It’s NOT all about the Benjamins

by lee_drutman on October 19, 2009 · 10 comments

in Uncategorized

(this is post #2 in a series on money and influence in politics)

I like the Center for Responsive Politics. I really do. They make an incredible amount of data on lobbying and campaign contributions very easily available, which makes it much easier to be a political scientist studying this stuff.

They also make it really easy for journalists to add figures about how much such and such special interest contributed to such and such candidate, the implication being that campaign contributions are trying to buy votes.

I just did a Google News search on “Center for Responsive Politics.” Dozens of articles from today’s papers are using the Center’s data. Here’s one from today’s Kansas City Star: “In health care overhaul effort, millions are being spent to influence Congress.”

The money-buys-votes story is in many ways intuitively appealing because it sort of makes sense. Interest groups want outcomes and have money. Legislators can deliver outcomes and want money so they can get re-elected. Sounds like a good trade.

Plus: why would these groups be spending all this money if it weren’t having at least some impact? They’re not stupid, are they?

What’s interesting, however, is that political scientists have been trying for decades to show a systematic relationship between campaign contributions and roll call votes. But they can’t seem to find one.

Here is Steve Ansolabehere, John de Figueiredo, and James M. Synder Jr., from a 2003 review of almost 40 studies on the question:

Overall, PAC contributions show relatively few effects on voting behavior. In three out of four instances, campaign contributions had no statistically significant effects on legislation or had the ‘wrong’ sign – suggesting that more contributions lead to less support.

Also: A very worthwhile new book, Lobbying and Policy Change concludes that there is no systematic relationship between lobbying resources and the likelihood that a “side” would get what it wanted (after an exhaustive study).

But should we even expect there to be a systematic relationship between money and outcomes? I see six reasons why we shouldn’t (there are probably more):

1)There is often a lot of money given on multiple sides of an issue. So a lot of the influence may depend not on how much is given, but what the competition looks like.

2)Giving a lot of money might be a sign an industry is in trouble. If healthcare industry spending reaches record numbers this election cycle (which it may), this will be because the industry was on the defensive. So maybe the more money given, the less likelihood a side has of succeeding.

3)A lot of money is given to members of Congress who are already supportive. Consider this piece from Friday’s Atlanta Journal-Constitution, which reports on two conservative, anti-reform Georgia Representatives who are getting showered with healthcare industry money. But are they being bought?

4)Different companies have different campaign contribution strategies. In my interviews, I found that some company lobbyists loved having a PAC because they thought it got them access. But others hated it, because it meant they were always being harassed to contribute. Without a systematic strategy, it’s hard to prove a systematic result.

5)Any company with a PAC is lobbying on multiple issues in a given election cycle, and it’s hard to tie their campaign contributions to any single issue.

6)Campaign contributions are only one of many strategies that lobbyists employ.

Political scientists often conclude that because there is no systematic relationship between money and votes, special interests don’t have much influence and everything is A-OK with democracy.

But just because the tools of multivariate regression can’t detect a systematic relationship doesn’t mean that all this money is wasted. Again: are groups really that stupid? Do we think that, if only they had gotten themselves a subscription to the American Political Science Review, they could have saved themselves millions of dollars?

The key point here is the sixth point: Campaign contributions are only one of many strategies that lobbyists employ. In my interviews, I surveyed lobbyists on a bunch of tactics they use, asking them to rank them in importance from one to seven.

The results are here.

Campaign contributions ranked near the bottom. This actually feels about right to me. My best guess is that campaign contributions help to make access a little easier and are a good token of friendship. At the very least, it means lobbyists will get some face-time at a fundraiser.

But access does not guarantee results. It might get one in the door. But once in the door, there are a whole lot of steps between a handshake and a presidential signing. There are also several other ways to get in the door besides money.

The rest of this series is going to look beyond the money and fills in the blacks in the hand-wavey “and then lobbyists undermine reform!” story. Lobbyists do a lot more than orchestrate campaign contributions, and they work very, very hard to shape and pass legislation in many perfectly legitimate ways. But the results can still be troubling.

In my next post, I’m going to shift to the lobbyists’ perspective on things. I hope this will provide a more nuanced understanding of the process, and lead us towards some often-overlooked sources of influence and power.


Millard F. October 19, 2009 at 6:28 pm

Who in the world thinks that money is the only strategy for influence? Are any political scientists really that naive?

Maybe they are.

David Cass October 19, 2009 at 6:57 pm

I loved this post and I’m looking forward to the next.

I just wanted to add, though, that if we do prove one day that campaign contributions have no influencing effect, it doesn’t mean they shouldn’t be banned/outlawed/ended. Contributions have other effects, perhaps more serious, than influencing or quid-pro-quo corruption.

I know that I, personally, would prefer my Rep and Senators to be doing a great job informing themselves and representing me to the best of their abilities–not cold-calling potential donors from some Capitol Hill basement…

andrew b October 19, 2009 at 6:59 pm

Does the literature take into account both lobbying expenditures and campaign contributions or just the campaign contributions? Both sets of data are disclosed, although my understanding is that the lobbying data isn’t as complete since a lot of lobbying work doesn’t meet the disclosure threshold.

Anonymous October 19, 2009 at 7:17 pm

An alternative title for the post might be “It’s not all about the votes.” While I haven’t read the Ansolabehere et al review piece, I assume that they note there are multiple potential points in the policymaking process where lobbyists can attempt to influence policy outcomes. Indeed there is evidence (in the APSR no less) that there are circumstances where PAC money influences policy outcomes.

Erik October 19, 2009 at 7:37 pm

The above point is very important. If you want to assess the influence of money you need to look at the content of bills, gatekeeping, etcetera, not just at who votes how.

ZC October 19, 2009 at 9:44 pm

One thing to note about the Ansolabehere et al (2003) study, though, is while it draws on 40 studies, its conclusions are heavily weighted by one – Grenzke (1989). Ansolabehere and co. report 491 coefficients, but, out of 40 studies, 100 of those coefficients come from the Grenzke study alone.

Fortunately, that Grenzke (1989) survey still looks to me like a pretty solid piece of work. I’d be curious, though, what other campaign finance geeks think about it.

ZC October 19, 2009 at 9:47 pm

Just as a follow-up … if you drop that one study (Grenzke 1989) from the mix, the proper statement would be, “In two out of three instances, campaign contributions had no statistically significant effects on legislation …” It grows in other words from 1 in 4 significant to 1 in 3 significant.

Doug Hess October 20, 2009 at 8:50 pm

I wish I had time to read up on the research on this topic. There are so many cases that anybody that has worked on the Hill should be able to point to of money trading hands for favors (not just votes) that I am always surprised that people think this can be studied with data sets alone. I mean just how much change (and how often) would $$ need to lead to before you say it really happens? Presumably it doesn’t bring about lots of the most overt kinds of change but certain kinds of players still engage in it, if you’re hoping to stay in the game. You need good interviewing skills and the research tools of an old fashioned investigative reporter to get at this question is my suspicion, but I should read more about those quant. models.

Doug Hess October 20, 2009 at 8:56 pm

Hi, Lee: Thanks for posting that list. Why didn’t you ask about lobbyists organizing/advising clients regarding whot to give money to? I’m not sure what all is covered by “organizing a fundraiser”, but in my mind it’s only one kind of way lobbyists use money.

Lee Drutman October 21, 2009 at 11:57 am

A couple of comments in response:

1. “Does the literature take into account both lobbying expenditures and campaign contributions or just the campaign contributions?”

Most of the literature is on PAC contributions, which are tied to particular members. There is less literature on lobbying expenditures. It is true, the data is limited — a lot does not get reported. There is also no data on how groups allocate their lobbying resources across targets.

2. “An alternative title for the post might be “It’s not all about the votes.””

True, and I probably should have dwelled on this more. There are a lot of other outcomes besides votes that lobbyists care about. Much more important is what gets into the legislation in the first place.

3. “Why didn’t you ask about lobbyists organizing/advising clients regarding who to give money to?”

A great idea. I really should have done this. Next time.

Comments on this entry are closed.

Previous post:

Next post: