportation is one of the more corruption-prone sectors (Kyriacou et al., 2015). Kottasova (2014) ranked it among the top three in this regard—in large part, because it involves large and complex construction projects on which it can be difficult to impose adequate and consistent quality control, management, and evaluation measures. Further, because most infrastructure projects require official government approval and therefore facilitate rent-seeking behaviors, the sector tends to be dominated by a small number of monopolistic firms with close links to government officials. As with the study of other forms of corruption, most existing studies of the impact of corruption on infrastructure and transportation have dealt with transition and developing countries (Al-Saidi, 2018; Gong, 2011; Kenny, 2009a, 2009b; Mashali, 2012; Quesada et al., 2013; Sancino et al., 2018; Tanzi and Davoodi, 1998).

Transportation infrastructure is obviously important for the developed world too, where this sector has also been shown to be corruption-prone. Thus, there is also ample anecdotal evidence of corruption in the various departments of transportation (DOTs) in the US states too. In one recent case, an employee of the Georgia DOT was charged with accepting bribes (US DOT, 2015); in another, three former employees of the South Carolina DOT and a contractor were charged as part of a six-year corruption and kickback scheme that cost taxpayers more than US$400,000 (Flach and Cope, 2016). These anecdotes invite more systemic research into the impact of public corruption on these agencies and the quality of transportation infrastructure they produce. This is the main objective of this study. Moreover, the relative uniformity of DOTs across the US in terms of funding, financing, and management facilitates this study (Neshkova and Guo, 2012).

The corruption literature generally defines corruption as the misuse of public office for private gain. There is a large volume of cross-country studies identifying the hazardous impacts of corruption on government spending, economic growth, and social equality. However, few studies have linked public corruption to the quality of public infrastructure, particularly in developed countries. In addition, theoretical and empirical studies on the relationship between public corruption and transportation infrastructure are especially rare in the US context. To fill the gap, the primary purpose of this study is to examine whether public corruption has links with the quality of transportation infrastructure. The comprehensive panel data sample covers all 50 US states from 2002 to 2008. The main finding of the article is that corruption reduces the quality of US state highway infrastructure.

The following sections discuss the relevant literature, develop our hypothesis accordingly, and describe our benchmark models and data. After discussing the findings from the analyses, we conclude and provide some policy implications.

Literature review and hypothesis

The literature on the effects of corruption on transport infrastructure is connected to studies about the effects of corruption on public investment, which primarily focus on bribery in the public investment sector. The public investment sector, including transport infrastructure, meets all the necessary conditions for corruption. It allows public officials discretion, attracts rent-seeking activities due to the large rents involved, and conceals malfeasance through secretive transactions with limited competition. Given the conditions, the managers of enterprises in this sector sometimes paid commissions, or bribes, to public officials in charge of public investment projects and expected favorable decisions from them in exchange for such bribes (Kenny, 2009a, 2009b; Sohail and Cavill, 2008).

In this perspective, Kyriacou et al. (2015) and Tanzi and Davoodi (1997) argue that there is a positive association between the extent of corruption and the size of public investment in a society. They provide relevant empirical evidence, mostly from developing countries. Hessami (2014) and the Organization for Economic Cooperation and Development (OECD, 2014) extend the literature by showing that the practice of giving bribes in the public investment sector has been rampant not only in developing countries, but also in the developed world. The OECD also emphasizes that transportation is among the most corruption-prone industries, along with the extractive industries, construction, and the storage sector.

Another stream of literature about the effects of corruption on public investment emphasizes the cost overrun of public investment projects, mainly due to bribe payments, which also applies to transport infrastructure. Enterprises providing public officials with commissions or bribes have incentives to recover the cost of bribes by overpricing their bids (Tanzi and Davoodi, 1997). A series of studies finds a positive association between the unit cost of construction and the extent of corruption in a country (Flyvbjerg et al., 2003; Kenny, 2006; OECD, 2008). Cantarelli et al. (2010) apply the same argument to explain the frequency of cost overruns in large-scale transport infrastructure projects. The World Bank (2011) also confirms that roughly one quarter of the 500-plus bank-funded roads have involved overcharges, and the average overcharge is 40% of the project price, with 8% in the US state of Florida, 15% in South Korea, 15–60% in Tanzania, and 20–60% in the Philippines.

Corruption has adverse effects on the various outcomes of public investment projects. Kenny (2006) suggests a new perspective for studies because bribe payments are weak indicators of the development impact of corruption. Low institutional quality resulting from corruption leads to misconduct in public investment management (Chakraborty and Dabla-Norris, 2009; Dabla-Norris et al., 2010; Grigoli and Mills, 2014; Haque and Kneller, 2008). Thus, in the perspective of governance and public management, corruption should have negative impacts on input and output management, project selection efficiency, maintenance, and the quality of construction in selected sectors as a consequence, which arguably deserve investigation (Kenny, 2009b; Tanzi and Davoodi, 1997).

In this regard, corruption results in low efficiency and reduced public investment products, such as poor project selection, delays in the design and completion of projects, and waste (Flyvbjerg et al., 2003; International Monetary Fund [IMF], 2015). In particular, the airport sector has attracted academic attention in this regard. For example, according to Yan and Oum (2014), airports turn out to be less productive in more corrupt countries because corruption lowers the strictness of government monitoring of airports. Randrianarisoa et al. (2015) find consistent results with a sample of 47 European airports. Calderon and Serven (2004) provide evidence of the negative effects of corruption on infrastructure in general, while Kyriacou et al. (2019) focus on the negative impact of corruption on transport infrastructure investment efficiency in a sample of 34 countries.

The effects of corruption on the quality of transport infrastructure separately from other general public investments have received less attention from scholars. The argument of Crescenzi et al. (2016) that investments in motorways should generate lower returns in weak institutional contexts is relevant for this discussion, although it does not directly refer to corruption. Tanzi and Davoodi (1997) provide anecdotal evidence on the negative effects of corruption on transport infrastructure, which captures the quality of transportation through an examination of paved roads in good condition and railway diesels in use.

Drawing upon the existing literature, there are at least three theoretical mechanisms through which public corruption may influence the quality of transportation infrastructure. First, public corruption reduces the resources available for building and maintaining transportation infrastructure. If, for example, an official pockets a payment intended for the government (i.e. embezzles public transportation funds for personal use), the available resources for the transportation sector are effectively reduced by the value of the payment, and therefore the quality of transportation infrastructure is similarly reduced.

Second, beyond depriving transportation infrastructure of funding in this way, corruption also decreases the efficiency of resource utilization in the transportation sector. This perspective is consistent with both principal–agent theory and the bureaucratic inefficiency model. The former theory views government officials as agents working on behalf of the public—the principals—to implement public policies and manage public service programs. However, the problem of agent opportunism may arise, that is, the agents may pursue their own interests in preference to those of the principals. Corrupt public agents violate the ideal principal–agent relationship and, in so doing, become less accountable to the citizens they are meant to serve and less likely to use resources efficiently.

Turning now to the bureaucratic inefficiency model (Niskanen, 1975), the idea here is that bureaucrats are interested in maximizing their own utility. The utility function of bureaucrats generally includes such aspects as salary, staff size, power, patronage, outputs of the bureau, ease of managing it, and so on, all of which are positively related to the size of the budget. Access to a discretionary budget makes possible various non-productive activities, such as expanding staff unnecessarily (Williamson, 1964), reducing the efforts of individual staff (Wyckoff, 1990), excessive risk aversion (Peltzman, 1973), and corruption (Wintrobe, 1997). Under the model of a maximized discretionary budget, public officials may be inclined to gratify their selfishness through corrupt practices and to waste public funds on unproductive activities. In sum, inefficient utilization of public resources can diminish the quality of transportation infrastructure (Reinikka and Svensson, 2005).

Third, based on rent-seeking theory, public officials may ignore citizens’ needs and preferences and distort government resource allocation. They are more likely to spend public resources on items for which they can obtain larger rents or bribes. Liu and Mikesell (2014) find that a US state with a higher level of public corruption is likely to spend more on capital construction items because these items generate more bribes for corrupt public officials. In the transportation sector, it is possible that high-level corruption may create a bias toward capital investment at the expense of regular maintenance needs. This kind of resource misallocation may result in poor infrastructure quality. Based on the preceding theoretical discussions, we develop the following research hypothesis:

Hypothesis 1: Public corruption is negatively associated with the quality of transportation infrastructure.

Methodology and data

Model specification

We model the quality of transportation infrastructure as a function of public corruption, which serves as our main test variable, and a number of control variables, as follows:

{Transportation Quality}_{i t} = β_{0} + β_{1} {(Corruption)}_{it} + \sum_{j = 1}^{J} a_{j} X_{jit} + γ θ_{i} + ψ ω_{t} + ε_{i t}

where:

{Transportation Quality}_{i t}

is a measure of the US state highway transportation infrastructure quality in state i and in year t;

{Corruption}_{i t}

captures the level of public officials’ corruption for state i in time t;

X_{jit}

is a matrix of control variables and

a_{j}

is a vector of coefficients for these empirical controls;

θ_{i}

is the state fixed effect to control for unobservable state attributes, such as landscapes and geographical locations;

ω_{t}

is a time-specific effect to control for periodic changes in national business and political cycles; and

ε_{i t}

is a random error term.

Dependent variable

In the US states, roads are the predominant component of state transportation infrastructure systems. Nearly 80% of state transportation funding is used to support road planning, construction, operation, and maintenance (Bartle and Chen, 2014; Goetz, 2007). Therefore, our dependent variable is road quality (Good Road). It is defined as the percentage of acceptable roads in state-administered highway systems. We assume that roads are in acceptable condition when their International Roughness Index (IRI) falls below 170 (US DOT, 2010). The IRI is a widely used civil engineering measurement. Lower IRI values are associated with higher ride and road quality. The quality of roads is considered acceptable below an IRI value of 170, and good when IRI values fall below 95. IRI values between 95 and 170 define fair roads (Chen, 2017, 2018). We rank the 50 US states according to our index of road quality by averaging state indexes for the period 2002–2008. The 10 states with the worst road quality are Georgia, Florida, Kentucky, Tennessee, Wyoming, Oregon, Montana, Utah, Ohio, and Arizona, and those with the best quality are New Jersey, California, Idaho, Maryland, Kansas, Hawaii, Oklahoma, Arkansas, New Mexico, and Vermont (for details, see Table 1).

Table 1.

Rankings of the US states (on average, 2002–2008).

Ranking	Corruption (population)	Corruption (employee)	Road quality	Ranking	Corruption (population)	Corruption (employee)	Road quality
1	Oregon	Nebraska	New Jersey	26	Texas	Connecticut	Wisconsin
2	New Hampshire	Oregon	California	27	Georgia	Oklahoma	Indiana
3	Nebraska	New Hampshire	Idaho	28	Massachusetts	Rhode Island	Connecticut
4	Colorado	Minnesota	Maryland	29	Oklahoma	Maryland	Colorado
5	Minnesota	Iowa	Kansas	30	Wyoming	Massachusetts	Iowa
6	Utah	Colorado	Hawaii	31	West Virginia	West Virginia	South Carolina
7	Iowa	Utah	Oklahoma	32	Virginia	Virginia	Texas
8	Kansas	Kansas	Arkansas	33	Pennsylvania	New York	Washington
9	Washington	Washington	New Mexico	34	Tennessee	Hawaii	Nebraska
10	Nevada	Wisconsin	Vermont	35	New York	Montana	Alabama
11	Wisconsin	North Carolina	Rhode Island	36	Hawaii	New Jersey	North Carolina
12	North Carolina	Vermont	West Virginia	37	New Jersey	Tennessee	Minnesota
13	Michigan	New Mexico	Louisiana	38	Ohio	Alaska	North Dakota
14	Indiana	Wyoming	Pennsylvania	39	Florida	Ohio	Virginia
15	New Mexico	Michigan	Alaska	40	Delaware	Delaware	Nevada
16	Vermont	Nevada	Massachusetts	41	Montana	Alabama	Arizona
17	South Carolina	Indiana	Missouri	42	Illinois	Pennsylvania	Ohio
18	Arkansas	Arkansas	Michigan	43	Alabama	South Dakota	Utah
19	Arizona	South Carolina	Maine	44	Missouri	Missouri	Montana
20	California	Idaho	Mississippi	45	Kentucky	Illinois	Oregon
21	Rhode Island	Maine	New York	46	South Dakota	Florida	Wyoming
22	Idaho	Georgia	New Hampshire	47	Alaska	Kentucky	Tennessee
23	Maryland	Arizona	South Dakota	48	Mississippi	North Dakota	Kentucky
24	Connecticut	Texas	Delaware	49	Louisiana	Mississippi	Florida
25	Maine	California	Illinois	50	North Dakota	Louisiana	Georgia

Note: The two corruption indexes are from the least corrupt states to the most. The road quality index is from states with the best road quality to the worst.

Key independent variable

To measure the extent of public corruption across the US states, we rely on the Report to Congress on the Activities and Operations of the Public Integrity Section, published by the US Department of Justice (US DOJ, 2002). The report includes the number of public officials convicted of violating federal corruption laws annually, state by state. All federal, state, and local governors, legislators, judges, and other public employees are subject to investigation by the Public Integrity Section (PIS). Multi-year panel data are available for the 50 states, from which data for the period from 2002 to 2008 were selected for inclusion in this study. These data were appropriate for this study because the report defines public corruption as “crimes involving abuses of the public trust by government officials,” which is consistent with the academic definition of corruption, or misuse of public office for private gain (US DOJ, 2002), as well as because they cover most corruption cases across the US states.

While we tested the relevance and validity of the methodology used by Liu and Mikesell (2014),¹ we also note here that some studies have questioned the reliability, relevance, correctness, and validity of the PIS data (e.g. Cordis and Milyo, 2016; Zhang and Kim, 2017). However, many others have used the data to capture the extent of public corruption across states (Butler et al., 2009; Glaeser and Saks, 2006). We ranked the 50 states according to our indexes of corruption by averaging state corruption for the period 2002–2008 according to the number of convictions per 10,000 public employees and per 100,000 members of the general population. According to the first measure, or the corruption variable in our benchmark model, the 10 most corrupt states were, in order, Louisiana, Mississippi, North Dakota, Kentucky, Florida, Illinois, Missouri, South Dakota, Pennsylvania, and Alabama, and the 10 least corrupt were Nebraska, Oregon, New Hampshire, Minnesota, Iowa, Colorado, Utah, Kansas, Washington, and Wisconsin (for details, see Table 1).

Empirical controls

The study adds a set of control variables. We use two variables to capture state variations in economic conditions: the real per capita state gross domestic product (GDP) (Ln Real GDP Per Capita) and annual GDP growth rate (GDP Growth Rate (%)). The variable of Ln Highway Funding is utilized to control for the effect of the amount of resources available for state highway transportation. It is measured as real per capita state total highway revenue. We also control for the size of state highway systems (Ln Highway Size), state urbanization rate (Urban Pop (%)), and the share of annual vehicle mileage travelled (VMT) attributable to heavy trucks on state-administered highway systems (Truck Share of VMT (%)). Finally, to control for environmental factors, we use the variables of Temperature and Precipitation. Temperature captures average winter temperature based on the months of December, January, and February. Precipitation captures the annual average Palmer Drought Severity Index, which ranges from –10 to +10, with negative and positive signs indicating dry and wet, respectively. Table 2 summarizes the descriptive statistics of all the variables.

Table 2.

Descriptive statistics: variable definitions and summary statistics.

Variable	Description	N	Mean	Std. Dev.	Min.	Max.	Data source
Good roads (%)	% of acceptable roads in state highway systems (IRI < 170)	350	82%	11%	48%	99%	Highway statistics
Overall road condition score	The total weighted road condition score (very good, good, fair, mediocre, and poor)	350	0.66	0.08	0.45	0.89	Authors’ construction
Corruption (employee)	(Number of convictions/numbers of state public employees) * 10,000	350	0.52	0.41	0	2.63	US DOJ
Corruption (population)	(Number of convictions/numbers of state population) * 100,000	350	0.36	0.32	0	2.5	US DOJ
Truck share of VMT	Heavy truck share of annual VMT (%)	350	11.0%	4.0%	3.0%	24.0%	Highway statistics
Urban population	% of population living in urban areas	350	67.95	15.27	37.8	99.1	Highway statistics
Highway size	Length of state-administered highway (lane miles)	350	168,129	112,198	9420	669,321	Highway statistics
Highway funding	Real per capita state total highway revenue per capita (2012 US$)	350	91.13	118.75	34.42	487.14	Highway statistics
Real GDP per capita	Real per capita state GDP (2012 US$)	350	37.34	15.15	6.24	72.48	Bureau of Economic Analysis
GDP growth rate	Annual growth rate of state GDP (%)	350	2.5%	2.4%	–5.5%	13.3%	Bureau of Economic Analysis
Precipitation index	Annual average Palmer Drought Severity Index	350	0.52	2.17	–6.11	7.19	National Climatic Data Center
Average temperature	Average winter temperature based on the months of December, January, and February	350	33.22	10.77	7	62.6	National Climatic Data Center

Estimation method

We analyze the impact of corruption on the quality of highway transportation infrastructure with the panel data of 50 US states in the period 2002–2008. This period is chosen for two reasons: (1) data availability; and (2) consistent reporting standards for state road quality. Due to the panel data structure, the study employs a two-way panel estimator with state and year dummies to control for both state- and year-invariant unobserved heterogeneity. The variance inflation factor (VIF) test confirmed that multicollinearity should not be a problem for this study since all variable scores for the VIF test are much less than 10. Due to the results of the Hausman tests, we used the fixed-effect estimator for model estimation.

We expect there should be a time lag between the occurrence of public corruption and its effect on the quality of transportation infrastructure for three reasons. First, federal prosecution of public corruption cases usually begins several years after the corrupt activities take place. Second, infrastructure construction projects typically require several years to complete. Third, the use of lagged values of public corruption avoids the potential endogeneity between corruption and the quality of transportation infrastructure of public agencies: on the one hand, corruption can compromise the quality of transportation infrastructure of state agencies; on the other, poorly performing public agencies tend to be vulnerable to corruption. To address these concerns, we used lagged values for our corruption variable.

Empirical findings

Main results

Table 3 present the main regression results on the effects of public corruption on the condition of state-administered highways (again, those with IRI values under 170 are in acceptable condition) of the 50 US states in the period 2002–2008. Models 1 to 4 are estimated with different lags of corruption. Model 5 is the full model with the inclusion of all lagged corruption variables. Across the five models, we find that the four-year lag corruption variable is statistically significant and negative in Models 4 and 5.² As predicted by the hypothesis, the variable of public corruption was significantly negatively associated with the road-quality variable (Good Road) and significant at the 0.1% confidence level. This finding confirms the negative impact of public corruption on road quality, which can be explained in terms of the inefficiency and resource misallocation resulting from public corruption. There might be one plausible reason why the effect occurs four years after the corruption conviction: public officials’ corrupt transactions may occur during the early bidding and construction period. However, transportation construction projects typically take several years to complete. The corruption effects may not materialize until the completion of transportation construction projects.

Table 3.

Effect of corruption on the quality of US state highways (panel two-way fixed effects).

Variables	Model 1	Model 2	Model 3	Model 4	Model 5
Key independent variable
Corruption (public employees)_first year lag	0.005 (0.005)				0.002 (0.006)
Corruption (public employees)_second year lag		–0.002 (0.003)			–0.001 (0.003)
Corruption (public employees)_third year lag			–0.004 (0.004)		–0.002 (0.004)
Corruption (public employees)_fourth year lag				–0.008*** (0.002)	–0.007*** (0.002)
Empirical controls
Ln (real GDP per capita)	–0.148*** (0.025)	–0.138*** (0.022)	–0.140*** (0.030)	–0.145*** (0.029)	–0.146*** (0.031)
GDP growth rate (%)	0.004*** (0.0005)	0.003*** (0.0004)	0.004*** (0.0004)	0.004*** (0.0004)	0.004*** (0.0005)
Ln (highway funding)	0.065*** (0.015)	0.061*** (0.013)	0.063*** (0.013)	0.063*** (0.012)	0.064*** (0.014)
Ln (highway size)	–0.699** (0.291)	–0.702** (0.291)	–0.705** (0.286)	–0.676** (0.285)	–0.685** (0.282)
Truck share of VMT (%)	–0.004 (0.066)	–0.014 (0.063)	–0.003 (0.063)	–0.010 (0.068)	–0.004 (0.064)
Urban population (%)	–4.25e–05 (0.0005)	–3.06e–06 (0.0005)	–2.01e–05 (0.0004)	–2.32e–05 (0.0005)	–5.35e–05 (0.0005)
Temperature	0.004** (0.002)	0.004** (0.002)	0.004** (0.002)	0.004** (0.002)	0.004** (0.002)
Precipitation	–0.0003 (0.0005)	–0.0003 (0.0005)	–0.0002 (0.0005)	–0.0002 (0.0005)	–0.0002 (0.0005)
State fixed effects	Y	Y	Y	Y	Y
Year fixed effects	Y	Y	Y	Y	Y
R-squares	0.741	0.742	0.746	0.786	0.794
Observations	350	349	348	347	347
Number of groups	50	50	50	50	50

Note: ***p < 0.01; **p < 0.05; *p < 0.1.

Concerning our control variables, we found that state GDP growth rate correlates positively with road quality. However, state real GDP per capita has an opposite sign. To interpret the results, we may apply the resource-dependency theory that a state with a higher level of economic affluence is likely to have more resources for maintaining a higher quality of transportation infrastructure. However, it is also possible that highways and transportation infrastructure are more likely to be used and worn in a state with a higher growth rate in economic activity. The level of state highway funding associates positively with road quality for the simple reason that resources are available to maintain and improve road quality in those states. Highway size is correlated negatively with road quality. This may be because a larger highway network increases the level of task difficulty (e.g. road maintenance).

Robustness checks

We used various procedures to check the robustness of our results.³ First, instead of the number of convictions per 10,000 public employees, we used the number of convictions per 100,000 people in the population as a proxy for corruption level. We found that there is no substantive difference in the estimated results across the variations of the corruption measures.

Second, we employed an alternative dependent variable: overall road condition score. State road conditions consist of five categories: very good, good, fair, mediocre, and poor. Each category is assigned with a weight point from 1 to 5, with 1 representing the worst condition (poor) and 5 representing the best condition (very good). Then, the number of miles in the corresponding category is multiplied by the weight assigned, which leads to the total weighted miles of road. To calculate the final overall road condition score, the total weighted miles of road is divided by the total graded road length and the maximum weight (5). It should be noted that the overall state road condition score is between 0 and 1 (a continuous variable). The new estimation results remained almost unchanged.

Third, we ran two-step difference Generalized Method of Moments (GMM) regressions to ensure the robustness of the empirical results. One of the most important reasons we applied GMM to check the robustness of our regression results is a potential endogeneity problem due to the corruption variable. It is generally understood that research on corruption should address the problem, and instrumental variable regressions are the most frequently used methods for this. In this regard, Arellano and Bond (1991) find that GMM provides relevant and exogenous instruments for endogenous variables, which means that GMM offers valid instruments and addresses the endogeneity problem. To check the robustness of the regression results, we use difference GMM and find that it provides consistent results.⁴

Conclusion

In this study, we investigated the effects of public corruption on the quality of transportation infrastructure across the US states. From a theoretical perspective, we observed that corrupt officials: gratify their selfishness by wasting resources on unproductive activities; are not accountable to citizens and political leaders and therefore have less incentive to use resources efficiently; and tend to allocate resources inefficiently, in particular, by directing them toward new capital investments rather than toward the maintenance and improvement of existing infrastructure. We accordingly hypothesized that inefficiencies and misallocation associated with public corruption would worsen the quality of transportation infrastructure not only in developing countries, but also in the advanced world. We presented strong empirical evidence to support our predictions even in the context of US states. Our findings indicate that government corruption has a lagged negative impact on the quality of state roads.

The findings presented here contribute to the public management literature in several significant respects. First, while the existing literature has devoted considerable attention to the political, social, and economic consequences of corruption in a cross-country setting, our research has focused instead on the relationship between public corruption and the quality of transportation infrastructure in a sub-national government context. As expected, we find that the former impairs the latter. In this regard, this study makes a definite contribution to the existing literature by providing evidence that the prevention of corruption is crucial to improving the quality of transportation, or a critical policy outcome. Second, using different lags of corruption variables, we demonstrated empirically that the corruption effects take some time to materialize, at least in the context of transportation infrastructure. Third, we have solidified the theoretical basis for understanding the detrimental effect of corruption on transportation infrastructure. Thus, we show that public corruption may squander resources meant for building transportation projects, misallocate public resources, and diminish the productivity and efficiency of public sector agencies, at least those in the transportation sector. Further research is needed to determine whether these findings can be generalized to other public agencies.

The study has limitations. Since the US DOJ refused to disclose the disaggregated data of the conviction measures, it was not possible for us to pinpoint the numbers of convictions directly related only to transportation infrastructure corruption cases. Thus, we do not argue that the coefficients of the corruption variable in our benchmark models capture the exact amount of corruption effects (which took place just in the sector of transportation infrastructure) on the quality of transportation infrastructure.

However, the PIS of the US DOJ argues that its conviction measure database captures the corruption cases of US public officials comprehensively, which included: accepting bribes; awarding government contracts to vendors without competitive bidding; accepting kickbacks from private entities engaged in or pursuing business with the government; overstating travel expenses or hours worked; selling information on criminal histories and law enforcement to private companies; mail fraud; using government credit cards for personal purchase; sexual misconduct; falsifying official documents; theft of government computer equipment for an international computer piracy group; extortion, robbery, and soliciting bribes by police officers; possession with intent to distribute narcotics; and smuggling illegal aliens (US DOJ, 2002). We think that multiple items are specifically relevant to transportation infrastructure according to the corruption literature, such as accepting bribes, awarding government contracts to vendors without competitive bidding, and accepting kickbacks from private entities engaged in or pursuing business with the government. We also reviewed the annual reports of the PIS over the period 2002–2008 from which we originally collected the conviction numbers for this study. The PIS used to choose some “representative” corruption cases annually and describe them in its annual reports, although not all the conviction cases were chosen for these descriptions. We found that there are a substantial number of cases relevant for transportation infrastructure each year.

This study also has important policy implications. To begin with, since corruption diminishes the quality of transportation infrastructure, fighting and preventing it must be a focus of all efforts to improve infrastructure performance. A variety of key anti-corruption strategies may be worth pursuing in the context of a given organization, including: strengthening the ethics training of public officials; promoting transparency with respect to resource allocation; increasing public scrutiny of government contracts and procurement procedures; enforcing stricter penalties on corrupt practices; and limiting political influence on hiring and promotion decisions (Lewis, 2006; Piotrowski, 2014). The problem of corruption is age-old, and developed societies need to be reminded that it is not confined to the transitioning and developing world. In either context, continued vigilance is required—as is continued study—if corruption is not to compromise the agencies that are meant to serve the public.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by KDI School of Public Policy and Management.

ORCID iDs

Can Chen

Cheol Liu

Notes

Author biographies

Can Chen is an Assistant Professor in the Department of Public Policy and Administration at Florida International University, USA. His current research interests include state and local infrastructure finance, fiscal transparency, and corruption. He has published widely in public administration and public budgeting and finance journals.

Cheol Liu is an Associate Professor at the KDI School of Public Policy and Management, Republic of Korea. His research interests are in budgeting, revenue administration, performance management, corruption, and e-leadership. He is the Head of the Experiment Lab for Public Management Research (EXPMR).

Jekyung Lee is an Assistant Professor at the Hainan University–Arizona State University Joint International Tourism College (HAITC), Hainan University, China. His current research focuses on financial management policy and performance budgeting. His research has been published in journals including Public Budgeting & Finance and The Journal of Public Budgeting, Accounting, and Financial Management.

References

Al-Saidi

(2018) Evaluation of decentralization and commercialization in the urban water sector of Yemen. International Review of Administrative Sciences. Epub ahead of print 15 September 2018. Available at: https://doi.org/10.1177/0020852318780984

Arellano

Bond

(1991) Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. Review of Economic Studies 58(2): 277–297.

Bartle

Chen

(2014) Future issues in state transportation finance. In: Willoughby

Katherine

Rubin

(eds) Sustaining the States: The Fiscal Viability of American State Governments. New York, NY: Taylor & Francis, pp. 211–233.

Butler

Fauver

Mortal

(2009) Corruption, political connections, and municipal finance. The Review of Financial Studies 22(7): 2873–2905.

Calderon

Serven L (2004) The effects of infrastructure development on growth and income distribution. World Bank Policy Research Working Paper 3400.

Cantarelli

Flyvbjerg

Molin

EJE

,et al. (2010) Cost overruns in large-scale transportation infrastructure projects: Explanations and their theoretical embeddedness. European Journal of Transport and Infrastructure Research 10(1): 5–18.

Chakraborty

Dabla-Norris

(2009) The quality of public investment. IMF Working Papers 9/154. Available at: https://www.imf.org/en/Publications/WP/Issues/2016/12/31/The-Quality-of-Public-Investment-23110

Chen

(2017) Does money matter for infrastructure outcomes? The effects of public infrastructure finance on state infrastructure quality. Journal of Public Budgeting, Accounting and Financial Management 29(3): 375–408.

Chen

(2018) Maximizing efficiency in state infrastructure finance: The role of competition, citizen monitoring capacity, and institutions. American Review of Public Administration 48(8): 915–928.

10.

Cordis

Milyo

(2016) Measuring public corruption in the United States: Evidence from administrative records of federal prosecutions. Public Integrity 18(2): 127–148.

11.

Crescenzi

Di Cataldo

Rodriguez-Pose

(2016) Government quality and the economic returns of transport infrastructure investment in European regions. Journal of Regional Science 56(4): 555–582.

12.

Dabla-Norris

Allen

Zanna

,et al. (2010) Budget institutions and fiscal performance in low-income countries. IMF Working Papers 10/80.

13.

Flach

Cope

(2016) Former S.C. transportation employees charged with corruption. Available at: https://www.thestate.com/news/local/article92365127.html (accessed 28 July, 2016).

14.

Flyvbjerg

Bruzeliua

Rothengatter

(2003) Megaprojects and Risk: An Anatomy of Ambition. Cambridge: Cambridge University Press.

15.

Glaeser

Saks

(2006) Corruption in America. Journal of Public Economics 90(6): 1053–1072.

16.

Goetz

(2007) State departments of transportation: From highway departments to transportation agencies. In: Plant

(ed.) Handbook of Transportation Policy and Administration. New York, NY: Taylor & Francis, pp. 121–144.

17.

Gong

(2011) An “institutional turn” in integrity management in China. International Review of Administrative Sciences 77(4): 671–686.

18.

Grigoli

Mills

(2014) Institutions and public investment: An empirical analysis. Economics of Governance 15(2): 131–153.

19.

Haque

Kneller

(2008) Public investment and growth: The role of corruption. Center for Growth and Business Cycle Research Discussion Paper Series 98.

20.

Hessami

(2014) Political corruption, public procurement, and budget composition: Theory and evidence from OECD countries. European Journal of Political Economy 34: 372–389.

21.

International Monetary Fund (IMF) (2015) Improving public investment efficiency in the G-20, Washington.

22.

Kenny

(2006) Measuring and reducing the impact of corruption in infrastructure. Policy Research Working Paper Series 4099, The World Bank.

23.

Kenny

(2009a) Measuring corruption in infrastructure: Evidence from transition and developing countries. The Journal of Development Studies 45(3): 314–332.

24.

Kenny

(2009b) Transport construction, corruption and developing countries. Transport Reviews 29(1): 21–41.

25.

Kottasova

(2014) World’s most corrupt industries. CNN, December 3. Available at: http://money.cnn.com/2014/12/02/news/bribery-foreign-corruption/

26.

Kyriacou

Muinelo-Gallo

Roca-Sagalés

(2015) Construction corrupts: Empirical evidence from a panel of 42 countries. Public Choice 165(1/2): 123–145.

27.

Kyriacou

Muinelo-Gallo

Roca-Sagalés

(2019) The efficiency of transport infrastructure investment and the role of government quality: An empirical analysis. Transport Policy 74: 93–102.

28.

Lewis

(2006) Governance and Corruption in Public Health Care Systems. Working Paper no. 78. Washington, DC: Center for Global Development. Available at: http://www1.worldbank.org/publicsector/anticorrupt/Corruption%20WP_78.pdf

29.

Liu

Mikesell

(2014) The impact of public officials’ corruption on the size and allocation of US state spending. Public Administration Review 74(3): 346–359.

30.

Mashali

(2012) Analyzing the relationship between perceived grand corruption and petty corruption in developing countries: Case study of Iran. International Review of Administrative Sciences 78(4): 775–787.

31.

Neshkova

Guo

(2012) Public participation and organizational performance: Evidence from state agencies. Journal of Public Administration Research and Theory 22(2): 267–288.

32.

Niskanen

(1975) Bureaucrats and politicians. Journal of Law and Economics 18(3): 617–643.

33.

OECD (Organization for Economic Cooperation and Development) (2008) Competition in the Construction Industry. OECD Policy Roundtables, DAF/COMP 36. Paris: OECD.

34.

OECD (2014) Foreign Bribery Report: An Analysis of the Crime of Bribery of Foreign Public Officials. Paris: OECD.

35.

Peltzman

(1973) An evaluation of consumer protection legislation: The 1962 drug amendments. Journal of Political Economy 81(5): 1049–1091.

36.

Piotrowski

(2014) Transparency: A regime value linked with ethics. Administration & Society 46(2): 181–189.

37.

Quesada

Jiménez-Sánchez

Villoria

(2013) Building local integrity systems in Southern Europe: The case of urban local corruption in Spain. International Review of Administrative Science 79(4): 618–637.

38.

Randrianarisoa

Bolduc

Choo

YY,

, et al. (2015) Effects of corruption on efficiency of the European airports. Transportation Research Part A: Policy and Practice 79: 65–83.

39.

Reinikka

Svensson

(2005) Fighting corruption to improve schooling: Evidence from a newspaper campaign in Uganda. Journal of the European Economic Association 3(2/3): 259–267.

40.

Sancino

Sicilia

Grossi

(2018) Between patronage and good governance: Organizational arrangements in (local) public appointment processes. International Review of Administrative Sciences 84(4): 785–802.

41.

Sohail

Cavill

(2008) Accountability to prevent corruption in construction projects. Journal of Construction Engineering and Management 134(9): 729–738.

42.

Tanzi

Davoodi

(1997) Corruption, public investment and growth. IMF Working Paper 97/139.

43.

Tanzi

Davoodi

(1998) Roads to nowhere: How corruption in public investment hurts growth. IMF Economic Issues No. 12.

44.

US DOJ (Department of Justice) (2002) Report to Congress on the Activities and Operations of the Public Integrity Section. Washington, DC: Public Integrity Section, Criminal Division, US Department of Justice.

45.

US DOT (Department of Transportation) (2010) 2008 status of the nation’s highways, bridges, and transit: Conditions and performance. Washington, DC.

46.

US DOT (2015) Former Georgia Department of Transportation employee charged with accepting bribes. August 14. Available at: https://www.justice.gov/usao-ndga/pr/former-georgia-department-transportation-employee-charged-accepting-bribes

47.

Williamson

(1964) The Economics of Discretionary Behavior: Managerial Objectives in a Theory of the Firm. Englewood Cliffs, NJ: Prentice-Hall.

48.

Wintrobe

(1997) Modern bureaucratic theory. In: Mueller

(ed.) Perspectives on Public Choice: A Handbook. Cambridge: Cambridge University Press, pp. 429–454.

49.

World Bank (2011) Curbing Fraud, Corruption, and Collusion in the Roads Sector. Washington, DC: The World Bank Group.

50.

Wyckoff

(1990) Bureaucracy, inefficiency, and time. Public Choice 67(2): 169–179.

51.

Yan

Oum

(2014) The effect of government corruption on the efficiency of US commercial airports. Journal of Urban Economics 80: 119–132.

52.

Zhang

Kim

(2017) Do public corruption convictions influence citizens’ trust in government? The answer might not be a simple yes or no. The American Review of Public Administration 48(7): 1–14.