Public Funding for Political Campaigns

Institute for Justice Research Brief

What Does Research Say About Public Funding for Political Campaigns?

David M. Primo
Associate Professor of Political Science
University of Rochester

August 2010

The issue of public funding for political campaigns is heating up as the U.S. Supreme Court considers whether to take up the case of Arizona Freedom Club PAC v. Bennett, a First Amendment challenge by the Institute for Justice to Arizona’s public funding law. Legal challenges to similar laws in Connecticut and Florida continue to work their way through the courts, and the U.S. Congress is considering the Fair Elections Now Act, which would provide congressional candidates with government money to run for office. Public funding advocates often make bold claims about the benefits of these systems, but scientific evidence supporting these claims is scarce. This research brief examines the evidence about the effects of public funding.


1. Do “matching funds”the focus of Arizona Freedom Club PAC v. Bennetthave an effect on the speech of independent groups and candidates who refuse to accept government subsidies?

Yes, matching funds cause candidates to alter the timing of their speech in campaigns, infringing First Amendment rights.

Arizona’s public funding system, known as “Clean Elections,” provides additional public subsidies—so-called “matching funds”—to publicly funded candidates whenever their privately funded opponents or independent groups raise or spend more than a limit set by the government.[1] For example, in a multi-member House race, once a privately funded candidate has raised or spent up to the government cap, every additional dollar he raises for or spends on his own speech results in an additional dollar of matching funds for all of his publicly funded opponents (minus 6 percent to account for fundraising costs). Moreover, if a group independent of the privately funded candidate spends money speaking out in his favor, his publicly funded opponents are again entitled to matching funds. In effect, then, when independent groups or privately funded candidates speak in amounts deemed excessive by the government, the government attempts to “level” that speech by directly subsidizing their political and ideological opponents. Those challenging the law in Arizona Freedom Club PAC argue that matching funds unconstitutionally burden the speech of independent groups and candidates who refuse public funds for their campaigns.

The only statistical research on the effects of matching funds on candidate speech shows that many privately supported candidates facing the prospect of triggering matching funds for their opponents alter the timing of their speech as a result.[2] Specifically, Arizona candidates at risk of triggering matching funds delay their fundraising and spending until the last possible minute in a campaign, holding fire in order to postpone the distribution of matching funds and make them less useful to opponents. Late fundraising and spending of candidates facing the prospect of matching funds far outstrips that of other privately funded candidates not threatened by matching funds—by four to one in the two weeks before Arizona House primary campaigns and, in competitive races, by three to one in the two weeks before the general election. This demonstrates that candidates are acting to minimize the harm caused to their campaigns by their own speech as a result of matching funds distributed under Arizona’s “Clean Elections” act.

For a May 2010 review of Clean Elections programs, the General Accounting Office interviewed candidates and independent groups who also said that to avoid triggering matching funds, independent groups and privately supported candidates delay raising and spending money on speech until late in the campaign.[3] Indeed, earlier research by political scientist Michael Miller shows that this now is common practice in Arizona.[4] Every single privately funded candidate he surveyed raised concerns about matching funds, and some informed Miller that they put off spending until the last minute to avoid helping the opposition with matching funds. As one privately funded candidate put it, “Every dollar I spend over the threshold starts feeding the alligator trying to eat me . . . . I sent out a lot less mail and held a lot less events than I would have but for my hands feeling like they were tied under this system.”[5]


2. Does public funding improve citizen perceptions of government?

The only statistical study to examine the effect of state public funding laws on perceptions of government found that public funding has, in some cases, a small negative effect.

Reformers often claim that replacing private voluntary donations to campaigns with public funding will “clean up” politics. If so, we would expect that citizen perceptions of government would improve as a result. However, in the only study to statistically examine the effect of state public funding laws on perceptions of government, my co-author, Jeffrey Milyo, and I demonstrate that these laws had a small, but negative, effect.[6] We examined survey questions asking citizens whether they believed that they had a say in what government does, whether public officials care about what people like them think and whether they find politics complicated. After controlling for individual- and state-level factors that could muddy the findings, we demonstrate that citizens in states with public funding programs were less likely to believe that officials care what people like them think and less likely to believe that they had a say in what government does.[7]

The small negative effects may be due to the fact that public funding programs rarely live up to the expectations set by reformers. When few positive effects of “better” elections or policymaking materialize, this may lead to increased disillusionment or, at best, have a neutral effect on perceptions of government. Moreover, the effects may be negligible because many voters are not even aware of the laws. More than a decade ago, Maine and Arizona enacted one type of public funding system, Clean Elections programs that prohibit participating candidates from receiving private contributions in exchange for full public funding. Yet in a recent survey conducted by the GAO, barely half of the respondents in those two states were even “a little aware” of these laws.[8] And, of those in Maine who were aware of the laws, only 20 percent said the laws increased their confidence in government, while 15 percent said the laws decreased their confidence in government. The comparable figures in Arizona were 26 percent and 22 percent. The rest were unsure or said Clean Elections laws had no effect.


3. Does public funding affect voter participation?

The strongest evidence shows that public funding has no effect on or reduces voter turnout.

The most rigorous examination of this question to date studied turnout in all 50 states, controlling for a variety of factors that could affect turnout rates.[9] In a working paper, two co-authors and I find no effect of public funding (both full- and partial-funding systems) on turnout in gubernatorial elections and a modestly negative effect on turnout in legislative elections. Looking specifically at Clean Elections systems for legislative candidates, we estimate that they lead to a reduction in turnout of about two percent.

In another working paper, Miller studies Clean Elections in Maine and Connecticut and shows that voters who have already turned out to vote are more likely to cast a vote in a legislative race (rather than abstaining from that particular race) if at least one of the candidates has accepted public funds. Miller does not suggest that more voters are heading to the polls as a result of Clean Elections, only that fewer voters are failing to cast votes in certain races, and this effect is small—a 1.5 to 2 percentage point reduction in such ballot “roll-off.”[10] In fact, Miller focuses on roll-off in part because he does not expect that public funding of state legislative races will affect turnout: “Even in the fully funded state legislative elections, it is unrealistic to assume that down-ticket races alone can affect the number of citizens who turn out to vote.”[11]

The results of previous research on public funding and turnout, which had methodological limitations, are mixed.[12] The strongest evidence, therefore, points to public funding at best having a tiny positive effect in reducing roll-off, and at worst having a negative effect on turnout overall.

There are at least two reasons why public funding may not increase turnout, as reformers promise. First, public funding programs may depress turnout indirectly: When the promised effects on policymaking are not borne out, trust in government and citizens’ beliefs that they can make a difference are depressed, in turn lowering turnout. Second, public funding may lead to spending reductions in competitive races, which may depress turnout compared to a privately funded campaign. Further research is necessary to examine these possibilities.


4. Does public funding increase the competitiveness of elections?

There is no consistent body of scholarly research establishing that public funding increases campaign competitiveness.

Some measures of electoral competitiveness increased after Arizona and Maine implemented Clean Elections programs—but it is not clear that Clean Elections was the cause. Furthermore, although some measures of electoral competitiveness increased, this was not the case for all or even the most important measurements.

For example, studies show that Maine saw more candidates running for office and Arizona saw more contested races (those where candidates have opponents) after reform. This is not surprising: One of the basic principles of economics is that if you subsidize something, you get more of it. Pay politicians to run for office and more will. Data also suggest that there are more races with narrow vote margins. But the same data show that incumbents are reelected just about as often after reform as before, so challengers are no more successful at gaining office, contrary to reformers’ hopes. Indeed, one study concludes that the 2004 House election in Arizona was “something of a disappointment to campaign finance reformers: The percentage of incumbents in competitive races in 2004 was the same as it was in 2000 (about 36 percent), declining from a post-1990 record of 47 percent in 2002.”[13] Although such before-and-after snapshots are instructive, it is very difficult to determine whether Clean Elections caused these changes. Other factors may be at play, and these studies do not attempt to account for them. For example, during the same time period, Maine changed contribution limits in significant ways, and Arizona implemented term limits.

So what do statistical studies that do attempt to control for such complicating factors show? Overall, there is limited evidence that either Clean Elections or other public funding programs that offer only partial funding have appreciably affected competitiveness, if we focus on measures like narrow vote margin or reelection rates. For example, two co-authors and I find little effect of public funding programs on competitiveness in gubernatorial elections and note that the modest positive effects of the laws on legislative elections, as identified by other authors, are likely to dissipate over time due to candidates strategically adjusting to the law, among other factors.[14] Political scientist Neil Malhotra finds that Clean Elections had some pro-competitive effects in the 2000 Arizona Senate election, but his approach has some methodological limitations.[15] For instance, Malhotra uses only one year of data during which the Clean Elections law was in force—the first year, 2000—so we do not know if these results are part of a pattern or apply only to 2000. Economist Thomas Stratmann, in a study of 42 states, estimates that partial public funding in Minnesota and Clean Elections in Maine reduce vote margins dramatically (by 15 percentage points) but have no effect on incumbent reelection rates.[16] However, Stratmann does not control for unobserved or immeasurable factors in the states that may drive vote margins, such as the pool of quality candidates, so these results may be tapping more than the effect of public funding. Finally, the GAO study finds no evidence that incumbent reelection rates in Maine and Arizona were affected by Clean Elections. Moreover, although the GAO determined that vote margins appeared to decrease in recent years, they were not able to attribute those declines to Clean Elections.[17]


5. Does public funding reduce the perception or reality of “special interest” influence in politics? What about corruption?

Survey and interview evidence does not support the claim that special interest influence has been reduced in politics.

The aforementioned GAO survey asked Maine and Arizona residents whether they believed Clean Elections has reduced special interest influence. About the same percentage of respondents believed that interest group influence had increased—17 percent in Maine and 24 percent in Arizona—as believed it had decreased—19 percent in Maine and 25 percent in Arizona. The GAO also interviewed 11 candidates for office in both Maine and Arizona and asked them about their perceptions of the role of interests in the wake of Clean Elections. In Maine, only four candidates believed interest group influence had decreased, and all of these were publicly funded candidates. In Arizona, only two candidates believed this to be the case. Results for interviews with interest groups were not markedly different.[18]

For political scientists and economists, these results are not surprising. First, there is very little systematic evidence that campaign contributions affect the decisions of elected officials.[19] Groups may have many motivations for giving, but it is not realistic to think that an elected official will change his or her vote on a piece of legislation based on a small contribution. In fact, if it were the case that money could move votes, corporate political action committees (PACs) would presumably max out their contributions to elected officials each year. Yet, corporate PACs rarely hit donation ceilings at the federal level.[20] Second, campaign contributions are only one way that interests can participate in the political process. Lobbying, for instance, is an important way that interest groups transmit information about the likely consequences of legislation. Even legislators elected through Clean Elections receive this information.



As Congress considers the Fair Elections Now Act and the U.S. Supreme Court decides whether to hear Arizona Freedom Club PAC v. Bennett on the merits of matching funds, public debate will focus on questions about the effects of public funding systems and claims that they improve perceptions of government, encourage competitiveness in electoral races, and encourage participation in the political process.[21] In answering these questions, the importance of social scientific evidence is paramount. While statistical evidence is never perfect and should be rigorously scrutinized, it is important that the best evidence be brought to bear on questions that speak to fundamental First Amendment concerns.

Comparing the claims and promises made by public funding advocates with the actual evidence demonstrates that public funding programs have delivered few, if any, of the benefits promised by their promoters, and they have certainly not resulted in the fundamental transformation and rebirth of confidence in government the promoters sought. On the other hand, the cost of such programs—not only in terms of their negative effect on the timing and nature of political speech in the states with such programs, but also in terms of wasted public resources—has been demonstrable and real. In other words, public funding is a program that promises much, delivers little, and raises real constitutional and policy problems.

[1] In primary elections, matching funds are awarded on the basis of privately funded candidates’ spending beyond a government cap, while in general elections, they are awarded on the basis of funds raised beyond the cap. For both types of elections, independent groups’ spending in support of a privately funded candidate counts against that candidate’s cap.

[2] David M. Primo, Expert Report, McComish v. Brewer, 605 F.3d 720 (9th Cir. 2010) (No. 10-15165).

[3] U.S. Gov’t Accountability Office, GAO-10-390, Campaign Finance Reform: Experiences of Two States That Offered Full Public Funding for Political Candidates 79–80 (2010).

[4] Michael Miller, Gaming Arizona: Public Money and Shifting Campaign Strategies, 41 PS: Pol. Sci. & Pol. 527, 527–32 (2008).

[5] Id. at 528–29.

[6] David M. Primo & Jeffrey Milyo, Campaign Finance Laws and Political Efficacy: Evidence From the States, 5 Election L. J. 23, 23–39 (2006).

[7] Public funding had no effect on responses to the question regarding whether politics is too complicated.

[8] U.S. Gov’t Accountability Office, supra note 3, at 76, 101–02.

[9] Jeffrey Milyo, David M. Primo & Matthew L. Jacobsmeier, Does Public Financing of State Election Campaigns Increase Voter Turnout? (2009) (working paper) (on file with authors).

[10] Michael G. Miller, Citizen Engagement and Voter Behavior in Publicly Funded Elections (2010) (working paper), available at

[11] Id. at 8.

[12] David M. Primo & Jeffrey Milyo, The Effects of Campaign Finance Laws on Turnout: 1950-2000 (2006) (working paper) (on file with authors); Donald A. Gross & Robert K. Goidel, The States of Campaign Finance Reform (2003).

[13] Kenneth R. Mayer, Timothy Werner & Amanda Williams, Do Public Funding Programs Enhance Electoral Competition?, in The Marketplace of Democracy: Electoral Competition and American Politics 245, 259–60 (Michael P. McDonald & John Samples eds., 2006).

[14] David M. Primo, Jeffrey Milyo & Tim Groseclose, State Campaign Finance Reform, Competitiveness, and Party Advantage in Gubernatorial Elections, in The Marketplace of Democracy, supra note 13, at 268, 268–85.

[15] Neil Malhotra, The Impact of Public Financing on Electoral Competition: Evidence from Maine and Arizona, 8 State Pol. & Pol’y Q. 263, 263–81 (2008).

[16] Thomas Stratmann, The Effect of Public Financing on the Competitiveness of Elections (May 2009) (working paper) (on file with author).

[17] U.S. Gov’t Accountability Office, supra note 3, at 34. For a critique of the GAO study, see Michael G. Miller, After the GAO Report: What Do We Know About Public Election Funding? (2010) (working paper) available at

[18] U.S. Gov’t Accountability Office, supra note 3, at 73–75.

[19] Stephen Ansolabehere, John M. de Figueiredo & James M. Snyder, Jr., Why is There So Little Money in U.S. Politics?, 17 J. Econ. Persp. 105–30 (2003).

[20] Jeffrey Milyo, David Primo & Timothy Groseclose, Corporate PAC Campaign Contributions in Perspective, 2 Bus. & Pol. 75–88 (2000).

[21] See, e.g., Fair Elections Now Act, S. 752, 111th Cong. (2009); and Fair Elections Now Act, H.R. 1826, 111th Cong. (2009).

Email Address
Please enter a valid email address

Institute for Justice
901 N. Glebe Road, Suite 900
Arlington, VA 22203
Tel 703.682.9320, Fax 703.682.9321
© 1997-2015