Skip to main content

As part of our ongoing focus on interest group advertising and dark money in elections, we are pleased to welcome our first guest post by Conor Dowling (University of Mississippi) and Amber Wichowsky (Marquette University).


Total outside group spending in federal elections with no disclosure of donors saw a 60-fold increase between 2006 ($5.17 million) and 2012 ($310.8 million). This “dark money” can in large part be explained by increased spending by nondisclosing entities—particularly 501(c)(4) and 501(c)(6) groups that are not required to disclose the identities of their donors as long as their primary purpose is not “political.” In 2010, 501(c)(4) groups outspent Super PACs by a 3-2 margin. And though Super PACs and 527s are required by federal law to disclose the identities of their donors, many of these groups are able to hide the original source of their funding by attributing contributions to these tax-exempt groups rather than to individual donors, effectively creating a “shell game” that makes it more difficult to follow the money trail.


It is unclear whether this increase in “dark money” is a surprise to Justice Kennedy, who in writing to uphold disclaimer and disclosure laws argued: “The First Amendment protects political speech and disclosure permits citizens and shareholders to react to the speech of corporate entities in a proper way. This transparency enables the electorate to make informed decisions and give proper weight to different speakers and messages.” Or to current House Speaker John Boehner, who told reporters after the Court’s decision in Citizens United was announced: “I have always believed that sunshine was the best disinfectant…Let the American people decide how much money is enough…Sunshine really does work if you allow it to.” Since Citizens United, the Supreme Court has had several opportunities to strike down or weaken disclosure requirements, but it has not, giving confidence to advocates of transparency in campaign finance. Indeed, in the recent McCutcheon decision, Justice Scalia wrote that with “modern technology, disclosure now offers a particularly effective means of arming the voting public with information.”


In denouncing the increase in undisclosed campaign spending, critics of “dark money” argue that voters would evaluate campaign communications differently if they had more information about the messenger behind the message. But is this true? Are there advantages to anonymity? Do voters’ opinions shift in response to greater campaign finance transparency? Several recent experimental studies, including our own, suggest that the answers to these questions may be yes.


First, to the advantages of anonymity. Recent experimental studies manipulated a negative ad’s sponsorship to test the effectiveness of ads sponsored by unknown groups. (Research in this area tends to focus on negative advertising because the bulk of ads sponsored by groups are “negative.”) In one study, participants were randomly assigned to watch an ad that attacked a fictitious congressional candidate’s record on crime sponsored by a different fictitious candidate, the National Rifle Association (NRA), or a fictitious, unknown group. Although those already favorable to the NRA were more persuaded by the NRA-sponsored ad than the candidate-sponsored one, the unknown group ad was persuasive regardless of participants’ feelings about the NRA, suggesting participants found the unknown group ad more credible than the NRA-sponsored attack. Another study found a candidate was less likely to be penalized for an attack ad if it was sponsored by an unknown group. In our own work, we also find that candidates can escape voter backlash for “going negative” if the attack ad is sponsored by an unknown group. Further, when we showed our subjects a vague attack that did not include any partisan content, participants evaluated the candidates differently if they were told the sponsor was either the Republican or Democratic Party, but made no such distinction between unfamiliar conservative and liberal groups. Thus, despite the media scrutiny of these non-disclosing groups, it appears that many voters remain unaware of their ideological or policy positions.


Does campaign finance disclosure correct for these advantages of anonymity? Here the evidence is more suggestive. We have conducted two studies to test the effects of greater campaign finance transparency. In one study, we showed subjects an attack ad sponsored by an unknown group and then randomly varied the amount and format of information about the group’s donors. We found that the attack ad was less effective at moving participants’ opinions of the candidates when participants were provided additional information about donors, suggesting that voters may discount an attack ad when they have more information about the financial interests behind the message. But we also found some evidence that the form of disclosure may matter, too. In particular, the attack ad was less effective when participants were subsequently shown a list of the top five donors to the group, a treatment that we modeled after congressional legislation proposed back in 2010 as part of the Disclose Act. However, presenting subjects with a news article that also included this information produced no such effect.


More recently, we examined the potential effects of disclosure in non-federal elections. In this study, participants were asked to read a newspaper account of a fictional state legislative race and were then randomly assigned to one of seven conditions that varied the amount of information about the donors to an outside group. Subjects were either told that the outside group supported the Democratic candidate or the Republican candidate. We were particularly interested in whether information about out-of-state donors or donations from a wealthy individual are especially meaningful to voters, and whether campaign finance disclosure has a greater effect on evaluations of Republican or Democratic candidates. As the figure below shows, we found that candidate evaluations changed the most when subjects were told the outside group received the majority of its contributions from out-of-state donors (note the separation between the out-of-state D and out-of-state R bars). There was a smaller treatment effect for the conditions where subjects were told about a wealthy contributor to the outside group (note the smaller separation between the individual D and individual R bars). And we observed very little difference in candidate evaluations for our conditions in which subjects were simply given information about how much the outside group had spent, but were not provided any additional information about the group’s donors.


The figure displays averages (dots) with 95% confidence intervals for each treatment condition. For example, participants who were told out-of-state spending by outside groups benefited the Democratic candidate (Out-state D) were more likely to vote for the Republican candidate than those who were told out-of-state spending by outside groups benefited the Republican candidate (Out-state R).


These experimental studies suggest that outside groups have an advantage when they sponsor negative ads: given their unfamiliarity and innocuous-sounding names, few voters link their attacks back to the candidates, and in some situations, may even find their claims to be more credible. This research also provides suggestive evidence that under some conditions voters do care about where the money comes from, and use that information to make more informed judgments about candidates. What form campaign finance disclosure should take, however, remains an open question. Should individual donors to the outside group be identified during the ad? Or, might voters care more about aggregate data, such as how much particular industries and/or interests are spending to influence electoral outcomes? Future work in this area (to paraphrase Speaker Boehner) should be able to shed light on the best way to produce sunshine that will act as a disinfectant.