L
as Vegas is known for many things, but probably not as the epicenter of political advertising. In 2011 and 2012, however, more than 60,751 presidential ads aired in the Las Vegas media market. They totaled nearly two million seconds of television time, amounting to about 23 uninterrupted days. Sin City featured an additional 55,000 ads for other elections ranging from City Council and Justice of the Peace to the U.S. House and Senate. To watch every political ad aired on local broadcast stations in Las Vegas last year would have required 43 days of television viewing.1
Is that a lot? Elections today are often described as featuring a “torrent of ads”, with the overall cost usually being called “obscene” or “insane.” Colbert King, a columnist for the Washington Post, blogged on Election Day that 2012 was
the most obscene display of campaign spending in history. “Obscene” is a carefully chosen word. What we have witnessed this election cycle is a stunning display of uncontrolled political desire. Luring that kind of money for lustful spending is—or should be considered—offensive to our sense of morality.
Whether election costs are offensive is a matter of opinion and debate, of course. If we reason that candidates and campaigns need such huge sums of money to buy the television ads they believe they need to win and can get that money only from sources who expect favors in return, then that’s a problem. If we reason that there is no necessary or simple connection between the sources of campaign funds and how legislators behave once in office, then there is less of a problem, or even no problem.2 But however we reason, we need a good handle on what elections actually cost, whether those costs are higher today than before, and why changes in spending occurred. It turns out that these questions make for a thorny thicket. But the thicket is worth clearing.
At the end of the day, 2012 was indeed a very expensive election, one of the more expensive ones in recent memory. Careful analysis suggests that cost may not in itself be a problem for fairness and democratic probity, but that other factors probably are.
How much did the 2012 election cost? What figures do we include in the total? What baseline do we evaluate those costs against in order to arrive at some reasonable comparison with past elections? These questions define a numerator and denominator problem: The total cost of a campaign (the numerator) and the appropriate comparison number (the denominator, say, GDP) are not clear and sometimes not easy to find. These numbers can also be arbitrary and subject to manipulation in service of a desired outcome.
Consider the numerator. We might decide to include only campaign expenditures as opposed to campaign fundraising. Lots of congressional campaigns, for example, leave piles of money in their war chests as a signal to potential challengers in future elections. According to Federal Election Commission (FEC) reports from the 2012 election, candidates for Federal office maintained a combined total of more than $230 million in campaign accounts as of February 2013. Expenditures, in contrast, are related directly to the cost of running for office in a particular election.
Choosing between expenditures and fundraising is easy, however. It is much harder to decide what constitutes an expenditure. Should we count the costs of the presidential party conventions, including the public money allocated for those occasions (which in recent elections also include costs for anti-terrorism preparations)? Should we include spending by unregulated interest groups like 527s, even if we don’t know exactly how that money was spent or whether it was relevant to a Federal campaign? Much of these data are easily available, but not for an extended time series. We have good data on elections, for example, after the 1970s congressional campaign finance reforms, but we have only sporadic and incomplete numbers for prior years.
Consider one of the costliest campaigns in American history: that of 1896. Herbert Alexander’s 1984 book Financing Politics pegged the cost of the 1896 presidential general election at a little more than $4 million, with McKinley spending $3.35 million to Bryan’s $675,000. Alexander used numbers as reported to the Congressional Record. George Thayer’s 1974 book Who Shakes the Money Tree? argued that the election totaled $7–8 million, with McKinley spending the bulk. This may seem like a small difference, but $4 million in 1896 comes to more than $108 million in today’s terms. Thayer’s estimates increase the potential cost of the election to more than $217 million in 2012 dollars. The number we choose has real consequences for any calculation of overall cost.
There is an additional complication. Most assessments of presidential elections look at candidate spending across time, but some elections include lots of non-candidate spending from interest groups and political parties. In 1980, for example, candidates for President reportedly spent $91 million. Parties and groups reported an additional $22 million in pro-candidate ads. Excluding those outside ads undervalues the cost of the 1980 presidential election by a considerable amount. Such concerns remain critical to evaluating the cost of recent elections. In the 1990s, parties used vast resources of unregulated “soft money”, meant ostensibly for party branding and state and local races, to advertise “issue-based” messages that touted their own Federal candidates or attacked the opposition. They spent more than $1.6 billion in soft money between 1992 and 2002 before Congress banned it.
Including interest group totals is obviously germane to outcomes, but it is not easy to track. As close observers of campaign finance recognize, loopholes in disclosure laws allow lots of expenditures by interest groups to escape notice. For example, if an interest group-sponsored television ad mentions or pictures a candidate for Federal office but does not tell viewers to vote for or against that candidate, it is reported to the FEC only if it appears within thirty days of a primary or sixty days of a general election. These are referred to in anodyne fashion as “electioneering communications.”
Ads outside these windows can be substantial in both number and cost. In the 2012 Montana Senate race, interest groups sponsored more than 38,000 television ads in the race between Democrat Jon Tester and Republican Denny Rehberg. Only 17,000 of these ads, however, aired within the thirty-day and sixty-day reporting windows. For the other 21,000 ads, interest groups could escape reporting requirements altogether by simply leaving off any taglines urging voters to cast a ballot for or against a candidate. And according to coding by the Wesleyan Media Project, of the 38,000 ads in the race, more than 27,000 did not include such a tagline.
Such ads were formally called “issue advocacy” prior to the McCain-Feingold reforms in 2002, which created the thirty-day and sixty-day reporting mandates. And in the elections between 1996 and 2002, groups spent millions on ads that avoided reporting. It’s impossible to know exactly how much was spent by interest groups in this fashion. The best data available, comprising television ad totals and scholars’ close observations of campaigns, put the total at about $137 million, but that surely underestimates the true number.
In more recent elections, interest groups and parties have been even more active, for reasons discussed below. In 2012, for example, candidates spent $1.4 billion running for President. Interest groups and parties added another $650 million for the presidential primary and general election campaigns.
While compiling campaign costs is the significant challenge, figuring out the denominator poses problems too, albeit less cumbersome ones. The basic problem is to figure out the most sensible comparison. One easy reference point is the size of the economy, as contained in GDP. This measure gives an easily accessible figure and allows for strong comparisons across elections. But there are reasons for caution with using this measure. For one, the economy is huge, and campaign costs are small in contrast. In 2012, Federal elections, inclusive of candidate, party and interest group spending, totaled as a conservative estimate about $4.3 billion. (This total excludes certain spending; the Center for Responsive Politics estimates the cost to be $6.3 billion.) This was the costliest campaign in the post-1970s era, even when adjusting for inflation. But the total still only accounts for about 0.03 percent of GDP. Using GDP, therefore, can shift the discussion to the apparent miniscule cost of elections. But this is no more useful as a stand-alone measure than pointing out that McDonald’s reported $7 billion in revenue in 2011 was only 0.05 percent of GDP, which might imply that hamburgers are not a popular food.
One might argue, then, that the better comparison is total dollars spent to total votes. But even here we need to specify what we mean by total votes. In the 2012 presidential election, about 129 million Americans voted. That works out to about $8.50 per voter if we use general election candidate spending as the numerator. (It comes to about $11.50 if we include party and interest group spending on behalf of a candidate in the general election.) But the vast majority lives in non-swing, non-battleground states and was not targeted by the campaigns, parties or interest groups. If we compare total campaign spending by Obama and Romney (excluding Romney’s primary election expenditures) to the 37 million voters in the states with the ten closest outcomes in the election, the total cost per voter rises to more than $29.
2012 in Context
W
ith such considerations in mind, where does 2012 fit in terms of overall cost? Because of the limitations of data, we should first compare presidential elections between 1888 and 2012 in terms of candidate expenditures only. Party and group expenditures can be added for the elections after 1976 to estimate a second (and better) total cost. We can also compare Federal election costs, comprehensive of presidential and congressional elections, between 1980 and 2012. We include the following in our totals:
Presidential candidate spending between 1888 and 2012. We can use Alexander’s totals from his 1984 book for totals before 1976, and FEC reports for totals between 1976 and 2012.
Congressional candidate spending by general election candidates between 1980 and 2012, obtained from FEC reports.
Party coordinated and independent expenditures between 1980 and 2012. Coordinated expenditures are pro-candidate ads jointly planned by parties and candidates. These are strictly limited with caps varying for presidential, Senate, or House races. Independent expenditures are pro-candidate ads produced and distributed without the benefiting candidate’s input. There are no caps on these, but they have only been legal for parties since 1996. Both forms of expenditures can only be funded with regulated contributions. For all this we can use data from FEC reports.
Interest group independent expenditures between 1980 and 2012. These are pro-candidate messages produced and distributed without input from the benefiting candidate. Prior to the Citizens United v. FEC decision in 2010, only registered Political Action Committees (PACs) and some smaller non-profits could sponsor these types of ads. Now, any interest group can produce them without limitations on the source of funding. We use data from FEC reports for these numbers.
Party soft money expenditures between 1992 and 2002. Parties used soft money in the 1980s, but they were not required to report these expenditures to the FEC until 1992.
Interest group issue advocacy totals, where possible, between 1996 and 2002. Data come from the Wisconsin Advertising Project, which tracked ads in Federal races in 1998–2002, and from an Annenberg Public Policy Center report3 for the 1996 estimates.
Interest group electioneering communications between 2004 and 2012, obtained from FEC reports.
Not every kind of number we can find deserves inclusion, however. It is probably not a good idea to include costs for the presidential nominating conventions, for example, because the Federal government provides funding for part of the conventions and city “host” committees can accept contributions in excess of the public grant. The conventions are election-related in the sense that they formally nominate the party’s candidate, but the expenditures for such events (especially since 1980) are hardly related to the outcome of the election itself. In other words, the multi-day television shows the parties produce to wow their audiences do not correlate with election outcomes. They more likely qualify as lobbying expenditures for donors to the host committees.
Nor does it make sense to include costs reported to the IRS by 527s between 2004 and 2012. For one thing, some of this spending is reflected in the electioneering communication totals reported to the FEC; including them risks double-counting. Second, it is incredibly difficult to know which expenditures by these committees are related to Federal elections. Finally, the analysis excludes expenditures by congressional candidates who lost primary elections and total disbursements in regulated funds (beyond independent and coordinated expenditures) by political parties. The reason is that the data are harder to assemble for elections in the 1980s.
For a denominator the analysis uses a metric developed by Steven Ansolebehere and his colleagues in a 2003 article.4 They use GDP in an election year, as “deflated” by GDP in the year 2000. For example, U.S. GDP in 1896 was $15 billion, which was just 0.16 percent the size of the economy in 2000. They transform campaign costs in an election year into “millions of 2000 GDP dollars” to reflect costs in comparison to national income. This makes sense if we hope to compare elections accurately.
The subsequent analysis shows clearly that the 2012 election was the costliest in recent memory. Three key trends stand out.
First, looking only at presidential candidate spending, 2012 and 2008 are essentially tied in terms of overall cost. (For both elections the analysis uses primary and general candidate election spending as the numerator. However, just using expenditures from the two-party nominees does not change the interpretation.) Both elections were the costliest presidential campaigns since 1936 and rank ninth out of the 32 elections between 1888 and 2012.
Second, if we include party and interest group spending in the totals after 1976, the 2012 election cost 23 percent more than the one in 2008 and rises to the sixth costliest election since 1888. The five more expensive elections were 1888, 1892, 1896, 1900 and 1936 (see chart below). If we expand our totals in 2012 to include all spending by the Democratic and Republican National Committees (instead of just their pro-Obama and Romney ads) and count convention spending, the election rises to the third most expensive contest since 1888, after only 1892 and 1896.
The comparison to 1892 and 1896 is important, and not only because these results surprise most contemporary observers. These elections only cost about $4 million but represented huge investments relative to the size of the economy at that time. Indeed, the 2012 presidential election in deflated GDP dollars only amounted to about two-thirds of the 1892 and 1896 totals. The 1896 election in particular is notorious for its cost, and for the blatant means by which Republicans leveraged big business for cash. These high costs inspired Progressive campaign finance reform efforts that resulted in the ban on corporate donations to candidates and parties in the Tillman Act of 1907. Ranking behind 1896, even a fair distance behind, is thus no real honor.
There is a corollary point worth considering. Even as 2012 ranks behind 1892 and 1896, one might focus on the historic change in the way money is spent. By some estimates, Obama and Romney devoted more than half of their budgets to television, radio, and internet advertising. While there is a debate in political science about the democratic value of such ads, almost no one accepts them as an ideal form of campaign discourse. Of note, such hit-and-run messaging did not exist, absent the printed pamphlet, in the late 19th century, and 1896 in particular featured the high-level oratory and barn-storming candidacy of William Jennings Bryan. The high cost of 2012 in the context of a strong focus on brief messages might lead one to isolate the Obama-Romney race as the more problematic one.
The third key assessment comes from looking at election costs between 1980 and 2012, inclusive of all Federal elections. Here we see a rise in costs that is nearly unchecked. Comparing 2012 with 1980, for example, we see a nearly 143 percent jump in costs; and 2010 compared to 1982 was nearly 41 percent higher. The 2010 midterm was the costliest in the eight midterms since 1982, and 2012 was by far the costliest presidential election year in the nine elections since 1980.
The data are clear: Elections have grown considerably more expensive in recent years, even when accounting for costs in terms of GDP. Costs are rising fairly sharply even when we exclude certain expenditures from 527s and congressional primary losers. My analysis is also fairly conservative in the costs incurred by political parties, excluding any regulated spending by them that was not explicitly allocated for the promotion or attack of a Federal candidate. And note that parties are more active today than in the 1970s and 1980s, meaning that excluding some party spending will make us prone to underestimate how much costs have spiked.
Why?
K
nowing where the 2012 elections stand in the time series is important because we have to know the numbers to talk intelligently about whether costs are high or low, steady or rising. But this analysis really only prepares us to ask the questions that matter most: Why have the costs of presidential and Federal elections spiked so sharply? Does the answer to that question say anything important about fairness and justice in American democracy?
Let’s begin to get at that question by considering the fact that congressional candidate expenditures since 1980 remained largely flat once we deflate them to GDP dollars—albeit with 2010 and 2012 being the costliest of the 17 elections in that span. Expenditures for congressional candidates in 2010 and 2012 were more than 20 percent higher than the average between 1980 and 2008. Candidates in congressional elections raise substantial sums from PACs, which can be maintained by corporations, trade associations and unions, but PAC contributions to candidates (again scaled to GDP) have not budged in the past 32 years.
Spending jumps for presidential campaigns are far more pronounced. The total candidate cost of the 2004 election, for example, was double the cost of the 2000 election (in inflation-adjusted dollars), and candidates in 2008 and 2012 spent another 50 percent more than candidates paid out in 2004.
The jump in total costs since 2000 reflects in part the slow death of the presidential public funding system. Since 1976, candidates could opt into primary and general election public funding in exchange for a limit on total expenditures. The limits were fairly severe, with hard caps on spending in primary states and a total limit on general election expenditures. Every major candidate for President opted into the system until 2000, when Governor George W. Bush did not use public money for his primary election campaign. In 2008, Barack Obama opted out for both the primary and general elections, making him the first major party nominee to refuse the public grant for the general election. (Every other major candidate that year on both sides of the aisle, except former Senator John Edwards, also opted out.) By 2012, both Obama and Romney refused public money for the general election, and no major Republican candidate took such funds during the primary phase. The main consequence was to raise the total amount spent in pursuit of the presidency.
Ultimately, candidates decided to change course in this way for several reasons. One was to avoid being outspent by parties and interest groups. Both parties and outside groups, it turns out, are the real source of increased election costs.
Consider first the political parties. Between 1980 and 1990, the parties spent a total of $136 million on coordinated expenditures for congressional and presidential candidates. Parties expanded their use of soft money dramatically in the elections of 1994–2002, however, spending more than $1.7 billion in soft money and coordinated and independent expenditures in those five elections. That represents 13 times the total spent between 1980 and 1990.
The benefits of soft money are clearly illustrated in a scene from The West Wing. During the third-season episode “Gone Quiet”, staffers of fictional President Josiah Bartlett convene to plan strategy for Bartlett’s upcoming re-election campaign. When the conversation turns to Democratic Party advertising, Toby (the President’s communications director), Sam (Bartlett’s speechwriter), Bruno (the campaign manager) and Connie (Bruno’s assistant) acted out the following very realistic exchange:
TOBY: Look, we can’t spend soft money on a primary ad anyway, so…
SAM: No, he’s passing the magic words test.
TOBY: What magic words test?
SAM: The U.S. Supreme Court, Buckley v. Valeo. The court created a loophole by ruling [that hard money] only applies to communications that in express terms advocate the election or defeat of a clearly identified candidate for Federal office.
BRUNO: You don’t put “vote Bartlett” in the ad, you can pay for it with unmarked bills from a bank heist if you want to.
CONNIE: And we should know. There’s also footnote 52, where the Court said campaign-finance laws only apply to communications with the terms “vote for”, “elect”, “support”, “cast your ballot for”, “Smith for Congress”, “vote against”, “defeat”, “reject”, and that’s it. [pause] I’m savant-like.
TOBY: If it doesn’t use those specific words…
BRUNO: It is an issue ad.
CONNIE: You know what they say about money and politics.
SAM: No.
CONNIE: It’s like water on pavement.
SAM: Why is it like water on pavement?
CONNIE: That’s a good…
BRUNO: It finds every crack and crevice.
SAM: The standard ought to be, does the ad try to influence the outcome of the election? If so, you can’t use soft money, period.
BRUNO: Well, zippity-do-dah, Sam.
SAM: Excuse me?
BRUNO: That isn’t what the standard is. And I think we should run in the same election as everybody else.
Beginning with the 2004 elections, parties were banned from raising and spending soft money. As parties often do, however, they adapted to new restrictions, in this case by focusing more energy on raising regulated contributions from a wider base. They expanded their use of joint fundraising with congressional and presidential candidates, for example. Between 2004 and 2012, the parties spent $1.5 billion in coordinated and independent expenditures.
All told, beginning in the 1990s parties asserted a more dominant role in Federal elections. The reason is relatively simple: Elections after 1994 held more consequence. There was an old assumption in Washington that Republicans had the advantage in presidential elections, but Congress belonged to the Democrats. That changed in 1992 (with the election of Bill Clinton) and 1994 (when Newt Gingrich helped orchestrate the Republican takeover of the House and Senate). During the same period the parties drifted further apart ideologically. What emerged in the elections of the 1990s was a fierce struggle between bitterly polarized parties over control of Congress and the White House—a struggle that continues today.
The numbers tell a powerful story. In 1996, the Republicans controlled the House with 230 seats, 12 more than the 218 needed to control the chamber. As the election approached that year, estimates put likely Republican seats at 217, with 34 others too close to call. The years 1998, 2000, 2002, 2006 and 2010 featured similar levels of uncertainty. The same was more or less true of Senate elections. Parties thus had great incentive to mobilize resources for these elections, and particularly for the handful of close races that would determine the balance of congressional power.
Interest groups also responded in the mid-1990s with heightened investments in Federal elections. They too leveraged the “magic word” loophole. They did so for the same reason the parties did: high electoral stakes and razor-thin party control of Congress. The AFL-CIO, for example, spent tens of millions on “issue ads” attacking newly elected Republican Congressmen. Pharmaceutical companies backed shadow groups like Citizens for Better Medicare and the United Seniors Association, both of which advertised on behalf of Republicans. Business backed the Chamber of Commerce and the Business Roundtable. Groups like MoveOn.org, Media Fund and Progress for America heavily backed Kerry or Bush in the 2004 presidential election. These latter groups formed as 527s, exploiting an IRS loophole in McCain-Feingold to raise millions from wealthy individuals.
The interest group side of this story is a bit complicated. There have always been loopholes in elections laws that allowed for issue advocacy, but one had to run a real risk of punitive sanction to take advantage of them. So long as the Supreme Court believed that limiting interest-group electioneering was a constitutionally defensible argument, interest groups’ fears of sanctions or lengthy legal battles tempered their overall participation. Indeed, getting wealthy donors to contribute to these efforts was often a challenge for the groups’ leaders.
All of that changed in January 2010 when the Supreme Court handed down its decision in Citizens United v. FEC. The specifics of Citizens United are beyond the scope of this article.5 Suffice it to say for present purposes that what began in 2010 was a slow reduction in the barriers to entry for interest groups in backing candidates for Federal office. Any real threat of legal trouble for a group seeking to bolster a Federal candidate evaporated. Hoops to clear in raising dollars for “issue ads” or “magic word” ads disappeared. Unions, pro-business associations, corporations and ideological groups felt free to raise and spend whatever sums they felt necessary. Moreover, when the FEC in 2010 sanctioned the creation of Super PACs, candidates’ close associates were free to form organizations that raised unlimited cash for them in the next election.
Consider just one figure in that regard. PACs have been around in large numbers since the 1970s, contributing highly regulated dollars to candidates’ campaigns. In almost every election since 1980, PAC candidate contributions have dwarfed the amounts spent by interest groups to advertise on behalf of candidates. In the 1980s and early 1990s, for example, the contribution-to-advertisement ratio was nearly 11 to one. That ratio declined fairly quickly in the late 1990s and early 2000s. By 2012, however, for the first time on record, interest group ads outpaced PAC contributions and did so by more than 2.6 to one.
Dollars alone hide one other trend: Interest groups drive up costs not only because they are spending more but also because they pay more for ads than candidates do. Federal candidates are required by law, specifically Section 315(b) of the Communications Act of 1934 (as amended), to get a discount on radio and television ads: the “lowest unit rate.” This applies to ads aired by candidates within 45 days of a primary and 60 days of a general election. Interest groups do not similarly benefit.6 The intention of the rate discount for candidates is in the larger spirit of Section 315(b), that all candidates have the means to reach voters during an election campaign.
The Wesleyan Media Project uses cost estimates from CMAG/Kantar Media, and those data tell a powerful story regarding differential costs by ad sponsors. In 2012, candidates for President sponsored more than 822,000 ads on local broadcast stations across the country, paying an average of $527 per spot. (This datum obviously hides a large diversity in ad costs across large and small media markets). Interest groups, however, paid out an average of $791 a spot for just under 500,000 airings. In House races, the spending was $494 for candidates and $969 for groups, and $575 and $684, respectively, in Senate races.
Ad costs are exceptionally hard to get a handle on, though, making strong longitudinal comparisons of ad costs difficult. One might wonder, for example, whether election costs are increasing in part because broadcasters are charging more for ads (and at rates that outpace inflation). In principle, the “lowest unit” rate would only mitigate that problem, since broadcasters set the lowest rate and are free to increase it from year to year. No one really doubts that broadcast stations make a mint off competitive election campaigns, but it is really too hard to verify empirically how ad costs (and specifically political ad costs) are changing over time.
T
he cost of a campaign (or an election cycle) is not in itself a troubling trend. To be sure, campaign costs are up dramatically by any reasonable measure, but greater levels of spending for the most part have legitimate explanations: the up-for-grabs nature of a more or less evenly divided electorate, the passion of one’s cherished beliefs and so on. What deserves our greatest scrutiny is rather the allocation of that higher spending by the various players in American politics. Elections since 1994 have demonstrated that parties and interest groups are more dominant forces in electoral politics than at any point since the early 20th century. And since 2010, interest groups in particular have emerged as even more dominant than the party committees.
Some lopsided figures help make the point. If one over-counts party costs, counting all disbursements from the Democratic and Republican Party committees (including transfers to state parties and voter mobilization efforts) and under-counts interest group costs—say, by counting only pro-candidate ads (excluding issue advocacy and voter mobilization)—interest groups nearly matched the parties in the amount spent in the 2012 elections. That is $1.2 billion for parties compared to $1.15 billion for groups. That dollar-for-dollar match is unprecedented. In every election with good data on total party spending, parties have outspent groups by upwards of fifty to one. In just two elections since Citizens United, interest groups have come to match parties in the resources deployed toward electing or defeating Federal candidates.
To summarize succinctly, costs are higher because of both supply and demand factors. The supply of cash more readily flows from interest groups (and did for parties in the 1990s) because of lower barriers to entry (that is, less stringent campaign finance laws). At the same time, the demand for cash is higher than ever due to our evenly divided and ever-shifting political landscape.
The power that parties can play in American politics is not a serious cause for concern. Indeed, one can argue that stronger parties that can discipline members and deliver close votes are good for American politics. Parties are a centripetal force. They represent broad-based philosophies of government, have transparent finances, are accountable through their candidates, and have long histories in American politics. Few were fans of six- and seven-figure checks funneled through the parties by a loophole in campaign finance laws, but that soft money period of the 1990s seems quaint in comparison to the more interest group-centered politics we have now.
Indeed, today’s campaign finance system incentivizes unaccountability. As campaign costs shift to interest groups, two problems are paramount. First, candidates have every reason to foster the creation of Super PACs as parallel campaign committees. Both Romney (with Restore Our Future) and Obama (with Priorities USA) had Super PACs organized and run by former aides. Every Republican presidential primary candidate did too. And this is not limited to campaigns for the White House. The Democrats used their House Majority PAC and Senate Majority PAC to help congressional candidates; the Republicans relied on the Congressional Leadership Fund and the YG Action Fund. The existence of Super PACs skirts close to the ban on coordination between candidates and groups. In truth, all of these committees are essentially arms of party leadership, organized and run by close associates of incumbent politicians.
This is problematic because the primary purpose of these groups is to funnel resources into competitive elections while masking intentions. It is one thing for voters to see ads sponsored and paid for by the Democratic and Republican National Committees. We can process messages carrying party labels through our grasp of and predisposition toward each party’s ideology. But when groups like Restore Our Future, or Winning Our Future, or the Red, White and Blue Fund put ads on television, voters know little about those groups’ purposes at the moment of exposure. Knowing that all of these groups are closely associated with particular candidates, or funded by the candidates’ former staff or even family, might alter the way we process the message. None of this is nefarious or necessarily corrupt, but it’s slimy and does the voting public no good.
Some evidence indicates that expensive campaigns can produce higher turnout and more knowledgeable voters, but interest groups do not contribute to these effects by simply increasing the cost of a campaign.7 Dollars can have differential impacts. Some recent scholarship, for example, shows that an interest group’s negative ads can shield the benefiting candidate from any backlash.8 While interest groups take the low road in sliming an opponent, the candidates can claim to maintain a positive and civil discourse.
There is a second problem. While Super PACs disclose donors, many other groups are not required to do so. These are so-called 501(c) organizations that receive non-profit status from the IRS. 501(c)5 and (c)6 groups are unions and business trade groups, and 501(c)4s are ostensibly social welfare groups. Many such organizations (Americans for Prosperity, American Future Fund and Americans for Responsible Leadership, for example) claim to promote social welfare by airing television ads that promote or attack a particular candidate. But of course these groups are not really non-partisan. The result is that voters are now flooded with “dark money” ads from groups with no visibility or political identity beyond generically vague names and bare websites.
Consider some of the ad totals. According to the Wesleyan Media Project’s coding methods, there were more than 420,000 “dark money” ads in 2011 and 2012. These were from sponsors like 501(c)4 nonprofits that do not disclose their donors to the FEC. These ads aired at an estimated cost of $321 million. Moreover, 240,000 additional ads aired (at an estimated cost of $208 million) from groups that only partially disclose their donors. Compare this to 237,000 ads (at an estimated cost of $207 million) from groups that have full disclosure requirements (such as Super PACs). All told, the bulk of ads from interest groups in the 2012 election cycle aired from out of the shadows.
This is depressing because it makes it nearly impossible to know who bankrolls these vast efforts. It also raises troubling questions about what such donors expect in return for their investments. Indeed, while scholars have devoted considerable attention to the relationship between PAC contributions and incumbent voting behavior (finding very little by way of vote-buying), we do not know what millions invested in pro-candidate super PACs will do to the psyche of legislators. Lawrence Lessig in Republic, Lost labels this a “dependency corruption”, in which legislators gravitate to donor interests almost subconsciously.
The efforts of super PACs and other groups in elections are not without real consequence. A growing consensus among political scientists holds that ads can move voters in small doses, especially in races lower down the ballot (and sometimes even in presidential elections).9 The effects are not large, nor do they endure very long, but they can move enough votes in close races to arguably determine the outcome.10 As control of Congress hangs in the balance of a mere dozen or so House races and a few Senate races, many of which are determined by one or two percentage points, “dark money” can make a huge difference.
There may be some good news, however. Change need not come from a constitutional amendment or a different Supreme Court. Citizen-centered change is possible now. What is striking about the campaign finance environment today is that so much interest group spending is channeled via television advertising. (This was one knock on Republican Super PACs in 2012. As President Obama diversified his voter outreach to include sophisticated online and peer-to-peer efforts, pro-Romney Super PACs ran a relatively old-fashioned effort based on mass-media advertising.) Voter-based appeals are easier to blunt, however, than direct lobbying efforts in the halls of Congress and in the reg-writing bureaucracies of the Executive Branch. If voters refuse to heed ads from groups they do not recognize, or if they attempt to decipher the origins and intentions of such groups, then the impact of such ads (and the money that funds them) might decline to the point that many big donors give up the game. We are only victims of chicanery to the extent that we refuse to take ownership of our own vote.
Of course, some might argue that American voters are too dumb, too lazy or too apathetic to be serious consumers of political information. If so, we’re doomed. But if enough voters are serious about being strong citizens, it would make a difference. Justice Antonin Scalia’s advice, offered in early 2012 to a group of lawyers, may have real resonance: Voters can take charge of America’s elections and limit the influence of big donors by simply turning off the television or changing the channel. Call it remote control.
1All references to political ad airings and cost from 2012 are from the Wesleyan Media Project.
2This question is the subject of decades of research in political science. The balance of the evidence suggests the latter view, though that is by no means the consensus perspective among scholars.
3Deborah Beck et al., “Issue Advocacy Advertising During the 1996 Campaign” (Annenberg Public Policy Center, 1997).
4Stephen Ansolabehere, John M. de Figueiredo and James M. Snyder Jr., “Why Is There So Little Money in Politics?”, Journal of Economic Perspectives (Winter 2003).
5See the section, “Campaign Finance after Citizens United” in the July/August 2010 issue of The American Interest, as well as other articles in that issue.
6The complexities of the lowest unit rate are beyond the scope of this article. They involve, for example, different “lowest unit” rates for ads aired at various times of day. And sometimes candidates pay more than the lowest rate to ensure their ads are not preempted by non-political advertisers who are willing to pay more.
7See, for example, Michael Franz, Kenneth Goldstein, Travis Ridout and Paul Freedman, Campaign Advertising and American Democracy (Temple University Press, 2007).
8Deborah Jordan Brooks and Michael Murov, “Assessing Accountability in a Post-Citizens United Era: The Effects of Attack Ad Sponsorship by Unknown Independent Groups”, American Politics Research (May 2012); Christopher Weber, Johanna Dunaway and Tyler Johnson, “It’s All in the Name: Source Cue Ambiguity and the Persuasive Appeal of Campaign Ads”, Political Behavior (September 2012).
9Michael Franz and Travis Ridout, The Persuasive Power of Campaign Advertising (Temple University Press, 2011).
10Alan Gerber, James G. Gimpel, Donald P. Green, and Daron R. Shaw, “How Large and Long-lasting Are the Persuasive Effects of Televised Campaign Ads? Results from a Randomized Field Experiment”, American Political Science Review (February 2011).