* Some guys think they have found a reason behind why Illinois is so corrupt. Springfield is “isolated” from the state’s population center…

Campante and Do used a sophisticated statistical model to determine which capitals are the most isolated from their states’ population centers. They compared that measure of isolation with a database of convictions on federal corruption charges between 1976 and 2002. (Using federal convictions avoids the problem that some states may have more corruption convictions simply because local prosecutors are more aggressive. In addition, a long time period minimizes possible partisan bias).

The results showed an impressive effect, they wrote: “more isolated capital cities are associated with more corruption.”

* The accompanying graph…

* Explanation…

The graph illustrates their findings. The states with the highest number of corruption convictions relative to their population size are at the top of the chart. Those with the most geographically isolated capitals are at the left-hand side.

The most corrupt state capitals – Jackson, Miss., Baton Rouge, La., Nashville, Tenn., Pierre, S.D., Springfield, Ill., and Albany, N.Y., for example – are all more isolated than average. Nashville is the least so, being a major city in its own right although distant from other population centers in the state. Springfield and Pierre rank as the two most isolated on the list. The less isolated the capital (the further to the right), the more likely it is to rank low on corruption.

Isolation doesn’t explain everything, of course. Some states, such as Oregon, Washington and Vermont, have unusually low levels of corruption. But the impact of isolation appears strong.

Correlation does not equal causation. Keep in mind that most people are busted for corruption in this state outside Springfield. Rod Blagojevich was recorded cutting crooked deals in his Chicago campaign office. The overwhelming number of people indicted by the feds are low-level types out in secretary of state offices or whatever scattered throughout Illinois.

If Springfield’s alleged isolation was the reason for corruption, you’d expect to see more busts at the Capitol. Hasn’t happened.

Also, Springfield may be a few hours away from the population center, but it’s hardly isolated. Peoria, St. Louis and other mid-sized cities are nearby. Pierre, on the other hand, is literally out in the middle of freaking nowhere. But that capital is almost completely empty when they’re not in session - far more so than our own.

* Bogus…

What might cause the relationship between isolation and corruption, the researchers asked. One possibility was that newspapers, which provide most coverage of state governments, may be less likely to cover the capital when it is further from their circulation areas. So they examined the content of 436 U.S. newspapers, searching for references to state government. Sure enough, “in states where the population is more concentrated around the capital,” the study found “more intense media coverage of state politics, and therefore greater accountability.”

Yes, the Chicago newspapers do tend to play up local politics over Springfield, but they’ve done some excellent work as well and have top notch reporters assigned to the Statehouse.

       

46 Comments - Dirt Digger - Wednesday, May 23, 12 @ 10:20 am: Why the focus on corruption anyway? Blagojevich getting arrested was the most constructive thing he did for Illinois; when he actually tried to run the state it was demonstrably worse than having no governor. Our problem isn’t corruption, it’s that most of our politicians are incompetent to the point of total worthlessness. - ZC - Wednesday, May 23, 12 @ 10:26 am: Right around Blagojevich’s impeachment I played with a (much) less complicated model, and I kind of found the same thing though : states whose capitals were not in the top-five most populated cities, looked a little more corrupt on average. If you look at the states on the low end of the corruption-scale, an obvious alternate hypothesis arises: “Oh, those are just a lot of the states you’d expect to be less corrupt: Oregon, Minnesota, Wisconsin, Iowa.” But on the corrupt end of the scale, it’s a strange mixing. I wonder if they’re on to something, somehow. - Oswego Willy - Wednesday, May 23, 12 @ 10:27 am: “It’s Science …” - mark walker - Wednesday, May 23, 12 @ 10:30 am: Good that newspaper reporting also correlates with less corruption. As it should. - AnonAtty - Wednesday, May 23, 12 @ 10:31 am: The state capital should no longer be in Springfield, irrespective of the results of this statistical model. Springfield is a run-down, crime ridden, sorry excuse for a capital. School buses and politicians alike leave as quickly as they come. - CircularFiringSquad - Wednesday, May 23, 12 @ 10:31 am: We did not believe it was possible to have “academic research” dumber than Dick Simpson..but these guys take the cake..remember Simpson poured over newspaper clips for 40 years, found 1,500 folks on govt payrolls — suburban cops, et al — and claimed how corrupt IL is. He forgot to mention his universe is over 1 million. He admits that in his recent comparison with other federal districts no one looked at actual caes…they just too DOJ data and started to blab. Who shelled out the cash for this “research” - Oswego Willy - Wednesday, May 23, 12 @ 10:35 am: NEW STUDY - SQ MILE SIZE OF STATE CAPITALS IN SYNC WITH CORRUPTION OUTPUT STATE CAPITOLS WITH “DOMES” MORE CORRUPT “DOME”LESS CAPITOLS MOST CORRUPT STATES WITH MOST CORRUPTION SEEM TO HAVE MOST CONVICTIONS STATES WITH MOST RIVERS & STREAMS HAVE MOST BOATERS, STUDY FINDS - CircularFiringSquad - Wednesday, May 23, 12 @ 10:36 am: Sorry I answered my own questions Filipe R. Campante is Assistant Professor of Public Policy. He is interested in political economy and economic development, with special emphasis on understanding the constraints that are faced by politicians and governments beyond elections and formal “checks and balances”. His research has focused on the constraints imposed by the spatial distribution of population, the media, political protest, lobbying, and campaign contributions, and their effects on corruption, governance, polarization, fiscal policy, and political instability. Born and raised in Rio de Janeiro, Brazil, he holds a PhD from Harvard University, an MA from the Pontifical Catholic University of Rio de Janeiro, and a BA from the Federal University of Rio de Janeiro, all in economics. - Secret Square - Wednesday, May 23, 12 @ 10:37 am: Reminds me of the old joke about the following experiment: a guy drinks whiskey and soda one day, gin and soda the next day, and vodka and soda the day after that. He got drunk each time, so it had to be the soda that did it, right? - Cassiopeia - Wednesday, May 23, 12 @ 10:38 am: One of the professors who conducted this study is from Brazil and the other is from Singapore. I’m not sure where thier understanding of American politics came from other than textbooks and looking at data. - Conservative Republican - Wednesday, May 23, 12 @ 10:39 am: ==Yes, the Chicago newspapers do tend to play up local politics over Springfield, but they’ve done some excellent work as well and have top notch reporters assigned to the Statehouse. == True, but newspapers, a declining if not dying medium, have scaled down on personnel and resources for years. (To compare, the Trib had many fully staffed foreign news bureaus around the world in Col McCormick’s heyday, a thing of the past.) And despite that, they remain news outlets that are superior in quality to TV news. It seems to me that the newspapers cover corruption as an offshoot of official investigations or indictments (and when they occur, they suck up news space, e.g. Blagojevich), rather that expose corruption as an adjunct of their independent news investigations. PS, I think the location/corruption relationship isn’t worth a thought. E.g. Tennessee: the only other “population center” bigger than Nashville is Memphis, which doesn’t dominate the state like Chicago dominates Illinois. Even if true, you are not going to move to many state capitols to the biggest cities anyway. - Oswego Willy - Wednesday, May 23, 12 @ 10:39 am: ===Springfield is a run-down, crime ridden, sorry excuse for a capital. School buses and politicians alike leave as quickly as they come.=== This message is brought to you by the Springield Chamber of Commerce. “Of Course, I have never been to Springield, but that is what I hear …” is next. - He Makes Ryan Look Like a Saint - Wednesday, May 23, 12 @ 10:43 am: Corruption is a result of a direct DISCONNECT of the elected officials and the voters. Things that would get a state worker fired, is overlooked for legislators (see WICS News Report on Per diem from last night). Until this disconnect and more accountability for elected officials is dealt with, there will continue to be corruption. - Secret Square - Wednesday, May 23, 12 @ 10:51 am: “One of the professors is from Brazil and the other is from Singapore.” They must have studied political science from the Russian professor who predicted that by 2010 the U.S. would undergo another civil war and be divided up among China, Mexico, Canada and the European Union. Illinois and the rest of the Midwest were supposed to go to Canada, according to his map. - Anonymous - Wednesday, May 23, 12 @ 10:53 am: ==Springfield is a run-down, crime ridden, sorry excuse for a capital.== I wouldn’t want to be out walking around the White House after dark either. - Aaron - Wednesday, May 23, 12 @ 11:02 am: This study was on The Monkey Cage last week, and one might add that with Chicago’s history, the conclusions wouldn’t hold for Illinois, which of course might be suspected since study results are not uniform to populations. That’s why its science. However, to claim that less media coverage could lead to more corruption is not bogus. Are there other reasons for corruption? Of course, but that doesn’t mean the correlation doesn’t exist, and understanding the correlation, even if there is no related causal factor is still important for exploring relationships. What I’d like to see is the effect of more than just newspapers. - Political Junkie - Wednesday, May 23, 12 @ 11:04 am: ==Springfield is a run-down, crime ridden, sorry excuse for a capitol== Your right Chicago has no crime. (heavy snark) - Carl Nyberg - Wednesday, May 23, 12 @ 11:06 am: I’m more suspicious of the methodology to measure corruption than the thesis that being away from media attention encourages corruption. Here’s what I learned about corruption in Navy Recruiting Command. Reducing or increasing corruption is simple. If you want more corruption, decrease transparency and weaken rules against conflicts of interest. If you want less corruption, increase transparency and strengthen rules against conflicts of interest. - Carl Nyberg - Wednesday, May 23, 12 @ 11:09 am: Here’s what I’ve learned about politics in the real world. Zero corruption is not a useful goal. Institutions have to be able to perform their primary mission. Putting too much emphasis on avoiding any corruption is hare-brained if it means undermining the primary mission of the institution. - Ray del Camino - Wednesday, May 23, 12 @ 11:10 am: Not just “some guys” who found this effect, but researchers at Harvard’s Kennedy School. Nor is it a simple correlation, if you look at the paper. It is backed up by solid data and a review of the literature. It seems like pretty smart stuff to me: http://web.hks.harvard.edu/publications/workingpapers/citation.aspx?PubId=8379 - Rich Miller - Wednesday, May 23, 12 @ 11:11 am: ===It seems like pretty smart stuff to me:== Yeah, ignore where almost all the crimes took place and it’s probably fantastically smart. - ZC - Wednesday, May 23, 12 @ 11:12 am: Hey, wait, I wonder if people aren’t still pouring on from that ill-advised op ed Rich posted from the biologist. I’d better stick up for my discipline here … You can agree or disagree with the findings here, but this is -not- junk science, it’s a genuinely interesting correlation, and it deserves looking at. It’s not out of the blue either. George Ryan (who knows something about corruption) said something relatively similar, about why Illinois has a culture of corruption - I google searched for the quote, couldn’t find it right now. The argument I think is not so much about relative isolation but something that I suspect might be correlated, and the same thing people such as Kevin Phillips have alleged against Washington D.C.: that the town’s chief product is nothing -but- politics. It’s a city devoted disproportionately to one pasttime (as opposed to, say, Boston MA). The argument is that does things to the people who live, breathe and eat there. And it might have “ripple effects” outside the state capitol, when they’re not there. That all said, I’m suspicious of the finding too - in part because if you just dropped out three states - aforementioned MA, UT, and CO - the graphic above would look a heckuva lot less impressive. Just eyeballing it, the relationship wouldn’t go away though. - In 630 - Wednesday, May 23, 12 @ 11:18 am: Thanks for posting the link to the full paper, Ray. That really is interesting stuff, at least, for people open to it. - Dirty Red - Wednesday, May 23, 12 @ 11:22 am: = The overwhelming number of people indicted by the feds are low-level types out in secretary of state offices = Another Derrek Smith zinger! Is it that we’re so corrupt, or are our federal prosecutors more active than others? - Cincinnatus - Wednesday, May 23, 12 @ 11:24 am: - Oswego Willy - Wednesday, May 23, 12 @ 10:27 am: “It’s Science …” And shiny! I am pretty sure that moving the capital from Springfield to Chicago would significantly decrease corruption because of Chicago’s shining example… - Observer - Wednesday, May 23, 12 @ 11:28 am: So what we could take from this study is that if we move the state capitol to Chicago then politics in Illinois will be less corrupt. I find that hard to believe. - wordslinger - Wednesday, May 23, 12 @ 11:29 am: –They compared that measure of isolation with a database of convictions on federal corruption charges between 1976 and 2002.– That would presumably include federal corruption convictions of 29 Chicago alderman, 15 Cook County judges in Greylord, plus those in Operation Haunted Hall, Silver Shovel, Hired Truck, etc. How does distance from Springfield factor into those? - Yellow Dog Democrat - Wednesday, May 23, 12 @ 11:31 am: @Rich - I see an even bigger problem with the study. While the researchers relied solely on federal conviction data, they didn’t not exclude it to federal convictions of STATE OFFICIALS. In other words, a federal conviction of a Chicago alderman, Cook County judge, or DuPage County employee would be counted. A) This makes the geography of the state capitol moot; B) I don’t think anyone could argue there’s a shortage of media coverage of public corruption in Northeast Illinois. - Springfield.... - Wednesday, May 23, 12 @ 11:33 am: Is perfect. Zero flaws. Right RM? - Colossus - Wednesday, May 23, 12 @ 11:54 am: I think ZC made a good point here about the disconnect between the Capitol’s business and the actual population center and the ripple effect - this would directly undercut Rich’s focus on the location of the conviction. Coin flip: YDD is right about local/county convictions making the geography moot. All in all, an interesting starting point for further research. Not every study is a homerun conclusion-producing exercise, the majority are like stretching before the marathon: a necessary first step to the long hard work of getting to the end. - lincoln's beard - Wednesday, May 23, 12 @ 12:06 pm: I haven’t read the study, but note that in a major population center, you tend to get a more active, more prominent federal prosecutor. How often is it mentioned that Patrick Fitzgerald’s jurisdiction doesn’t even include Springfield? Can anyone here even name the U.S. Attorney for the Central or Southern District of Illinois? - Soccertease - Wednesday, May 23, 12 @ 12:10 pm: I hope those Campante and Do guys didn’t get paid much for their overthinking and overanalysis. My observation over 50 years is that it’s not Springfield that is corrupt, it’s Chicago politicians coming to Springfield making IL corrupt. - cover - Wednesday, May 23, 12 @ 12:26 pm: Any reason the researchers excluded Alaska, which has by far the most isolated capital? - 47th Ward - Wednesday, May 23, 12 @ 12:28 pm: Interesting study, but as Rich noted, correlation doesn’t equal causation. But here’s an even more interesting angle: when capitols are far removed from population centers, elected officials tend to be far removed from their spouses. If an elected official cheats on a spouse, in my book that makes them personally corrupt. If they cheat on their spouses, why would they have any qualms about cheating the system? If someone is personally corrupt, it wouldn’t be a stretch to see that correlate to public corruption. Of course, it would be difficult to get accurate data, so my theory is simply based on some personal anecdotal evidence and doesn’t prove anything. Sort of like this study. It’s interesting though. - olddog - Wednesday, May 23, 12 @ 12:29 pm: Since my academic research was in another field, I can’t really evaluate the methodology here. But I just skim-read the PDF file, and I’ve got a question about the findings: Where’s the discussion of Alaska? If I were testing a hypothesis about corruption in states with isolated capitals, I think I might want to take Alaska in account. They’ve had a significant amount of corruption there, and Juneau is even more isolated than Springfield is. At least we’ve got roads leading to Pawnee, New Berlin, Illiopolis and Chicago. - Ray del Camino - Wednesday, May 23, 12 @ 12:56 pm: The paper is not about Illinois, which is one data point out of 49. You can’t extrapolate from its general finding to one case. It is not saying that if you moved the state capitol to Chicago corruption would go down. Its findings are more subtle and its claims are modest. ZC’s and Colossus’s points well taken. - Wensicia - Wednesday, May 23, 12 @ 12:59 pm: non sequitur - Rod - Wednesday, May 23, 12 @ 1:32 pm: Clearly the roots of corruption in Springfield are not to be found in its supposed isolation. The cause is the Sangamon River, as is well known pollution in the river caused blue baby syndrome in river towns. Other unknown syndromes that manifest themselves in various forms of political corruption were also likely caused due to various political figures visiting Carpenter Park, but further study is required. It would be wise that the General Assembly authorize an appropriation for that purpose this session. - reformer - Wednesday, May 23, 12 @ 2:41 pm: Chicago media give much more coverage to aldermen and the mayor than to City legislators and the governor. Relative lack of attention to state government helps explain corruption. If the Chicago media had scrutinized legislative tuition waivers more consistently over the years, they would’ve been long gone. - Yellow Dog Democrat - Wednesday, May 23, 12 @ 2:41 pm: @ZC - I just ran through them in my head, and I’m guessing that only about a dozen state capitals are also their state’s largest city or at least qualify as a major metro area. I’m not saying there is no “Company Town” phenomenon, just that it would not explain the supposed disparity in corruption. I mean, honestly, do we really think that folks speeding down I-55 from Chicago or up from Southern Illinois breath a deep sigh of relief and whisper to themselves its okay to violate the public trust because they just crossed the Sangamon County line? - Retired Non-Union Guy - Wednesday, May 23, 12 @ 4:22 pm: lincoln’s beard @ 12:06 pm: I can name one of the Southern ADA’s … but only because he’s a distant relative - Kid Kansas - Wednesday, May 23, 12 @ 4:23 pm: “Lies, damn lies, and statistics” Mark T. from across the river. - Esquire - Wednesday, May 23, 12 @ 5:12 pm: I never knew that Cheyenne, Wyoming was close to a population center of any type. When I visited, the state capital had less than 50,000 residents. It was also located in a county that was fairly close to being the far Southeastern corner of the state. - zatoichi - Wednesday, May 23, 12 @ 5:49 pm: That’s a nice looking graph. Must have taken these guy’s grad assist a solid 15-20 minutes to chart all that info into Excel. So, is the Pearson Product-Moment Correlation Coefficient about a (-.20)? That puts the correlation coefficient at .04. Man, that’s tight predictive stats. - Langhorne - Wednesday, May 23, 12 @ 5:51 pm: Egghead nonsense. I pity the kids taking out student loans to take their classes. - Ray del Camino - Wednesday, May 23, 12 @ 8:09 pm: How many of you guys read the paper? This is a lot like the climate-change deniers dissing the science ’cause they don’t understand it. Sorry, comments for this post are now closed.