9 responses so far ↓

• Donny Shaw // Mar 27, 2008 at 5:34 pm

  Perhaps we need an analysis of how all votes on a particular issue (i.e. eliminating tax breaks for oil companies) over a particular period of time (say, the 110th session of Congress) relate to trends in money coming from an industry that would be affected by the vote. If we can find a proposal that has been voted on several times recently, we can use it as a sort of control. So, using something like MAPLight’s “$ Near Votes” feature, we could see how variations in the amount (and timing) of money given to a party changed how well the party stuck together for the vote. This would work best if we had at least three instances of a vote. That way we maybe could detect a trend and start to rule out coincidence. This kind of analysis still couldn’t give us hard proof of quid pro quo, but it seems to me that it could produce some strong and interesting new evidence.

• aliza // Mar 27, 2008 at 5:37 pm

  This is interesting, but I’m not 100% clear on what your point is. Is it that people should assume B from A? That it is obvious that votes would change for a given party based on the amount of money received by that party relative to the other party?

  My [giving away my identity as the friend who said the thing] thought when I answered the question was that a) there is still room in my idealistic mind for the possibility that even if a given politician received money from a given interest group, it’s possible (I’m not saying probable) that it wouldn’t effect their vote and b) that the money could be going to other things besides congressional campaigns…i.e. executive branch campaigns, state parties, etc.

  Now of course, it’s likely that the 2 headlines essentially mean the same thing, especially in this case with oil, but I don’t know if it’s necessarily always the case. Sometimes there are other interest groups that give more money on the other side, or other issues which cause votes to NOT reflect contributions. And also, sometimes, such as in certain Farm Bill debates, party lines are not as important as other divides –such as fiscal responsibility or regional differences; or votes depend more on who has the majority than how members actually want to vote.

  But maybe I was missing your point…

• Dan Newman // Mar 27, 2008 at 6:29 pm

  Regarding the headlines:

  A) Big Oil Finances the Republican Party
  B) Congressional Votes Correlated with Big Oil Contributions

  These hypothetical headlines, or findings, are not the same. Finding (A) addresses the path of money–who it comes from, and who it goes to.

  Finding (B), in contrast, correlates inputs (the money) with outputs, or results (votes favorable to oil companies). Finding (B) addresses the relationship of campaign money to specific results in the world–actions taken by legislators that affect gas prices, coal plant pollution, and so on.

  In MAPLight.org’s view, campaign contributions bias our legislative system. Candidates who take positions contrary to industry interests are unlikely to receive industry funds and thus have fewer resources for their election campaigns than those who vote in favor of industry.

  A correlation between money given and votes cast–whether the vote was along party lines or not–suggests that money is biasing our legislative system–-either by affecting the issue positions of legislators once they are elected, or by supporting the (re)election of lawmakers who already hold an issue position favorable to the monied interest. This influence of money remains regardless of how much or how little party affiliation correlates with votes on a given bill.

  On May 7, 2007, the Senate passed an amendment to block Americans from importing prescription drugs from Canada and other countries. Drug manufacturers wanted this amendment to pass. They gave more than twice as much money, on average–$100,104–to each Senator voting Yes, to pass this amendment, as they did to each Senator who voted No–$37,887. (http://www.maplight.org/map/us/bill/60939/SBE4xw/votes/vote-294039 )

  This striking correlation between money and votes exists on this amendment despite the vote not falling closely on party lines. You can see each Senator, their party, and how they voted, here: http://www.maplight.org/map/us/bill/60939/SBE4xw/votes/votedetail-294039

• Joshua Tauberer // Mar 28, 2008 at 6:14 am

  Donny- Yes, that’s precisely the type of thing that needs to be done. But are there issues that were voted on in near identical form multiple times? I found one (academic) paper on the subject about some bill in the 90s voted on twice, and it found (iirc) no significant effect of money. If you can find the votes, let’s do the analysis!

  Aliza- Btw, thanks for being my guinea pig for the headlines. / I guess the point was that while the headlines seem different, the facts that back them up are not independent. They are two sides to the same coin. Framing it one way or another is a mere matter of framing the issue. / When party lines aren’t as important in votes, as you raise, it lends itself to more careful numerical analysis. So actually that’s a good thing for establishing the relation between money and votes.

  Dan- We’ve talked about this a lot so I know I’m starting to sound like a broken record player to you. I agree with you up to a point, specifically that “A correlation between money given and votes cast–whether the vote was along party lines or not–suggests that money is biasing our legislative system”. So I *do* think that there is a bias in the system, but I don’t think that the numbers being presented actually show that at all. Further, I agree with you that “This influence of money remains regardless of how much or how little party affiliation correlates with votes on a given bill.”, but disagree that the data *shows* the influence regardless of the correlation.

  For instance, I found (as we’ve talked about) one case where if you look just at contributions to one party, the money coming from supporting orgs is actually going more to those who voted against the bill. Given that politicians don’t usually change their party affiliation to get more money, clearly in this case there is no bias caused by money in the direction of corruption. However, if you look at the money across both parties, a correlation in the direction of corruption comes back. We know from the first part that there’s no money bias, but if you look at the overall correlation you find one. So clearly an overall correlation does not mean that there is in that case necessarily any bias.

• Joshua Tauberer // Mar 28, 2008 at 6:36 am

  More to Dan- The particular amendment you cite is a great example — I agree there’s something there. It’s a good example for two reasons: The vote is not at all along party lines, and the interest group is not giving money overwhelmingly to one party over the other. Of the 25 Members getting the most money from the interest group, they are almost exactly evenly split between Democrats and Republicans.

  These two criteria were not the case for the oil example or the vote I talked about in my last post (related to your analysis).

  Because those two criteria hold, after you break it down by party the correlation is still as huge and frightening. Average contributions:

  Voted Yes Voted No
  Democrats $113,164.40 $32,632.93
  Republicans $88,927.27 $54,667.64

• Ellen Miller // Mar 28, 2008 at 11:58 am

  Got to weigh in here with some real world stuff. Take a look at this re: Jay Rockefeller and Telco money. http://blog.wired.com/27bstroke6/2007/10/dem-pushing-spy.html

• Joshua Tauberer // Mar 28, 2008 at 4:31 pm

  Mmm, those are some great graphs.

• Skye Bender-deMoll // Mar 31, 2008 at 7:40 pm

  Hi Josh,
  Couple of responses:

  (Fist, a disclaimer. I am heavily involved in the OilMoney site, ‘tho mostly on the data backend end visualization, I didn’t actually do the analysis Josh is referring to, but am fairly familiar with the numbers and can give my personal opinion )

  >Of all the ways that money is bad for politics, it >doesn’t seem to
  >actually cause any Member to change his mind >on a vote, as far as the
  >academics who study this can find.

  Are you sure about this? My understanding from the list is that mostly researchers find what they are expecting, but if you do a meta-analysis of lots of studies, there is pretty strong relationship between money and votes, and some researchers are willing to give an estimate of how many votes would have gone the other way:

  Roscoe, D.D. and Jenkins, S. (2005) A Meta-Analysis of Campaign Contributions’ Impact on Roll Call Voting Blackwell Synergy - Social Science Q, Volume 86 Issue 1 Page 52 - March 2005 (Article Abstract) http://www.blackwell-synergy.com/links/doi/10.1111/j.0038-4941.2005.00290.x/abs/

  direct link to the pdf: http://campus.murraystate.edu/academic/faculty/mark.wattier/Roscoe&Jenkins2005.pdf

  ‘Tho I think most people agree that the impact of money and lobbying is mostly on what bills make it to the floor at all, and the language that is in them, more than on how reps actually vote.

  >It’s a very serious matter when one reports on the relation between
  >money and votes in Congress because we all *want* to believe in deep
  >corruption (what a scoop it is) and we are eager to accept such claims.

  Personally, I belive that there is pervasive culture of relatively shallow corruption. So money and power greatly influence what legislation gets written, and subtly bias the entire legislative process in favor of big business and moneyed interests, often at the expense of the interest of the general public. Most of the time this process takes place in a perfectly “legal” way, but it is a way that is at odds with the ideals of transparency and public democracy. And most people don’t know that it is going on. Only the flagrant violations of trust make news. (I do not have any sources to back up this opinion)

  >A) Big Oil Finances the Republican Party
  >B) Congressional Votes Correlated with Big Oil Contributions

  I’m not sure if this was a comment about Oil Change’s Report. I belied they claim “we found a strong correlation between the tendency to vote for Big Oil and increased amounts of money received from the oil industry” This seems like a fairly neutral claim, which doesn’t try to say anything about the direction of causality (i.e. if donors gave because of shared ideology or an expectation of a specific voting outcome). Definitely voting for the oil industry is framed as being a bad thing in the report, that is Oil Changes value judgement, which should be considered as a claim somewhat from the claims about a correlation. Which ever way the association goes, it indicates strong degree of alignment between the ideals of certain oil companies and certain politicians. I think that the nice thing about the tool is that you can use it to figure out which oil companies and which politicians seem most likely to be influenced and explore various possible explanations based on actual data.

  >We all already know that Big Oil gives more to Republicans than to
  >Democrats. If you didn‚t know, you know now. It‚s an interesting point;
  >no qualms there. Since the Republicans support big business and the
  >Democrats support the environment (whatever, you get the idea), it makes
  >sense. Fine.

  You may have known that, but I think that 90% of people in America don’t. I certainly didn’t know that until we ran the numbers. I also don’t think Democrats have a very good environmental record

  >But because of that, I know immediately that any vote related to Big Oil
  >is likely to go down with an uneven distribution of money coming from
  >Big Oil between Yes votes and No votes. In fact, it has nothing at all
  >to do with Republicans and Democrats having different views on oil in
  >particular. It‚s just that Republicans and >Democrats either all vote
  >together (on naming post offices) or vote against each other (on
  >everything else). The votes where everyone agrees are not relevant here:
  >you can‚t have an uneven distribution of money between Yes and No votes
  >when there aren‚t No votes to begin with.

  For the sample of Energy, Climate and War votes chosen by Oil Change for this project, there is actually a pretty good range of voting patterns. It helps that we are using an aggregate score across several votes, it widens the spread a bit. Scores are expressed as % of times voting with the interests of the Oil industry (only Yea and Nay are counted, not NV). The distribution is weakly bi-modal, so although there are definitely more “0%” and “100%” scores, there is a fairly wide range in the middle as well. Mean (across both chambers) is about 56%

  >Since the relevant votes are
  >almost always split on party lines, of course there is going to be a
  >correlation between money and votes.

  If you use the VoteTracker tool you can see that there is actually quite a range within parties as well. There are some democrats who clearly side with (and receive lots of money from) Big Oil. Sen. Mary Landrieu (LA), Sen. Jeff Bingaman (NM) and Rep. Dan Boren (OK) are examples. An argument based purely on a party split would need to be able to explain these cases as well.

  >Finding out the magnitude, in dollars, of the
  correlation doesn‚t change anything. It might as >well be reported as
  >„Big Oil Gives $XXX more to Republicans than to Democrats.‰ This
  >headline is less exciting, but it‚s the same thing.
  >Throwing in votes
  >just makes it sound more important, and it is >misleading because it
  >makes it sound like there is something new and >nonobvious to be learned.

  I would argue that it focuses the attention of the analysis on the actual documented behavior of the politicians, trying to steer the debate away from the quagmire of partisan infighting and back to crucial issues.

  >Let‚s look at what is being reported. I paraphrase from Follow the Oil
  >Money (sorry guys) [http://priceofoil.org/oilmoney_keyfindings]:

  >Why not just say „Republicans‰ took more >than 4.5 times more oil money
  >than „Democrats‰? (Well, the number may >change a bit of course, but
  >that‚s the idea.)

  Well one reason is I don’t think the causality is that straightforward, I certainly wouldn’t feel comfortable making that claim. Also, the current political climate is so partisan that framing the findings in terms of a party bias would immediately cause 50% of readers to reject the findings without even reading the article.

  >The votes have nothing to do with it unless it is
  >showed that the votes were something other than decided roughly on party
  >lines.

  Untangling causality is subtle and complicated. I’m sure you are aware that there are many possible explanations. For example, instead of looking at the party of the candidates, one might say that state of the politician is what is important: “Congress people from TX, LA, CO and other heavy oil producing states receive more money from Oil, and vote more in favor Oil”. Maybe not so surprising, but probably as least solid as the party relationship. (I have not run the numbers, but try sorting the tables on the Oilmoney vote tracker and take a look at what happens) Without a great deal of further analysis, I don’t think anyone pretends to understand the details of the causality. Hopefully more academics with the skills to set up statistical models with the necessary level of sophistication will get interested.

  In short, I think that if sites like MapLight and Oilmoney get people to ask the question “Why did my Rep votes that way, was it because of party allegiance or campaign contributions?” and provide some ways for them to look start looking at the data, then they are doing their job. Perhaps it could be done better, and it probably will be by some as-yet-unknown startup in the next few years.

  > It‚s another question entirely whether the money influenced the
  >votes, or whether votes influence future >contributions ˜ a question that
  > is unsolved.

  And probably unsolvable. Do I work because I was paid, or because I expect to get paid, or because I think the work is important and just needs to be done? All three are true.

  Anyway, thanks again for your comments. And once again, thank you for creating such a great data resource in GovTrack. Without the data, we couldn’t even be having this discussion.

  best,
  -skye

• Joshua Tauberer // Apr 1, 2008 at 8:37 am

  Skye, thanks for the link to the paper and your other comments. I’m not sure yet what my reaction to the paper is.

  But I don’t think my point came across entirely clearly. On the one hand the academics are trying to sort out correlation and causality. That’s not what I’m concerned with here, because as we all agree the analyses we can do ourselves patently don’t have anything to say about causality.

  The problem is that in this type of shallow analysis, there are a lot of things going on, and they’re all related — they’re all part of the same phenomenon. You have a bunch of variables to choose from in this phenomenon: votes, money, geography, ideology. But they’re four sides of the same coin. You can’t pick two, find a correlation, then pick two others, find another correlation, and say you’ve found two interesting facts. You can’t say one correlated pair is more interesting than another. It’s all the same thing: they all covary together.

  So my point is not about the numbers or the claim, but about the presentation. Why did FTOM *choose* to run with the headline about money and votes, and not a headline about money and party affiliation? The former is suggestive of something to do with influence that is simply not supported by the numbers reported. (You can see why it’s not supported by the numbers because a headline about the relationship between party and votes would *not* imply influence, and yet it’s an equally good way of interpreting the numbers.) That’s of course why the headline was chosen: suggesting influence is the point of the website, and the headline will draw more readers.

  If you want to get at influence, you have to do more - you have to put yourself in the shoes of the politician and see what might affect his behavior. Just because someone else gets more money than him from industry X doesn’t mean that’s going to affect his behavior, and not just because he might be the one moral member of Congress. The representative with the bigger wallet may have a bigger wallet because he’s been around longer, right? Or because he lives in the state where industry X is located. Or because industry X happens to be located in Los Angeles and the representative with the bigger wallet is a Democrat. Our poor representative knows these things as well as we do, and he knows that he can’t change his vote to get more money from industry X — the money in question has no influence on the poor representative.

  Some of these things are easy to check for, and that’s what you have to do before you start framing the correlation in terms of influence.

  Josh

Leave a Comment