Harvard survey: Measuring Illegal and Legal Corruption in American States

12/22/2014

Some Results from the Edmond J. Safra Center for Ethics Corruption in America Survey

by Oguzhan Dincer and Michael Johnston

Although corruption is not endemic in America as it is in several other countries, it does exist. According to the Justice Department, in the last two decades more than 20,000 public officials and private individuals were convicted for crimes related to corruption and more than 5,000 are awaiting trial, the overwhelming majority of cases having originated in state and local governments.¹ Understanding the causes and the consequences of corruption and designing the policies in the fight against it starts with measuring corruption itself. How do we measure corruption, an activity that requires secrecy? The most commonly used measure of corruption in American states comes from the Justice Department’s “Report to Congress on the Activities and Operations of the Public Integrity Section.” These data cover a broad range of crimes from election fraud to wire fraud. The measure, based on the Justice Department data, suffers from several significant problems, however.

1. The data report federal public corruption convictions; thus, corruption cases tried by state and local prosecutors are not included in the data. Federal prosecutors have considerable discretion over how much effort to put into investigating public corruption. Hence, the number of convictions depends not only on the level of corruption but also on levels of prosecutorial effort.² Prosecutors choose which cases to prosecute and which to decline so as to maximize their conviction rate and their visibility.³
2. The number of federal convictions is related to prosecutorial resources in a state.⁴
3. Partisan bias is likely in the prosecution of public officials by federal prosecutors, i.e., the U.S. attorneys. The decision to prosecute is up to the U.S. attorneys who are appointed by the President with the advice and support of the home state partisans. There is anecdotal as well as empirical evidence supporting this hypothesis. The unprecedented midterm dismissal of seven U.S. attorneys in 2007, for example, led to congressional investigations. Some were allegedly dismissed either because they did not pursue corruption investigations against prominent Democrats with sufficient vigor or because they did pursue investigations against prominent Republicans.⁵ Similar allegations have resurfaced in Virginia in 2014.
4. While data are reported annually, there is an unknown, and most likely variable, time lag between crimes and convictions.
5. The data give little to no indication as to the seriousness of a case.
6. The data cover only those officials who are caught and, of course, convicted.

There are other problems with measuring corruption by using conviction data, too. Over the three decades between 1980 and 2010, for example, South Dakota appears to be the most corrupt state—two and a half times more corrupt than New Jersey—as judged by federal convictions. This is quite surprising since the Dakotas were among the leading states in the movement against corruption in government that started in the late 19th century and continued through the 1930s.⁶ Prairie states such as the Dakotas, Minnesota, and Wisconsin are historically the least corrupt in the U.S. In fact, the only perceptions-based index measuring state level corruption in the U.S. ranked South Dakota as the least corrupt state in 1999.⁷ Any index based on convictions standardized for population is likely to be more variable in states with small populations, like the Dakotas, because a handful of cases will affect rates much more there than in, say, New York or Texas. For example, the number of convictions per 1 million people in Louisiana increased from 7 to 14 between 1989 and 1990 and declined to 2.5 in 1991—doubling in one year and decreasing more than fivefold in the next. Other states show similar variability in convictions, even though it is unlikely that underlying levels of corruption are changing so markedly.

An alternative to a measure based on convictions is a measure based on perceptions. Several international organizations such as Transparency International have constructed indices to measure corruption across countries based on the perceptions of experts since the mid-1990s. The only attempt to construct a perceptions based index for American states was by Boylan and Long in the late 1990s. They surveyed reporters covering state politics in each state and asked questions about public corruption. Unfortunately though, their survey was designed to construct an aggregate index covering all forms of corruption in all branches of government.

To be able to know where in state government to push for reform and to be able to measure the effectiveness of these reforms, we need to have a somewhat more disaggregated index measuring different forms of corruption in different branches of government. To construct such a disaggregated index, we surveyed the news reporters covering state politics in addition to the investigative reporters covering issues related to corruption during the first half of 2014. Boylan and Long make a compelling argument regarding why we should survey reporters rather than another group of professionals such as trial lawyers or small business owners to measure government corruption. Reporters have a better knowledge of state governments and spend a great deal of time observing the government officials and interacting with them. We identified close to 1,000 reporters through an extensive search of the internet and contacted them via email. Investigative Reporters and Editors, Inc. (IRE) helped us to contact the ones that we were not able identify on our own. We received a total of 280 responses. Unfortunately, in some states (Arkansas, Hawaii, Idaho, Massachusetts, Montana, North Dakota, New Hampshire, South Carolina, and South Dakota) only a small number of reporters responded to our survey. Hence while interpreting the results from these states we should be cautious. Surprisingly enough, we received no responses from Louisiana, which is historically one of the more corrupt states in America.⁸

Our main purpose is to construct perception-based indices measuring two specific forms of corruption across American states: illegal and legal. We define illegal corruption as the private gains in the form of cash or gifts by a government official, in exchange for providing specific benefits to private individuals or groups. It is the form of corruption that attracts a great deal of public attention. A second form of corruption, however, is becoming more and more common in the U.S.: legal corruption. We define legal corruption as the political gains in the form of campaign contributions or endorsements by a government official, in exchange for providing specific benefits to private individuals or groups, be it by explicit or implicit understanding.Such dealings are, in turn, one aspect of the broader issue of institutional corruption which, according to Lessig,

is manifest when there is a systemic and strategic influence which is legal, or even currently ethical, that undermines the institution’s effectiveness by diverting it from its purpose or weakening its ability to achieve its purpose, including, to the extent relevant to its purpose, weakening either the public’s trust in that institution or the institution’s inherent trustworthiness.⁹

According to several surveys, a large majority of Americans, both liberals and conservatives, think that donations to super PACs, for example, by corporations, unions, and individuals corrupt the government.¹⁰

We asked reporters how common were these two forms of corruption in the executive, legislative, and judicial branches of the government in 2013 in the state they cover in their reporting in 2013. The response scale ranged from “not at all common” to “extremely common.” For each reporter responding to the survey, we assigned a score of 1 if he/she chose “not at all common,” 2 if he/she chose “slightly common,” and so on. The score of 5 meant that the reporter responding to the survey perceived corruption to be “extremely common.” We then calculated the state scores as the median of these individual scores, which are presented in the tables and maps below.

Illegal Corruption in America

In none of the states is illegal corruption in government perceived to be “extremely common.” It is nevertheless “moderately common” and/or “very common” in both the executive and legislative branches in a significant number of states, including the usual suspects such as California, Florida, Illinois, New Jersey and Texas. Arizona is perceived to be the most corrupt state with legislative and executive branches both scoring 4. Among the states in which the legislative and executive branches are perceived to be corrupt, only in Florida and Indiana is illegal corruption in the judicial branch perceived to be “not at all common.” Idaho, North and South Dakota and the majority of the New England states—Massachusetts, Maine, New Hampshire, and Vermont—are perceived to be the least corrupt states with all three government branches scoring 1.

Executive Branch

In more than ten states, executive branches score 3 or higher in illegal corruption. In no state is illegal corruption in the executive branch perceived to be “extremely common” by the reporters. It is, on the other hand, “very common” in Arizona and New Jersey, and somewhere between “moderately common” and “very common” in Georgia, Kentucky, and Utah.

Illegal Corruption Executive

Legislative Branch

State legislators are perceived to be more corrupt than the members of the executive branches in a number of states. In almost half of the states, legislative branches score 3 or higher in illegal corruption. In ten states, illegal corruption in state legislatures is perceived to be “very common.” Among the states in which legislative branches score 4, New York and Rhode Island are particularly interesting since illegal corruption is perceived to be only “slightly common” in the executive branches in these two states.

Illegal Corruption Legislative

Judicial Branch

No states score 3 or higher in illegal corruption in judiciary. Nevertheless, even a score of 2 is still worrying since it is the judicial branch of the government that is supposed to try and convict the corrupt officials. In California, for example, illegal corruption in the judiciary is perceived to be somewhere between “slightly common” and “moderately common.” Both the executive and legislative branches in California are also perceived to be corrupt with the scores of 3 and 4, respectively.

Illegal Corruption Judicial

Legal Corruption in America

Legal corruption is more common than illegal corruption in all branches of government. Executive and legislative branches score 3 or higher in legal corruption in a large majority of states. In seven states, legal corruption in the judicial branch is perceived to be “moderately common” and in Nevada it is perceived to be “very common.” In ten states, both legislative and executive branches score 4 or higher and in Kentucky and New Jersey they score 5. In Illinois, Kentucky, Mississippi, and New Mexico legal corruption is perceived to be common not only in the executive and legislative branches but also in the judicial branch.

Executive Branch

In more than ten states legal corruption in executive branches is perceived to be “very common” or “extremely common.” Georgia, Kentucky and New Jersey, for example, suffer from not only illegal corruption but also legal corruption. Executive branches in these states score 4 or higher in both forms of corruption. In Kansas, Maryland, Maine, North Carolina, New York, Ohio, and South Carolina, on the other hand, while illegal corruption in the executive branches is perceived to be “slightly common” or “not at all common,” legal corruption is perceived to be “very common” or “extremely common.” Most of the states in which the executive branches are perceived to be corrupt also led the nation in television ad spending (all donors) on gubernatorial races that were extremely competitive in 2014. In Connecticut, candidates spent approximately $12.50 per eligible voter (more than $25 million total); in Illinois, $10 ($75.5 million total); and in Florida, $8.50 (close $100 million total). In Colorado, which is perceived to be one of the less corrupt states, on the other hand, television ad spending was only $4.5 per eligible voter (11.5 million total) although the race was extremely competitive.¹¹

Legal Corruption Executive

Legislative Branch

Legal corruption in the legislative branch is particularly worrying in almost all states. In almost half of the states, it is perceived to be either “very common” or “extremely common.” Only in a few states do legislative branches score 2 or less. Kansas, Maryland, North Carolina, Nevada and Wisconsin suffer from only legal corruption, while a dozen states including Alabama, Illinois, Kentucky and New York, in which legal corruption is perceived to be “extremely common,” suffer from both forms of corruption. In one state senate district in Nevada, for example, special interest groups spent close to $15 per eligible voter (more than $1.5 million) for television ads in 2014. In the entire state of Vermont, the total amount spent in the gubernatorial and lieutenant gubernatorial races was less than $1 million ($2 per eligible voter).¹²

Legal Corruption Legislative

Judicial Branch

Legal corruption in the judicial branch is more common than illegal corruption. Legal corruption in the judiciary is perceived to be more than “slightly common” in ten states and “very common” in Nevada, a non-partisan election state. Not surprisingly, states such as Alabama, Illinois, New Mexico, and Pennsylvania which hold partisan elections to elect judges at all levels score 3. Among the states that hold partisan elections, only in Texas is legal corruption in the judicial branch perceived to be “slightly common.” Judicial branches in Georgia, Wisconsin, and West Virginia, which elect their judges via non-partisan elections, score higher than 2. The majority of the states in which legal corruption in the judicial branch is perceived be “not at all common” elect their judges via merit selection. In Illinois (a partisan election state), in a retention election for one Supreme Court seat, television ad spending (only by special interest groups) per eligible voter in 2014 was twice as much as a contested election for two seats together in Michigan (a non-partisan election state) in which legal corruption in the judicial branch is perceived to be “not at all common.”¹³

Legal Corruption Judicial

Aggregating the Results: Most and Least Corrupt States

What are the most and least corrupt states, taking all three government branches into account? Although there is always some information lost in aggregation, which is particularly important in the measurement of corruption, it is still an important question. We simply add the median scores of each government branch and calculate the aggregate score of a state. The table below presents the states whose aggregate scores are in the highest (i.e., most corrupt) and lowest (i.e., least corrupt) quartiles. With respect to illegal corruption, Arizona is perceived to be the most corrupt state, followed by a second group of states, which includes California and Kentucky, and third group that includes Alabama, Illinois, and New Jersey. Idaho, Maine, Massachusetts, New Hampshire, the Dakotas, and Vermont are perceived to be the least corrupt states, followed by a second group of states that includes Michigan and Oregon. Kentucky is not only perceived to be illegally corrupt but also legally corrupt. With respect to legal corruption, Kentucky is the most corrupt state followed by Illinois and Nevada while Massachusetts is the least corrupt state followed Michigan, Oregon, and Vermont. It is all bad news for Alabama, Georgia, Illinois, Kentucky, New Mexico, New Jersey, and Pennsylvania as their aggregate scores are in the highest quartiles of both illegal and legal corruption. Not so bad news for the Dakotas, Idaho, Massachusetts, Michigan, Oregon, Vermont, and Wyoming, which are perceived least corrupt both illegally and legally.

Most and Least Corrupt States

Oguzhan Dincer is an Associate Professor of Economics at the Illinois State University and a Lab Fellow at the Edmond J. Safra Center for Ethics. He can be contacted at odincer@ilstu.edu.

Michael Johnston is Charles A. Dana Professor of Political Science at Colgate University and a Lab Fellow at the Edmond J. Safra Center for Ethics. He can be contacted atmjohnston@colgate.edu.

1. United States Department of Justice, “Report to Congress on the Activities and Operations of the Public Integrity Section,” 2012.

2. Richard T. Boylan and Cheryl X. Long, “Measuring Public Corruption in the American States: A Survey of State House Reporters,” State Politics & Policy Quarterly 3.4 (2003): 420-438.

3. Eric B. Rasmusen, Manu Raghav and Mark J. Ramseyer, “Convictions versus Conviction Rates: The Prosecutor’s Choice,” American Law and Economics Review 11.1 (2009): 47-78.

4. James E. Alt and David D. Lassen, “Enforcement and Public Corruption: Evidence from the American States,” Journal of Law, Economics and Organization 30.2 (2014): 306-338.

5. Sanford C. Gordon, “Assessing Partisan Bias in Federal Public Corruption Prosecutions,”American Political Science Review 103.4 (2009): 534-554.

6. Eric M. Uslaner, Corruption, Inequality, and the Rule of Law: The Bulging Pocket Makes the Easy Life (Cambridge University Press, 2008).

7. Boylan and Long, “Measuring Public Corruption in the American States.”

8. Louisiana is colored red in the maps.

9. Lawrence Lessig, “Foreword: Institutional Corruption Defined,” Journal of Law Medicine and Ethics 41.3 (2013): 553-555.

10. See, e.g. Brennan Center for Justice, “National Survey: Super PACs, Corruption, and Democracy,” New York University School of Law, 2012.

11. Based on the authors’ calculations using Chris Zubak-Skees, “State Ad Wars Tracker,” Center for Public Integrity, September 22, 2014.

12. Id.

13. Id.

source