The elections of 2012 sent two important signals to students of American campaign finance. First, the system of limited donations to candidates and party committees faces a crisis moving forward, its efficacy and purpose challenged by huge investments from outside interests. Indeed, the conventional means of financing presidential and congressional elections was shaken to its core in 2012, with more spending from outside groups than in any prior election. When Politico reported in early November of 2012, for example, that the previous month featured $500 million in independent expenditures and electioneering communications from outside groups, careful observers surely took note of such unprecedented volume.1 That figure was roughly equal to the combined spending by outside groups in the entire election campaigns of 2010 and 2008.
The second lesson refers to the long debate in political science about whether parties or candidates should be the center of our electoral process. The debate is old, but it now risks irrelevancy as a collection of well-funded outside groups has asserted its role as the primary pivot point in competitive election campaigns. Moreover, many of these groups disclose little to nothing about their donor base. Parties, in comparison, increasingly languish in total spending and are in peril of becoming mere supporting characters in federal elections.
This paper uses data from the Wesleyan Media Project (for ad spending in 2006, 2010, and 2012) and the Wisconsin Advertising Project (for ads in 2000, 2002, 2004, and 2008) to compare the levels of ad spending from outside groups and traditional party organizations across seven elections.2 It also uses Federal Election Commission reports since 1980 to demonstrate the historic levels of such spending in 2012. The paper is organized as follows:
It begins with a brief review of recent changes in campaign finance laws that liberated interest groups to advocate directly for or against candidates without restriction.
Second, the paper discusses the levels of spending by outside groups in the presidential primaries of early 2012.
Third, it extends that look into the general election phase, finding very high levels of ad spending by outside groups on behalf of Mitt Romney and congressional candidates, levels that make the flirtation with 527s in prior years seem quaint.
Fourth, the paper speculates as to the effect of such spending on election outcomes. Mitt Romney’s loss and the failure of Senate Republicans to capture control of the chamber are perhaps sobering reminders of the limits of spending in elections. On the other hand, it seems unlikely that groups will disengage in future cycles. Nor is it obvious that the spending had no effect.
In late 2010, I compiled a review of interest group electioneering in the most recent midterm elections – the first since the Supreme Court’s January 2010 decision in Citizens United v. FEC 558 U.S. 310 (2010) – and found the share of spending on television ads to be high, but lower than many might have expected.3 For one, as a share of all ads, outside spending in House elections was lower in 2010 than in 2000.4 And interest group ads in Senate elections were stable in 2010 (15% of all ads that year) in comparison to their collective efforts in 2008 (14%), before the Court’s decision in Citizens United. Moving forward in the wake of the 2010 campaign, one could reasonably doubt the long-term damaging effect of Citizens United, though the paper did express concern at what unfettered spending from groups would mean for traditional party organizations.
As it turns out, 2010 was more epilogue than prologue, a final point on years past when outside groups invested mightily in close elections, but where doing so was still tainted by questions of its appropriateness. As Rick Hasen notes:
It is true that before Citizens United, people could spend unlimited sums on independent advertising directly supporting or opposing candidates. But that money had to be spent by the individual directly…. [T]o avoid having your name plastered across every ad, [you could] give to the 527s, which claimed they could take unlimited money from individuals (including, sometimes, corporate and labor union money) …. These organizations were somewhat successful, but a legal cloud always hung over them.5
To some extent, such concerns may have persisted into the fall of 2010. For example, 83 Super PACs registered with the Federal Election Commission before the November 2010 elections, despite their initial sanction in the middle part of 2010. Prior to the 2012 campaign, in contrast, nearly 1300 such PACs had registered. Moreover, there were a number of important legal developments in 2010 and 2011 that cleared away any legal clouds over what interest groups could do post-Citizens United. As a consequence, fund-raising by interest groups was swift and successful in 2012, and this spending exploded in unprecedented ways.
Developments in Campaign Finance between 2010 and 2012
The Supreme Court’s decision in late January of 2010 in Citizens United v. FEC is without question one of the most famous cases in recent Court memory. The specifics of the case have been covered in detail elsewhere,6 but the outcome is simply put: it overturned all restrictions on spending by outside groups for candidate advocacy that is uncoordinated with candidates. To that effect, unions and corporations were freed to spend general treasury funds on pro-candidate advocacy, something they were previously prevented from doing. It left in place restrictions on candidate and party contributions (these were not at issue in the case) and famously reinforced the value of disclosure as a check on the appearance of corruption.7 But in wiping away the restrictions on how corporations, unions, trade associations, and non-profits could fund uncoordinated pro-candidate advocacy, it also raised concerns about the influence of wealthy interests on the conduct and outcome of elections.
The case had an immediate impact in the federal courts, as it emboldened opponents of campaign finance restrictions to pursue vigorously the reversal of other limits on spending by individuals and organizations.8 In one case decided just 3 months after Citizens United, the US Court of Appeals for the District of Columbia in Speechnow.org v. FEC 599 F.3d 686 (D.C. Cir. 2010) ruled that for political action committees (PACs) that register with the Federal Election Commission but engage only in uncoordinated candidate advocacy (that is, contribute nothing to federal candidates directly), long-standing restrictions on their fund-raising (i.e., a $5000 limit on contributions to the PAC from individuals) were not consistent with the Supreme Court’s January 2010 ruling. The case explicitly freed individuals to pool unlimited resources into organizations that directly sought the election or defeat of one or more candidates.9
These “independent expenditure-only” PACs came to be known colloquially as Super PACs. The FEC clarified the rules for such PACs in four subsequent Advisory Opinions:
2010–09, which allowed the Club for Growth to maintain both a Super PAC and a candidate contribution PAC.10
2010–11, which allowed corporations and unions to contribute unlimited amounts to Super PACs.
2011–11, the Stephen Colbert opinion, which allowed Colbert to promote his Super PAC–Americans for a Better Tomorrow, Tomorrow–during his television show on Comedy Central.11
2011–12, which allowed federal candidates to appear at Super PAC events and solicit up to $5000 per individual for the PAC. In so doing, the opinion did not define such appearances as coordination between federal candidates and the PACs.
The legal sanction for Super PACs was an incredibly important development. It allowed organizations bankrolled by huge checks to advocate explicitly for the election or defeat of a specific candidate. Prior to the development of Super PACs, any such organization would have had to register with the Federal Election Commission but abide by significant fund-raising restrictions. As Rick Hasen’s quote from earlier noted, outside groups in prior elections could only avoid these regulations by organizing as a 527 (or a non-profit 501c4), but such groups were forced to claim publicly a primary interest in issue advocacy and/or voter education. Indeed, three 527s heavily involved in the 2004 presidential election were subsequently fined by the FEC for hewing too closely to direct candidate advocacy.12
Some have questioned whether Super PACs – and the rise in spending by outside groups more generally – were “caused” by Citizens United directly, or whether instead they were inevitable. Thus the Speechnow case was in the legal system well before Citizens United was decided, and it was likely that the outcome would have been the same regardless. Second, as Matt Bai has argued, the explosion of outside spending might be better tied to the unintended consequences of McCain-Feingold, which limited how parties could raise and spend funds.13 He writes: “the rising amount of outside money was probably bound to reach ever more staggering levels with or without Citizens United. The unintended consequence of McCain-Feingold was to begin a gradual migration of political might from inside the party structure to outside it.”
These are fair arguments, and there is some truth to them. To Bai’s point, and as explained below, the relative power of the traditional party organizations has declined since McCain-Feingold and may never recover. Moreover, had Speechnow sanctioned the creation of Super PACs funded by large checks from individuals, much of the spending witnessed in 2012 might have happened without the Court’s ruling in Citizens United. But Citizens United was also the moment when the paradigm shifted. Prior to the decision, the goal of limiting union and corporate money in elections was the prevailing legal precedent, and it was a time-tested compelling state interest. Moreover, without Citizens United it seems incredibly unlikely that corporate and union money could have funded Super PACs, or any other group engaging in direct candidate advocacy.
The debate over the causal effect directly attributable to Citizens United is probably beside the point anyway. The collective signal sent from the Supreme Court in recent years is that long-standing justifications for limiting the funding of uncoordinated candidate advocacy no longer hold water.14
One caveat: an old distinction in campaign finance laws – the express vs. issue-advocacy distinction – was still relevant to a class of outside groups involved in the 2010 and 2012 elections. 501c4 non-profit groups, in line with tax law, do not count elections as their “primary purpose”–a classification that eludes clear definition – but they are now allowed to spend on candidate advocacy efforts as a consequence of Citizens United. Most importantly, and unlike Super PACs that register with the FEC, 501c4s do not disclose donors publicly. This is an attractive option for political operatives who want to spend aggressively in elections and use anonymity as a tool to raise large contributions from wealthy investors. To avoid charges that these organizations violate the IRS’ primary purpose test, however, many 501c4s continue to sponsor ads that do not directly call for the election or defeat of federal candidates, though many of their ads do praise or criticize them. The IRS has yet to take an aggressive stand on counting these latter ads as electioneering, and many 501c4s continue to assert such ads as primarily voter education efforts.15
The Republican Presidential Primaries of 2012
These developments in 2010 and 2011 formed the backdrop to the struggle among Republicans for their party’s presidential nomination in 2012. The complete story of the GOP nomination is beyond the scope of this paper, but the levels of advertising by major players in the race are worth serious consideration. Since the elections of 2000, the Wisconsin Advertising Project and the Wesleyan Media Project have tracked local broadcast advertising on television. The bulk of research on ads using these data has focused on the congressional and presidential general elections, but the presidential primaries of 2012 stand out in comparison to the advertising in the primaries of 2000, 2004, and 2008.
For 2000, this paper counts ads aired before mid-March of the election year (when both Bill Bradley and John McCain dropped out of their respective nomination contests) as primary election ads. For 2004, it includes any date prior to Super Tuesday (March 3), when John Kerry secured enough delegates to be the Democratic nominee. For 2008, the primaries ended in early June when Hillary Clinton finally dropped out of the Democratic contest. (The GOP contest had ended far earlier in the year, however.) In 2012, Romney secured the nomination in mid-April with the departure of Rick Santorum from the contest.
Consider Figure 1, which shows the percentage of ads from outside groups that mention a presidential candidate and that aired at some point prior to the dates above. Because inclusion in this graph is determined by a cutoff date, there is the possibility that some interest groups ads are not ideally counted in the numerator. For example, MoveOn.org aired 17,000 ads between November 4, 2003, and the end of February 2004. The organization did not endorse a candidate in the Democratic nomination contest, but it did spend the early part of 2004 aggressively attacking President George Bush. These might reasonably be counted instead as general election ads, but they are included here as primary ads because of their early air date. Regardless, over-inclusion has the effect of amplifying the percentages in the Figure, and it is apparent that in the elections of 2000, 2004, and 2008, interest groups were not significant players. (Excluding the MoveOn.org ads from 2004 lowers the total from interest groups in the primary election to 3.5% of the ads aired.) All told, outside groups prior to this most recent election largely ignored the airwaves at times when the parties were struggling to pick a nominee.16
The nomination battle for the GOP in 2012, however, featured just short of 60% of all ads from outside groups. Here, too, the totals include some ads that are likely general election pro- or anti-Obama ads, but excluding these affects the proportion only slightly (as demonstrated below). Table 1 includes top spenders in the pre-April period of the presidential election, including candidates. (It excludes some party ads that aired.) The table shows that the Mitt Romney campaign aired more than twice the number of ads of the next most active candidate, Rick Perry. In total, GOP candidates aired about 64,000 ads by the end of April 2012, with almost half coming from Romney.
|Sponsor||First aired||Last aired||Ads aired|
|Restore Our Future, Inc.a (run by former Romney aides)||12/8/2011||4/24/2012||49,661|
|Winning Our Futurea (run by former Gingrich aides)||12/28/2011||3/13/2012||11,558|
|Red, White, And Blue Funda (Pro-Santorum)||12/15/2011||4/9/2012||11,471|
|Make Us Great Againa (run by former Perry Chief of Staff)||11/1/2011||1/20/2012||6465|
|Citizens For A Working America Paca (Pro-Romney)||12/24/2011||1/21/2012||1287|
|Our Destiny PACa (funded in part by Huntsman’s father)||11/15/2011||1/20/2012||810|
|Santa Rita Super PACa (founded by Ron Paul Fund-Raiser)||1/9/2012||1/21/2012||744|
|Leaders For Families Super PACa (Pro-Santorum)||12/29/2011||1/3/2012||88|
|GOP Super PAC total|
The top 20 outside groups, however, aired 122,000 ads, nearly double the total of the candidates. The table bolds the entries that one might consider general election ads – those that only promote or attack Barack Obama – and it also identifies any group that registered with the FEC as a Super PAC. The total ads aired for those groups, all of whom but Priorities USA were committed to a Republican primary candidate, was a little more than 82,000. As noted in the table, these Super PACs had very close relationships with many of the candidates running for the nomination. Former aides, for example, ran Romney’s Super PAC. Perry’s former chief of staff ran his Super PAC, and Jon Huntsman’s father was a major donor to his. If one excludes the bolded entries and counts only the PAC and candidate ads, the former’s ads account for 56% of the total share.
One could argue that the presence of these ads was one of the causes of the extended nomination battle. Each of the second tier of candidates had a Super PAC that supported him and that provided significant air support. Super PACs backing Newt Gingrich and Rick Santorum, for example, each aired nearly double the number of ads of each of the candidates. To be clear, Super PACs have the ability to advocate expressly for or against candidates because they operate independently of any candidate committees, but 2012 was the first election where so many organizations with such close, personal ties to the candidates were so heavily invested in federal elections. They formed, without exaggeration, unregulated parallel campaign structures. The pretense of promoting issues or educating voters was no longer a required hoop to jump through, and it proved rather easy to avoid the legal definition of coordination. Moving forward, no major candidate for president will forego sending some staff or confidants outside the official campaign structure to set up a Super PAC that solicits and spends millions of dollars promoting his or her candidacy. The consequence is a sea change in the content and structure of presidential elections.
The Presidential General Election and Congressional Races
The historic rate of spending by outside groups continued into the general election phase of the presidential election, as well as into competitive House and Senate elections. Again, the time series comparison is critical. First, consider presidential general elections since 2000. Table 2 shows the percentage of ads from outside groups and political party committees in each campaign since 2000. In the Gore-Bush contest, parties were still free to use soft money (which comprised unregulated donations ostensibly meant for party branding or state and local races) to advocate for their candidates through “issue advocacy” messages. Both major party nominees also opted into the public funding system for the post-convention period, which limited how much they could spend in the fall campaign.17 Parties, as a consequence, were critical to the efforts of both candidates.
|Pro-Dem, %||Pro-GOP, %||Total, %||Pro-Dem, %||Pro-GOP, %||Total, %|
|2000||Of all ads||26.53||39.24||65.77||6.80||2.07||8.87|
|2004||Of all ads||19.14||15.06||34.20||15.45||3.92||19.37|
|2008||Of all ads||1.00||22.62||23.62||2.14||1.95||4.09|
|2012||Of all ads||0.59||5.66||6.25||6.13||24.92||31.05|
Nearly two-thirds of all the ads aired in the 2000 general election were sponsored by parties or aired by parties in coordination with the candidates. The Democratic Party helped air over half of all the pro-Gore ads in the race, for example, while the GOP sponsored over three-quarters of Bush’s ads. Outside groups, however, through either regulated independent expenditures or unregulated “issue advocacy,” accounted for <9% of the total ads (though pro-Gore groups accounted for 14% of his ads). This, remember, is the election before the passage of McCain-Feingold, motivated by both the historic levels of party soft money but also by investments from outside groups. As Table 2 makes clear, the former was far more pervasive than the latter.18
One of the striking features of Table 2 is the precipitous decline of party spending since the 2000 election. In 2004, the Republican and Democratic parties combined to air just over a third of all of the ads, a significant decline from 4 years earlier but still a hefty investment. Indeed, both candidates received significant help from their respective national party committees that year. In 2008, only John McCain relied heavily on party spending, with Barack Obama being the first candidate to opt out of the general election public funding system. The total between both parties was just under a quarter of all of the ads aired. This was heavily weighted toward the Republican ticket, with more than half of the ads aired for McCain funded by independent or coordinated party expenditures. In contrast, only 2% of the pro-Obama ads were party funded.
By the most recent election, with both candidates opting out of public funding, parties sponsored only 6% of the ads in the general election. Even for Mitt Romney, the Republican Party was responsible for only 12% of his ads. In the 2012 presidential election the party committees were a far less prominent voice on television.
The story for interest groups is not so linear, but still revealing. In all four elections, only twice did groups account for a significant share of the ads. In 2004, pro-Kerry groups sponsored 15% of all the ads in the general election (this was primarily from groups like MoveOn.org and the Media Fund), which amounted to over 25% of all pro-Kerry ads. In 2012, conservative groups sponsored one in every four ads aired since the end of April and over half of the pro-Romney ads. These latter investments represent the single largest interest group presence in the previous four elections.
Table 3 lists the major sponsors of the ads in the 2012 presidential election. Obama and Romney top the list, though note the disparity between the two campaigns, in which Obama aired over 2.5 times as many ads as Romney. Of particular interest are the ads aired from Crossroads GPS, American Crossroads, and Restore Our Future. Combined, they aired over 210,000 ads, roughly equal in number to the Romney campaign. The RNC sponsored just over 68,000 ads in the election, about one-third of the total from the three top conservative groups.
|Sponsor||Est. Cost||Ads aired||Affiliation|
|Restore Our Future, Inc.a||$79,030,550||67,996||REP|
|Priorities USA Actiona||$40,540,970||63,376||DEM|
|Americans For Prosperity||$38,169,250||44,108||REP|
|Republican National Committee||$25,054,890||35,822||REP|
|RNC & Romney, Mitt||$24,789,190||33,466||REP|
|Americans For Job Security||$14,019,060||10,410||REP|
|American Future Fund||$11,427,530||10,391||REP|
|DNC & Obama, Barack||$15,309,090||7210||DEM|
|Republican Jewish Coalition||$5,408,920||3422||REP|
|Planned Parenthood Action Fund||$3,391,650||3159||DEM|
|Concerned Women For America||$4,400,360||3132||REP|
|NRA Political Victory Fund||$3,019,040||3108||REP|
|SEIU & Priorities USAa||$1,194,420||2010||DEM|
|Ending Spending Funda||$3,127,830||1446||REP|
|Priorities USA Actiona & LCV Victory Fund||$532,710||1080||DEM|
|Checks And Balances For Economic Growth||$896,290||981||REP|
|60 Plus Association||$1,209,250||873||REP|
|Special Ops OPSEC Education Fund||$553,640||850||REP|
|People For The American Way||$664,750||829||DEM|
These totals represent a shift in the focal point of presidential campaigns. Parties have become somewhat marginalized since the previous two presidential elections. This has occurred either because of aggressive fund-raising by the nominee directly (in the case of Obama) or because of the deeper pockets of outside groups (in the case of Romney). The former shift may not trouble many, as the candidate can perhaps more easily control the flow of money for the general election than a candidate/party joint effort. But the presence of outside groups in the Romney campaign is extraordinary. John Kerry was helped in significant ways by groups in 2004, as noted above, but these groups came nowhere close to out-advertising the nominee.
This story of increased investments from groups is apparent also in House and Senate elections. Here there are seven congressional cycles of ad data to compare, and Figure 2 shows the percentage of ads from outside groups and political parties in House races between 2000 and 2012. In this time-series, parties still out-advertise groups, though the trend lines are moving closer together.19 Before 2008, parties averaged about one in every four House ads. The mean investment is one in five ads in the elections of 2008 – 2012. In contrast, outside groups have increased their ad buys in every House election since 2004, and the share of such ads was greater in 2012 than the “issue advocacy” election of 2000.
In one sense, these totals seem a bit low. To assert 19% of ads from outside groups as astronomical or troublesome might not strike many as convincing. But these totals lump together all ads from the primary and general election phases of each campaign, and they include ads in races that are not competitive. There are a lot of ads in the denominator, then, that are from races that are simply irrelevant to the party committees and outside groups. The comparison across time is the real measure, and outside groups were more aggressive in 2012 than even the elections of 2010 (the increase was a 50% relative change). This stands out even more given the highly competitive nature of the 2010 campaign but the relative stasis of the 2012 one. (As a sense of scale, 76 seats were rated by CQ Weekly as too-close-to-call or leaning in one direction in 2010, compared to 51 seats in 2012.)
The change in spending totals is even starker in Senate races, which are shown in Figure 3. Party committees sponsored only about 10% of the ads aired in the Senate elections of 2012, compared to nearly 30% from outside groups. These totals essentially flip the investments of groups and parties from the 2000 and 2002 campaigns. Prior to 2008, outside groups invested only about 5% of ads in Senate races. That share of spending has increased by about 500% in the 2012 elections.
Top ad spenders among outside groups in congressional races in the fall campaign are listed in Table 4, along with the number of House and Senate races in which they were involved. Pro-GOP spending outran pro-Democratic spending by about 44%, and there was almost $177 million spent in total on over 190,000 ads. These ad totals actually understate the level of spending by truncating the list to ads aired after August. Before September, outside groups sponsored nearly 150,000 additional Senate ads and 40,000 House ads. These numbers point also to a real demarcation in the timing of ads by sponsors. Parties, for example, aired over 80% of their House and Senate ads in September and October.
|Sponsor||Est. cost||Ads||Affiliation||# House races||# Senate races|
|House Majority PACa||$22,348,510||19,458||DEM||44||0|
|US Chamber of Commerce||$16,204,390||15,291||REP||17||11|
|Americans for Tax Reform||$7,668,420||10,599||REP||10||0|
|Congressional Leadership Funda||$10,078,340||7112||REP||11||0|
|American Action Network||$7,474,920||5389||REP||7||0|
|Patriot Majority USA||$3,547,540||4480||DEM||4||5|
|Club for Growth||$5,553,860||4207||REP||1||3|
|YG Action Funda||$4,434,680||3316||REP||8||0|
|Now or Never PACa||$5,785,490||2945||REP||1||4|
|Vote Vets Action Fund||$1,947,390||2803||DEM||0||7|
|League of Conservation Voters||$2,791,460||2634||DEM||2||2|
|Center for Individual Freedom||$1,834,250||2576||REP||6||0|
|National Federation of Ind. Bus.||$1,986,330||2514||REP||3||2|
|End the Gridlocka||$1,289,470||2252||DEM||0||1|
|American Hospital Association||$1,011,130||2015||REP||7||0|
|Independence USA PACa||$6,761,020||2530||DEM/GOP||5||0|
|60 Plus Association||$1,000,540||1724||REP||3||4|
Worth noting are four particular groups: Majority PAC (20,168 ads in 13 Senates races), House Majority PAC (19,458 in 44 House races), Congressional Leadership Fund (7112 in 11 House races), and YG Action Fund (3316 in eight House races). The first two were organized on behalf of Democratic candidates, while the latter two advocated for Republican congressional candidates. All four, however, were run by former staff of the congressional leadership or by former members of Congress. They are all Super PACs who were able to take advantage of the FEC’s Advisory Opinion 2011–12, which, as described earlier, allowed members of Congress to appear at Super PAC events.20 This in turn incentivized congressional leaders to set up parallel Super PACs that advocated for congressional candidates. While technically uncoordinated with the traditional party organizations and congressional campaigns, they acted much like the Super PACs in the GOP presidential nomination but on behalf of a broader array of preferred candidates. Moreover, congressional incumbents were able to endorse the actions of the organization, appear at their fund-raisers, and solicit small donations on their behalf.
As a final demonstration of interest group ad levels, Table 5 shows the top 20 House and Senate races in terms of total ads aired in September and October. The final column shows the percentage of these ads sponsored by outside groups unaffiliated directly with the parties or candidates. Top interest groups efforts in House races ranged from just 14% in Arizona’s 1st congressional district to 38% in Texas’s 23rd. Top Senates races featured a range of investments by interest groups, from a low of 0 in the Massachusetts Senate race – the Brown/Warren Pact21 – to a high of 42% in the Indiana Senate race. Again, these totals are for ads aired only in the fall, so they understate any additional investments earlier in the campaign.
|House||Dem ads||GOP ads||Group, %|
A Brief Consideration of Additional Data
The trends discussed thus far have focused on television ads alone. This is useful, of course, but it is instructive to consider also the FEC reports, which aggregate totals for spending by groups both on and off the air, including print media and direct mail, online ads, and peer-to-peer mobilization. Indeed, it is here that the levels of spending from outside groups seem particularly striking. Figure 4 first shows the total in reported independent expenditures by all groups, excluding political parties, between 1980 and 2010. Independent expenditures are any public communications that urge voters to cast a ballot for or against a candidate. They include, in the legal parlance, “express advocacy.”
The totals in Figure 4 cover all reported expenditures by regulated PACs, as well as 501c4 and Super PACs (the former accounted for large shares in 2008 and 2010). These totals (unadjusted for inflation) come to $640 million. Next, the figure shows the amount of reported electioneering communications since the FEC started collecting the data in 2004. These are ads aired close to an election (60 days before the general and 30 days before a primary) on broadcast television or on the radio that feature a federal candidate but do not urge viewers to vote for or against a candidate. These ads totaled $258 million between 2004 and 2010. Combined, the level of outside spending reported to the FEC between 1980 and 2010 was just shy of $900 million.22
In contrast, the level of spending by outside groups in the entirety of 2012 was $1 billion, primary and general election campaigns inclusive. As such, the 2012 campaign featured more pro-candidate spending by outside groups than in 30 combined years of spending.23
Whither the Parties?
One of the potential implications of such intense levels of spending in 2012 is the displacement of political parties as important players in federal elections. In raw totals, parties are still raising and spending plenty. All party committees in 2012 spent over $300 million on coordinated and independent expenditures, which was pretty close to their historic high of $330 million in 2004 and 2008. Nonetheless, such spending in 2012 paled in comparison to the $1 billion from outside groups. Figure 5 examines the ratio of party spending to group spending in federal races for all of the available data back to 1980. The data underlying the Figure are reported in the Appendix. The totals include independent expenditures by outside groups and electioneering communications, as well as estimated “issue advocacy” efforts between 1998 and 2002. (The source of these latter estimates is noted in the footnote of the Appendix.) Party electioneering includes independent and coordinated expenditure totals, as well as soft money disbursements from party committees between 1992 and 2002.
The graph includes soft money totals in two ways. One calculation includes only soft money disbursements from the congressional committees (i.e., DCCC, DSCC, NRCC, and NRSC). This avoids counting any soft money transferred from the national committees to state and local parties meant explicitly for non-federal races. FEC summary reports make this a bit difficult to discern. The second total (for presidential election years only) includes reported soft money disbursements from the DNC and RNC, though it excludes soft money transfers between these committee and their congressional counterparts.24
For the mid-term elections of 1982 and 1986, political party committees more than doubled the spending of interest groups, and they were roughly equal to interest groups in total expenditures in 1980 and 1984. In truth, these totals do not include much money (see the Appendix), as PACs spent only about $75 million in independent expenditures in the 1980s, and parties were prohibited from independently advocating for federal candidates, which limited their participation to capped coordinated expenditures.25 Still, parties out-paced groups in total dollars spent in this period.
Moving into the 1990s, however, parties increased their use of soft money accounts for pro-candidate advertising – which was still ostensibly considered “issue advocacy” and/or party-building under existing interpretations of federal campaign finance laws – but interest groups were not able to keep pace in the rise of party spending. This was despite increased levels of “issue advocacy” by interest groups in the elections of 1996, 1998, and 2000. Indeed, parties dramatically outspent interest groups in the elections between 1992 and 2002.
After 2002, however, in the elections of 2004–2012, the party-to-interest-group ratio plummets unabated, such that parties and groups were roughly equal spenders in 2008 and 2010–something not witnessed since the mid-term elections of 1982–and were outspent heavily in 2012. The decline in the ratio post-2002 is consistent at least with Matt Bai’s argument (noted earlier in the paper) that one unintended consequence of McCain-Feingold was the transfer of large contributions from party soft money accounts to outside organizations. Parties have struggled mightily in recent elections to keep pace with this growth in outside spending.
What do all of these developments mean? There is growing body of scholarship in political science that re-conceptualizes the party as something beyond the traditional party organizations.26 As such, the shift in relative spending from party committees to outside groups may not trouble close observers of campaign finance. The network of outside groups acts, in essence, as a series “shadow” party committees. Moreover, these groups have highly partisan goals. There is almost no group that actively advocates for candidates in both parties, for example, although some do find occasional friends to support on the other side of the aisle. The party-affiliated Super PACs like Majority PAC and House Majority PAC in particular are clearly operating with the same goals as the party committees.
Beyond that, the network of interest-group participants can all coordinate with one another, sharing polling data, strategy, and expertise. Some examples on the left include the Democracy Alliance and America Votes. And on the right, the annual conference held by billionaires Charles and David Koch plots strategy with a collection of conservative organizations.27 What this generally implies in contemporary American politics is that “the party” is less the hierarchical structure one imagines with the DNC and RNC at the top, and more an amalgamation of players with shared electoral goals and ideological dispositions.
Whether this is good or bad for parties is beyond the scope of this paper, but it is worth noting that the traditional party committees are becoming the odd men out in this arrangement. In fact, they are the only network participants who are hampered by fund-raising restrictions. They cannot coordinate with interest groups to share information and strategy during the heat of an election. When the ideological leaders of these groups assemble to plot long-term strategies and fund-raising, the party committees are generally excluded or remain bit players. This places the party committees in the unenviable position of “going it alone” in the planning and implementation of their candidate advocacy efforts while a deep network of affiliated outside groups, many with no requirement to disclose donors, work together to enhance the efficiency of spending.28
A Discussion of Impact
In the immediate aftermath of the 2012 election, several news reports posed the question of whether the levels of outside spending in the presidential and congressional elections made any difference.29 The outcome of the election awarded President Obama a second term, for example, despite the overwhelming levels of spending on behalf of Mitt Romney from outside groups. Moreover, eight House seats shifted to the Democrats, and Republicans failed to capture the Senate despite having a serious chance to pick up seats in Wisconsin, Virginia, North Dakota, Florida, Ohio, and Missouri. The year 2012 turned out to be a status quo election that happened alongside the most spending by outside groups in modern times. Did the spending from outside groups simply fall on deaf ears, then, or potentially even backfire?
This is surely one possibility, though the evidence aggregated to such a level does not refute the possibility of a causal effect. For one, a long-standing argument in political science is that campaigns, especially presidential ones, do little to change minds, but instead remind voters of their pre-existing dispositions towards candidates and the party platforms.30 That is, campaign outcomes are quite predictable, but it takes the long road of a presidential campaign to bring voters in line with the prediction.
For 2012, most forecasts of the outcome predicted a very slight Obama win in the popular vote.31 But for most of the campaign, the president out-advertised Governor Romney significantly. Consider Figure 6, which plots all Obama ads against all Romney and RNC ads in all 210 media markets between April 25 and Election Day. The plot shows significant Obama advantages in ads aired. Indeed, for every 100 ads aired in a market by Obama, the Romney campaign mustered only 53.
These huge advantages in ads could quite easily imbalance the election away from the forecast and award the president re-election by a somewhat confirmable margin. Think of it this way. By casting Romney is such a negative light and without an equally forceful counter-narrative, many voters could have been persuaded to ignore the generally bad economy and rising federal debt, in favor of a more likeable President Obama. After all, Obama’s ads did little to remind voters of the rising debt or high unemployment. It is generally in the presence of information imbalance that ads move the most votes. However, the figure excludes ads from outside groups, who – as was established earlier – sponsored more than half of all the pro-Romney ads in the general election. Figure 7 puts Democratic and Republican interest group ads back into the totals.32
The resulting plot still shows a systematic Obama advantage, but a significantly smaller one. Now, for every 100 Obama ads, voters saw 82 Romney ads. There are even a number of markets where the ad totals on behalf of Romney outnumbered the total on behalf of the president. As a consequence, it is possible that outside groups provided the boost necessary to retain or win some Romney votes that would have otherwise been moved by the great imbalance in Obama ads. Indeed, one plausible alternative is that outside groups did not advertise enough, at least in numbers necessary to give Romney an imbalance in key markets.
Such an assertion is mere speculation, however, and demands some empirical evidence. Such evidence could come in many forms. One ideal test is experimental. Are ads from outside groups less effective overall, suggesting that Romney’s heavy reliance on them was a strategic mistake? A number of studies conducted prior to the 2012 campaign explored such sponsorship effects in the context of fictional campaigns and generally found that interest group ads can be quite effective, perhaps more so than candidate-sponsored ads.33 This does not imply that outside groups backing Romney saw greater gains from their ads than the candidate’s own advertising, but it does make it harder to accept the counter-hypothesis that outside groups were ineffective across the board.
Another approach is observational and compares vote returns with ad buys across the 210 markets in the continental US. Franz and Ridout did this for the 2004 and 2008 elections and located significant effects in both years, with larger effects in 2008.34 Their models leveraged the fact that media markets do not line up with state boundaries to compare counties in non-battleground states exposed to ads from battleground markets (i.e., voters in Massachusetts seeing ads in the Boston market meant for voters in New Hampshire) with other voters in non-battleground states exposed to no advertising.35 The advantage of the approach is a natural experiment where ads are assigned in essentially random fashion to non-battleground voters in various levels, and where no other confounding campaign activity – beyond media coverage of the candidates – is present.36
I take this approach here to examine ad effects in the 2012 presidential election, and I show also the results from identical models using vote returns and ad buys in 2004 and 2008.37 This allows for a comparison of ad effect sizes across all three elections. The unit of analysis is the county in non-battleground states, and I control for the vote in the previous presidential election along with median income in the county and a host of demographic variables (percent male, white, African American, Asian, and Hispanic, as well as percent of population over 65 and younger than 25). I also include state-level fixed effects. The dependent variable is the Democratic nominee’s improvement in the county compared to the previous election. The key independent variable is the Democratic advantage in ads aired at three time points: the full general election, the fall campaign, and ads aired in October. (This allows for a test of whether ads airing late in the campaign have stronger impacts.) These include ads aired by all sponsors collapsed into a pro-Democratic minus pro-Republican measure. Because ads are purchased for local broadcast at the market level, counties clustered in the same media market have the same value on the independent variable.
The results are reported in Table 6. I show only the coefficient estimates for the ad advantage measures (full model results are available on request), but I show them in a few different ways. First, the effect for 1000 ads is the coefficient estimate of the ad advantage measure. In all instances, the effect is statistically significant. Second, I show the effect of a two-standard-deviation shift in the ads in either the Democratic or Republican direction. Third, I show the effect of flipping the market with the highest Republican ad advantage to one with the greatest Democratic advantage, holding all other variables constant.
|Effect of 1000 ads on Democratic county votesa|
|Effect of 2SD shift in ads|
|Effect across full range of measure|
The results suggest that ads were significantly related to county-level vote share, but at a lower rate than in 2004 or 2008. In 2008, for example, a 1000-ad shift in October rewards the party of the sponsor about 0.9% of the vote. In 2012, a similar shift brings only 0.37% of the vote. If anything, 2012 looks a bit more like 2004 in terms of the estimated ad effects. This is not unexpected, of course, given that each election featured a highly recognizable incumbent president, limiting the room for movement in voter perceptions. In that regard, the efforts of pro-Romney interest groups were a priori facing a challenge in persuasion. Still, when these groups were able to out-advertise Obama, or cut significantly into his ad totals in the market, the effort was rewarded with votes.
This analysis does not allow, however, for a specific test of the effect of outside group ads. This is because ad totals at the market-level are so strongly correlated with one another. That is, Obama’s ads are highest in the same places that Romney’s are, and because interest groups were such a critical component of his overall totals, interest group ads at the market-level are also strongly correlated with Obama’s and Romney’s. All one can do with these results is make the reasonable inference that outside groups contributed to these noted ad effects. To the extent pro-Romney groups closed the gap with Obama in total ads aired, then, it seems they helped cut into the president’s vote share.38
A Discussion of Implications on American Elections
What happens moving forward? For one, the prevalence of outside spending might be self-correcting if donors come to believe (the above investigation notwithstanding) that the investments paid little dividends. As with liberal funders of MoveOn.org and the Media Fund in 2004, who famously moved their resources after the election to building a network of outside groups focused on voter outreach and mobilization,39 so too may conservative groups invest in a different type of election infrastructure. This is all the more likely, given that outside groups pay higher costs for their ads than candidates do. Federal candidates get the advantage of the lowest unit rate on the purchase of broadcast ads, meaning Obama’s dollars went farther than American Crossroad’s or Restore Our Future’s.40
Indeed, the market created with the purchase of political advertising in many battleground states (which also featured competitive House and Senate races in many cases) raised the total cost of advertising spots, which disproportionately affected interest groups unable to take advantage of mandated lower rates for candidates. Any collection of strategists moving forward will surely argue for a much more efficient allocation of resources off television, either online or through peer-to-peer contacts. It is unlikely, then, that the spigot will turn off, and more likely that the money will find other routes designed to maximize effectiveness.
So what does this portend for candidates and parties moving forward? In the short run, they need help. The existing structure of rigid contribution limits acts as a handcuff. Parties are increasingly less relevant to the deployment of resources in competitive campaigns – and are unable to play the lead in the plotting of strategy between elections – and candidates are coming to understand that outside groups loom large in any competitive contest. On that point, the mere threat of outside spending in primary and general elections may force candidates to devote more time to raising money and even less time on policy-making or constituency outreach.
Of course, the ability of outside groups to challenge incumbents and bolster the prospects of long-shot challengers (as was the case with many Tea Party candidates in 2010) may be a positive of the current system, but even here such challengers come largely to depend on the hope that an organization bankrolled by wealthy investors comes to their defense. Indeed, such candidates cannot solicit such help, nor plan for it. All told, in a regulatory framework with no restrictions on outside independent spending, might it make sense to relax or rework the rules for candidate and party committees?
The answer is yes, but the solutions run the gamut. For progressives, a matching program on a similar scale to one for New York City elections is increasingly popular.41 The plan encourages candidates to raise small contributions and facilitates fund-raising by making each dollar translate into $4 or $6 (or whatever the match is set at). It also gains broader support politically when it provides a “floor” that bolsters challengers but does not establish a “ceiling” for total spending, the latter of which tends to advantage incumbents who have greater name recognition. On the other hand, the plan also requires the commitment of tax dollars, which is controversial, and it raises sensitive questions over how to set thresholds for qualification while avoiding messy bureaucratic structures in the plan’s implementation.42
Another alternative for candidates or parties is to raise significantly any contribution limits. This is in essence a move towards de-regulation of the larger system. While this might encourage donors to invest directly with candidates, and create more accountability with voters – who could punish candidates backed by large donors – it also amounts to solving a drug problem by distributing more drugs. The appearance of impropriety that concerns many in the current system would only be amplified.43 If the answer is putting the power to direct campaigns back in the hands of candidates or party committees, this solution (or a variant of it) might be more attractive, however. At the very least, raising contribution limits to parties and allowing parties and candidates to work together with fewer restrictions (coordination is strictly capped currently) is a moderate reform that is probably worth adopting.
This particular point is worth some further discussion. One of the major threats in the current system is the ability of outside groups to direct the narrative in a campaign. Because voters were exposed to more pro-Romney ads from groups than they were from Romney himself, this is no small problem. Matt Bai, in the same article cited earlier, makes this case persuasively:
Back in the days of soft money, a [presidential] candidate had ownership of his party’s national apparatus and the accusations it hurled on prime-time TV. He was responsible for the integrity of his argument, and his advisers ultimately controlled it. What the reform-minded architects of McCain-Feingold inadvertently unleashed, what Citizens United intensified but by no means created, is a world in which a big part of the money in a presidential campaign is spent by political entrepreneurs and strategists who are unanswerable to any institution. Candidates and parties who become the vehicles of angry outsiders, as Mitt Romney is now, don’t really have control of their own campaigns anymore; to a large extent, they are the instruments of volatile forces beyond their own reckoning.
It seems almost impossible to imagine any functional democratic theory of campaigns that would empower outside groups while also ensnaring candidates and parties in a strictly limited fund-raising paradigm.
If higher donor limits and easier coordination between parties and candidates is part of the solution, any reform that exacerbates the cost for groups to advertise on television might also be attractive. For example, consider a more aggressive lowest-unit-charge for candidates or even free television time. Both would free up resources for other forms of campaigning, while still allowing an affordable means of countering any messages from groups that do advertise.
Moreover, new disclosure laws should be central to any reform efforts. Super PACs disclose donations to the FEC already, but 501c4 groups do not.44 Indeed, it does not seem logical that shadow groups can be dominant players in competitive elections, often dominant enough to drive the issue debates in such contests. One need not embrace the traditional party organizations to understand that, comparatively, they are better for democracy than groups like American Crossroads or Priorities USA. Traditional party organizations have long-standing reputations, are over 150 years old, and can claim the vast majority of voters as supporters. They also have the obvious goal of winning control of government in line with a transparent political philosophy, and they disclose all donations. In contrast, what exactly do groups like the Center for Individual Freedom or Crossroads GPS want?
Does disclosure work, however? Does it have the effect of limiting the influence of large donors, for example? This is a common counter-argument to calls for more disclosure.45 The empirics as they exist now may not matter, however. Consider certain developments in technology, such as the SuperPacApp and AdHawkApp, which allow voters to let their smart-phones access campaign finance data for different ad sponsors. Point your phone at the television during an ad, and the app will identify any donor information available with the Federal Election Commission. This gives voters more power to access information that is otherwise somewhat complicated to find and understand. We simply do not know moving forward how easy access to disclosure will empower citizens, and so it seems worth the experiment.
A set of simplistic reforms, then, starts with empowering citizens with more information to make good decisions and then turns that responsibility over to citizens. If voters are not interested in who sponsors ads or funds a Super PAC, there may be little need to demand a bigger change in campaign finance laws. If voters come to parse the information in a campaign – throwing out messages from groups they do not recognize – the system might end up returning to one dominated by candidate and party messages. One goal of policy-makers and reformers should be to urge voters to do that parsing. Second, any reform should give candidates and parties more tools to raise money that can compete with the volume of outside spending. Candidates and parties should be at the center of the campaign dialogue, not outside groups.
All of this is ultimately a question of democratic theory. Who should control the messaging in our campaigns? Interested citizens that aggregate into active organizations should not be discounted, of course, but campaigns dominated by wealthy investors with no disclosure do not seem normatively ideal, either. Our elections currently face a crisis over enhancing and preserving the free speech rights of citizens while also allowing candidates and party organizations the space they need to make appeals to voters that break through the messaging haze. No proposed solution is easy to implement, and there is no cure-all. But not addressing it seriously is a cliff not of the fiscal sort, but of the democratic one.
|PAC independent expenditures||Party coord. exp.||Party indep. exp.||Elec. comm||Party soft money disbursements (congressional comm. Only)||Party soft money disbursements||IG “Issue ads"|
1“Outside groups spend over $500 million in October,” by David Leventhal. Politico, 11/2/12: http://www.politico.com/news/stories/1112/83218.html (Accessed November 4, 2012).
2Ad data from both projects count ads aired on local broadcast stations (NBC, CBS, ABC, and Fox affiliates) and national cable. Totals exclude ads aired on local cable stations. The number of markets included in the data varies in each year but includes all 210 markets since 2008 (and the 2004 presidential election).
3Michael M. Franz. 2011. “The Citizens United election? Or same as it ever was?” The Forum: A Journal of Applied Research in Contemporary Politics Vol. 8, Issue 4.
4On the other hand, interest groups more than doubled their share of ads compared to 2008 (from 6% to 13%).
5“The Numbers Don’t Lie,” Slate, by Rick Hasen, 3/9/12: http://www.slate.com/articles/news_and_politics/politics/2012/03/the_supreme_court_s_citizens_united_decision_has_led_to_an_explosion_of_campaign_spending_.html (Accessed 11/9/12).
6For example, Rick Hasen. 2011. “Citizens United and the illusion of coherence,” Michigan Law Review. 109: 581–623.
7Justice Anthony Kennedy wrote the majority opinion and said regarding disclosure: “With the advent of the Internet, prompt disclosure of expenditures can provide shareholders and citizens with the information needed to hold corporations and elected officials accountable for their positions and supporters. Shareholders can determine whether their corporation’s political speech advances the corporation’s interest in making profits, and citizens can see whether elected officials are ‘in the pocket’ of so-called moneyed interests.”
8The review below, for the sake of brevity, leaves out a longer discussion of ancillary challenges to campaign finance laws in the court system. The Supreme Court in 2011, for example, overturned part of Arizona’s clean election laws (McComish v. Bennett), which in turn invalidated related provisions in other states. The Court relied on similar logic as in Citizens United. There are a host of other challenges in the court system that concern major pieces of established campaign finance, from candidate contribution limits to party fund-raising restrictions.
9Presumably the case also freed corporations and unions to contribute unlimited amounts to these PACs, though the case concerned only contributions from individuals.
10See also Carey v. Federal Election Commission 791 F. Supp. 2d 121 (D.D.C. 2011), which allowed PACs unconnected to unions or corporations to run “hybrid” Super PACs that make candidate contributions (with regulated money) and independent expenditures (with unregulated money). The FEC has deadlocked (Advisory Opinion 2012–01) on whether to allow connected PACs (those run by unions and corporations) the same opportunity.
11See R. Sam Garrett. 2012. “Seriously funny: understanding campaign finance policy through the Colbert Super PAC,” St. Louis University Law Journal 56 (3): 711–723.
12These fines were small in comparison to the total money raised and spent by the three organizations, compelling many to argue that groups could internalize the fines as “the cost of doing business.” If so, the threat of an FEC investigation is not really a deterrent. Still, such negative publicity might have deterred donors from actively giving, which is one argument for the lower levels of outside spending in the congressional elections of 2006. For the facts surrounding the fines, see: http://www.fec.gov/press/press2006/20061213murs.html (Accessed 12/1/12).
13How did political money get this loud? by Matt Bai, New York Times Magazine, 7/22/2012, p.MM14.
14Bai’s piece misses a lot of nuance about the elections following McCain-Feingold. For one, the 2006 elections did not see a huge amount of outside spending (as will be demonstrated below). This is probably because the rules at the time were effective at restraining them, especially in the immediate aftermath of such public discussion of 527s in the 2004 election. If the entire jump in outside spending was attributable to McCain-Feingold, 2006 stands out as a glaring anomaly. Second, the Supreme Court’s 2007 ruling in Wisconsin Right to Life v. FEC 551 US 449 (2007) loosened some of these tight funding restrictions for groups and more importantly signaled that the Court was willing to rethink standing precedent.
15These ads were at issue in Van Hollen v. FEC earlier in 2012. Television or radio ads that mention or depict a candidate for federal office and that air close to elections are called “electioneering communications.” Groups are required to report these expenditures to the FEC, but they can easily avoid reporting the source of the funding. The lax regulation on donor disclosure was overturned by the District Court for the District of Columbia in March of 2012, but that ruling was overturned by the D.C. Appeals Court in September 2012.
16This is not to suggest that interest groups committed no resources in these campaigns. Unions, for example, were split over the appropriate Democratic nominee in 2004 and 2008, and lots of groups in the party’s network of allied organizations preferred one candidate. These groups invested in the primary race in other ways. The important point with Figure 1 is the clear change in the level of investment from groups on television ads.
17Because the time periods for the start of the general election include some months before the party conventions, advertising by the candidates as described below includes funds raised for their primary campaigns. That is of little consequence for the calculations, since all of the ads either promoted their own candidacy or attacked the other party nominee.
18It is important to note that nearly all of the party ads were funded with soft money. According to FEC reports, party committees reported only $20,000 in independent expenditures for the presidential race in 2000.
19In Figures 2 and and 3, I calculate the percentage of ads aired in all available markets within the year. Sometimes that is only the top 75 media markets (2000) or top 100 markets (2002 – 2006), as opposed to all 210 media markets (2008 – 2012). However, if I restrict the analysis in all years to the top 75 markets, the reported percentages in later years change only slightly. Also, totals count all ads aired in the election year, inclusive of all primaries and the general election.
20Another AO request (2011 – 21) initiated by Senator Mike Lee (R-UT) asked if Members of Congress could directly run a Super PAC provided the funds did not support the Member’s own candidacy. The FEC did not approve of this request.
21This was an agreement between the two candidates to donate campaign funds to charity if any outside organization sponsored advertisements in the race. The pact was successful in keeping ads off the air, but it did not prevent interest groups from finding loopholes in the agreement. See: http://www.bloomberg.com/news/2012–10–12/brown-warren-pact-undone-as-outside-groups-see-loopholes.html (Accessed 12/13/12).
22The inflation adjust total in 2011 dollars is $1.07 billion.
23These figures for all years exclude what is not reported to the FEC. This includes “issue advocacy” spending before 2004, and ads that feature candidates after 2004 that were publicly distributed outside the electioneering communication windows, as well as print and broadcast ads that do not meet the definition of reportable communications. One might question the inclusiveness of these totals, then. It is true that in years prior to 2012 there may have been significant sums of non-reportable electioneering that keep the totals down. But the same problem exists in 2012, and there is no reason to think that such expenditures were smaller than in previous elections.
24The data are accessible at http://www.fec.gov/press/press2003/20030320party/20030103party.html (Accessed 12/1/12).
25Campaign finance laws at the time prohibited parties from spending independently of candidates. This was overturned by the Supreme Court in 1996 in Colorado Republican Federal Campaign Committee v. FEC 518 US 604 (1996). Money spent in coordination with candidates, however, was and is capped. (Capped rates vary between House and Senate candidates, the latter of which depends on the number of voters in the state.)
26Gregory Koger, Seth E. Masket, and Hans Noel. 2009. “Partisan webs: information exchange and party networks.” British Journal of Political Science 39: 633–653.
27“Koch brothers postpone post election meeting,” Paige Lavender, The Huffington Post: http://www.huffingtonpost.com/2012/12/11/koch-brothers-_n_2277700.html (Accessed 12/14/12).
28Though how much data and strategy-sharing actually happens is not really known. However, anecdotal evidence suggests that it happens and is reasonably sophisticated. And it should be noted that parties are not shut out of the data-sharing that can happen. For example, a liberal organization known as the Atlas Project collects and shares voter-contacting information with subscribers. Subscribers upload their own contacting and outreach efforts to Atlas, and subscribers can access the data (for a fee) to see what other groups have done and are doing in campaigns up and down the ballot. Parties can access the data, as well, and this serves as a sort of go-around for limits of coordination between unregulated groups and the party committees. Still, parties cannot play the lead in such efforts and are essentially reduced to support players in the network of allied organizations.
29See, for example, Ken Vogel’s report, “The billion-dollar bust?” Politco, 11/7/12. http://www.politico.com/news/stories/1112/83534.html (Accessed 12/12/12).
30Andrew Gelman and Gary King. 1993. “Why are American presidential election campaign polls so variable when votes are so predictable”? British Journal of Political Science 23: 409–451.
31For a review of the major political science predictions, see http://www.washingtonpost.com/blogs/wonkblog/wp/2012/09/14/obama-leads-in-at-least-eight-of-13-election-forecasts/ (Accessed 12/12/12).
32Replicated plots for ads aired only in October show the same story.
33Deborah Jordan Brooks with Michael Murov. 2012. “Assessing accountability in a post Citizens United Era: the effects of attack ad sponsorship by unknown indepentent groups.” American Politics Research 40 (3): 383–418; and Christopher Weber, Johanna Dunaway, and Tyler Johnson 2012. It’s all in the name: source cue ambiguity and the persuasive appeal of campaign ads, Political Behavior 34: 561–584.
34Michael Franz and Travis Ridout. 2010. Political advertising and persuasion in the 2004 and 2008 presidential election, American Politics Research 38 (2): 303–329.
35The creative research design was borrowed from Gregory A. Huber and Kevin Arceneaux. 2007. “Identifying the persuasive effects of presidential advertising. American Journal of Political Science 51 (4): 957–977.
36One caveat concerns the identification of battleground states. This is not especially controversial in most instances, but candidates often target a second tier of states, making their exclusion as battlegrounds not exactly a random assignment of ads to counties. In all instances, I played with inclusive and exclusive lists of battleground states, and effects are quite robust. My list of battleground states in 2012 is: CO, IA, WI, OH, NH, VA, NC, and FL.
37This analysis excludes congressional races, mostly because county-level data are not yet available. For the presidential investigation below, the data are current as of early December and are subject to some change as vote returns are adjusted in many places. The county-level data were purchased from Dave Leip’s Election Atlas (http://uselectionatlas.org/). I should also note that the models do not include measures of candidate visits to different counties in the fall. This is partly because the data collection for such visits in 2012 is ongoing. However, their exclusion from the models in 2004 and 2008 do not change the coefficient effects for ads in any significant way.
38There is one caveat to that, however. It may not be entirely the case that the smaller effect sizes are because of the presence of an incumbent on the ballot. Might it be that ads overall were simply less effective this year? The measure treats 1000 more ads–up or down the measure–as the same, regardless of their varying levels of persuasiveness. If the measure lumps a bunch of ineffective ads in with more highly effective ones, the overall impact may be to weaken the coefficient size. Such a possibility in this election demands more investigation, however.
39See, for example, “Dem donors split on 2012 strategy,” Politico, by Kenneth Vogel, 11/11/10: http://www.politico.com/news/stories/1110/44980.html (Accessed 11/9/12).
40“Romney spent more money on ads and got much less,” by Tom Hamburger, Washington Post, 12/12/12: http://www.washingtonpost.com/politics/romney-campaigns-tv-ad-strategies-criticized-in-election-postmortems/2012/12/11/a2855aec-4166-11e2-bca3-aadc9b7e29c5_story.html (Accessed 12/13/12).
41Michael J. Malbin, Peter W. Brusoe, and Brendan Glavin. 2012. “Small donors, big democracy: New York City’s matching funds as a model for the nation and states,” Election Law Journal 11 (1): 3–20.
42Peter Wallison and Joel Gora 2009. Better parties, better government: a realistic program for campaign finance reform, AEI Press, pp. 63, 70, 72–73.
43Regarding the current system, there is a lot of misinformation in the public discourse about the influence of donors on candidates. Lots of political science research fails to find direct links between PAC donations and legislative voting, for example. See Stephen Ansolabehere, John M. de Figueiredo, and James M. Snyder 2003. “Why is there so little money in politics?,” Journal of Economic Perspectives 17 (1): 105–130. But, also, there is a tendency to talk about Wall Street donations or donations from certain industries, and to presume that such donations are jointly directed. The Center for Responsive Politics aggregates contributions from individuals by reported employer and industry, for example, and while this is helpful aggregate information, it facilitates arguments about industry influence. In reality, no individual can give a candidate more than $5000 in a 2-year election. Candidates might be influenced by such dollars, or fund-raising help that comes from bundlers, but it is often less appreciated that wealthy citizens have almost no means of leveraging that wealth in a candidate’s campaign directly.
44All is not rosy with Super PAC disclosure, however. They can claim 100% transparency, while taking contributions from groups that do not disclose. Imagine the Center for American Democracy PAC accepting and reporting a million dollar contribution from the Center for American Democracy 501c4. This sort of money laundering allows the PAC to claim fealty to the law, while also shielding public spotlight on its donor base.
45David Primo and Jeffrey Milyo. 2006. “Campaign finance laws and political efficacy: evidence from the States,” Election Law Journal 5 (1): 23–39.