Campaign Finance Project Proposal

Introduction

In January 2010 the Citizens United v. Federal Election Commission, 558 U.S. ­­__(2010) decision came down through the Supreme Court of the United States regarding campaign finance. The case reviewed some of the provisions in the Bipartisan Campaign Reform Act (BCRA) better known as the McCain-Feingold Act passed in 2002. The BCRA built upon the Federal Elections Campaign Act (FECA) of 1971 which was designed to limit the amount of money an individual or a corporation could donate to a federal campaign. The BCRA was designed to block the many loopholes of FECA. One of the ways the BCRA blocked these loopholes was to prohibit all corporations from engaging in what Citizens United refers to as “electioneering Communication.” Because Corporations were limited in their contributions to candidates they instead paid for advertisements for or against candidates independent of the campaign. These advertisements are known as independent expenditures, and the BCRA was designed to prohibit them. Specifically the BCRA prohibited both for-profit and non-profit corporations from funding any broadcast, satellite, or cable communication that mentioned any of the candidates for the 60 days leading up to the general election. Citizens United struck down this provision of the BCRA on First Amendment grounds.

The Citizens United decision caused a firestorm of controversy about the role of independent expenditures. Many criticized the court, including President Barack Obama in his 2010 State of the Union Address, because of the belief that corporations can now more effectively buy votes off of representatives. But apart from corporate influence on policy the Citizens United decision brings up interesting questions regarding its effects on campaigns.

Since 1970 reelection rates in the House of Representatives has been hovering around 95% (Krasno 1994) This striking incumbent advantage must be considered in evaluating all campaign finance reform. Such high reelection rates lead to less accountability in what was designed to be the most democratic institution in the Federal Government. This study will look at the effects of independent expenditures on House elections. It is purposed that removing the prohibitions on independent expenditures will be beneficial for viable challengers in House races. Removing such restrictions will also lead to a better knowledge of candidate name recognition among registered voters during the election.

Literature

Members of the House of Representatives in the United States Congress have had more and more secure reelections since the 1950s. During the 1950s the House had an 80% reelection rate with the incumbents winning an average of 60% of the vote. (Hibbing 1991) That number steadily grew, and only had a slight decrease during the early 1990s. (Smith 1996) Yet even at the decrease the reelection rate hovered around 95% with incumbents winning an average of 70% of the votes. (Hibbing 1991) The 2002 and 2004 federal election cycles were the least competitive in post World War II history. (Abramowitz et al. 2006)

Much of the problem with the high reelection rates are a result of the discrepancies between the amount of money incumbents can raise and spend as opposed to challengers. According to Abramowitz high House reelection rates are mostly due to the inability to raise funds. Even after being adjusted for inflation each dollar challengers spent on campaigns between the 1970s and 1980s bought less votes than the year before. (Abramowitz 1991) This is because incumbents have been able to exponentially raise more money than challengers each year. According to Smith, “House incumbents, who in 1976 outspent challengers by a ratio of 1.5 to 1, in 1992 outspent challengers by almost 4 to 1.”( 1996 pg. 1050) Magleby and Nelson summed up the problem: “both too much and too little is spent in congressional elections. Hundreds of House challengers are seriously underfinanced and largely invisible.” (1990 pg. 2)

Of course, other variables account for high reelection rates. These variables include redistricting and the polarization within the House. (Cover 1977) However, these variables are arguably less important. For example Abromowitz et al. refuted the redistricting hypothesis saying redistricting doesn’t matter. (2006) Yet, even with the consideration of other variables there is a wide consensus that the fundraising discrepancies are a major reason for the reelection rates.

Though there may be consensus that the inability of the challenger to raise sufficient funds is a major reason for the high reelection rate there is not consensus on what to do to solve the problem. The answer revolves around campaign finance reform. It has been shown that limits on campaign contributions decrease the amount of money candidates are able to raise. (Stratmann 2006) The question is whether regulating the ability to raise money inhibits challengers or if it puts them on more even ground with incumbents. According to Bonneau and Cann limiting spending inhibits challengers. “Rather than being an argument on behalf of campaign finance restrictions, we contend that diminishing marginal returns to campaign spending actually produce an argument against restrictions on fundraising and spending paired with the possibility that incumbent and challenger spending have differential effects.” (2010 pg.2)

Bonneau and Cann argue that incumbents receive fewer votes for each dollar that they spend than challengers do. This is because incumbents have greater diminishing marginal returns than challengers. (Bonneau and Cann 2010, Bonneau 2007a, Kenny and Burnett 1992 ) Jacobson takes it farther saying that the more money incumbents spend, the more votes they lose. (Jacobson 1990)  The longer incumbents are in office the greater the diminishing marginal returns. (Erikson and Palfrey 1998) Yet, there are others that say that there are no differences in the diminishing marginal returns between incumbents and challengers. (Gerber 1998, Ansolabehre and Gerber 1994, Stratmann 2006)  This is because incumbents spend more when they face high-quality challengers, and that those challengers receive a large part of the vote regardless of spending. (Levitt 1994)

With fewer restrictions on financing, incumbents will receive more money, but so will high-quality challengers. Assuming incumbents have a greater diminishing marginal return, this improves the chances of the challenger which will in turn cause the challenger to receive even more contributions because contributors donate to candidates who they feel have a good chance of winning. (Snyder 1990, Overton 2004, Bonneau 2007b) According to Aranson and Hinich campaign finance laws “discriminate against challengers to the advantage of incumbents. The challenger begins with a disadvantage, and the effect of the … limitations on individual contributions takes that disadvantage as a precondition and exacerbates it.”(1979, pg. 452) Fewer restrictions would not help challengers with low chances of success, but may greatly help challengers with high chances of success.

Fewer restrictions on contributions resulting in more spending also helps keep the public more informed because advertising is how many less-informed people gain knowledge about the campaign. (Patterson and McClure 1976) Stratmann (2006) argues that advertising makes the public skeptical of the election process, but Coleman and Manna refute the claim. They argue that high levels of spending inform the public about candidates, and have little negative effects. (Coleman and Manna 2000) “Contributions finance the provision of information about candidates to voters and increase the probability that the candidate who would be majority preferred, were voters perfectly informed, wins.” (Coate 2004 pg. 790) This then puts incumbents and challengers on more of an even playing field.

Most of the work done on campaign finance revolves around contributions not independent expenditures, but there is evidence that there is not a big difference between the two because corporations use independent expenditures to circumvent restrictions in contributions and expect the same results from them. (Sullivan 1996)

Hypothesis and Methods

This study will look at the effects of independent expenditures in light of the 2010 House elections. By comparing the 2010 elections to the 2006 elections, the only midterm elections that happened under the restrictions of the BCRA, we can see if the loosened restrictions of independent expenditures creates more competition between challengers and incumbents. This will be done by looking at three hypotheses.

  1. No restrictions on independent expenditures results in fewer discrepancies between money spent by or in behalf of high-quality challengers and incumbents.
  2. For each dollar spent incumbents have a greater diminishing marginal return in number of votes received than do challengers.
  3. The more money spent on television advertisements in an election, the more informed the registered voters are about candidate names and parties.

Hypothesis #1

This hypothesis will simply be to look at the total money spent by or in behalf of each candidate running for the House of Representatives in the 2006 and 2010 election cycle for the 60 days before the election except for the open-seat elections. Then the percentage of the total money spent during the 60 days that was spent by or in behalf of the challenger in each election will be calculated. This includes all money from contributions and all independent expenditures. Each challenger will have between 0 and 100 percent of the money spent in her behalf. The data collection will be simple because all money spent must be reported by law. All data on money spent by the challengers themselves can be found through the Federal Elections Commission at FEC.gov. Data on all money spent on television advertising, both on production and on the purchase of air time, on behalf of the candidate, that is all ads including the challenger’s name in a favorable light or her opponent’s name in a negative light, will be gathered from the TNS Media Intelligence Corporation (tns-mi.com), and also from the FEC.

The treatment in this test is the removal of independent expenditure restrictions. The treatment group is all 2010 elections, and the comparison group is the 2006 elections. The dependent variable is the percent of money raised by the challenger, and the independent variable is the quality of the challenger as judged by their percent of the vote in the Gallup polls before the 60 days. It is believed that when the distribution of the percentage of the money spent by each challenger is calculated a significant statistical difference will be seen between the 2006 and 2010 elections. It is believed that low-quality candidates, or challengers who poll below 35% with registered voters according to Gallup at the beginning of the 60 days, will not see an increase in their percentage. On the other hand, it is believed that high-quality challengers, those that poll at 35% or above, will see a significant increase in their percentage of total money spent. This is because it is believed that interest groups who have had to spread out contributions will now be able to focus all their funds on challengers who they see as viable.

Hypothesis #2

Here I look at the relationship between the dependent variable which is the change each candidate sees in the percentage of the vote she has between the beginning of the 60 days before the election and the election itself, and the independent variable which is the total money spent by or in behalf of that candidate. The polling data will be gathered from Gallup for the beginning of the 60 days, and from the final results at the end of the election. The data from hypothesis one will be used to see the money spent. I will calculate the OLS regression for both incumbents and challengers and will Log the results to account for the diminishing marginal returns. This will determine what percentage of the vote each dollar buys for challengers and incumbents.  We believe that the percentage of the vote each dollar buys for challengers will be much greater than it is for incumbents because the diminishing marginal return is much greater for incumbents.

Hypothesis #3

For this hypothesis we will conduct a simple pre-test post-test quasi-experiment in 60 random districts during the 2010 House elections. Sixty days before the election we will poll by telephone 300 random participants in each district about their knowledge of the candidates. We will ask them if they know the names of both the Republican and Democratic candidates. If they do not we will see if they recognize the names from a list. A point system will be set up to calculate their knowledge. If the participant knows the names of both candidates and the political party they belong to then they will receive five points. If they can name just the incumbent, but recognize the challenger’s name from a list and puts them both in the correct political parties they will receive four points. Three points will be assigned for recognizing both candidates from the list. Two points will be assigned to those who can recognize one of the candidates from the list, and zero points will be assigned to those who can’t recognize either candidate. An additional point will be taken off if they can’t put the candidates in the right political party. We will then repeat this process after the 60 days for 300 randomly selected participants in all 60 districts. The dependent variable then is the difference between the average score before and after the 60 days in each district. The independent variable is the total money spent on each election. Using regression to calculate the results we expect there to be a statistically significant positive relationship between the average point increase after the 60 days and the total money spent in the election during those 60 days. This will show a direct correlation between knowledge of the candidates and money spent in an election.

Validity

The biggest problems with these hypotheses are ones of internal validity which will make me hesitate to assert that  any causal relationships can be determined using these tests. For example, when comparing the percent of the money the challenger raised in 2006 and 2010 I hesitate to say that the change in independent expenditure restrictions was the cause of any statistical variation. The history problem is very prevalent. These were two different elections in different political climates. With hypothesis two we come across the problem of reciprocal relationships. The value of an incumbent dollar may be worth less solely because incumbents only spend more when they face high-quality challengers who would have gained even a larger percent of the vote if the incumbents hadn’t spent as much as they did. Unfortunately we can’t test what would have happened had the incumbent chose not to spend and compare.

With hypothesis three we run into several problems. First, there is a selection problem because those polled are only the ones that chose to respond to the phone survey. We also have instrumentation problems because we simply test name recognition which may not accurately relate the total information an individual has on the election. However, similar tests have been used in the past to test the information level and are generally accepted. (Coleman and Manna 2000) There are also other variables to consider. If candidates are spending more the race is most likely closer and covered more by the media which might account for the increase in name recognition.  For this purpose it is difficult to assert causal relationships.

On the other hand, because of the large sample size, and the anticipation of diminishing marginal returns statistical conclusion validity problems should be minimal. The constructs are easy to define, with the exception of the participant knowledge of the election construct, minimizing construct validity problems. The hypotheses should also not have many external validity problems due once again to the sample size. In two tests it is %100 and in the other it’s about 1/3 of all races and it is selected randomly.

It is believed that these tests will show a positive correlation between the release of independent expenditure restrictions and high-quality challenger opportunity. The challengers will be able to have a larger percent of the money spent spent in their behalf. The money they do spend will be more effective than incumbent money, and the increase in money spent in general will increase their name recognition. Though causal relationships won’t be determined this study will open the door to determine causation with further studies.

Work Cited

  • Abramowitz, A.I. (1991). Incumbency, campaign spending, and the decline of competition in u.s. house elections. The Journal of Politics, 53(1), 34-56.
  • Abramowitz, A.I., Alexander, B., & Gunning, M. (2006). Incumbency, redistricting, and the decline of competition in u.s. house elections. The Journal of Politics, 68(1), 2006.
  • Ansolabehere, S., & Gerber, A. (1994). The Mismeasure of campaign spending: evidence from the 1990 u.s. house elections. The Journal of Politics,56(4), 1106-1118.
  • Aranson, P.H., & Hinich, M.J. (1979). Some Aspects of the political economy of election campaign contribution laws. Public Choice, 34(3/4), 435-461.
  • Bonneau, C.W. (2007a). The Effects of campaign spending in state supreme court elections. Political Research Quarterly, 60, 489-499.
  • Bonneau, C.W. (2007b). Campaign fundraising in state supreme court elections. Social Science Quarterly,88, 68-85.
  • Bonneau, C.W., & Cann, D.M. (2010). Campaign spending, diminishing marginal returns, and campaign finance restrictions in judicial elections.(Unpublished manuscript, under review)
  • Coate, S. (2004). Political competition with campaign contributions and informative advertising. Journal of the European Economic Association, 2(5), 772-804.
  • Coleman, J.J., & Manna, P.F. (2000). Congressional campaign spending and the quality of democracy. The Journal of Politics, 62(3), 757-789.
  • Cover, A.D. (1977). One good term deserves another: the advantage of incumbency in congressional elections. American Journal of Political Science, 21(3), 523-541.
  • Erikson, R.S., & Palfrey, T.R. (1998). Campaign spending and incumbency: an alternative simultaneous equations approach. The Journal of Politics, 60(2), 355-373.
  • Gerber, A. (1998). Estimating the effect of campaign spending on senate election outcomes using instrumental variables. The American Political Science Review, 92(2), 401-411.
  • Hibbing, J.R. (1991). Congressional careers: contours of life in the u.s. house of representatives. The University of North Carolina Press.
  • Jacobson, G.C. (1990). The Effects of campaign spending in house elections: new evidence for old arguments. American Journal of Political Science, 34(2), 334-362.
  • Kenny, C., & McBurnett, M. (1992). A Dynamic model of the effect of campaign spending on congressional vote choice. American Journal of Political Science, 36(4), 923-937.
  • Krasno, J.S. (1994). Challengers, competition, and reelection: comparing senate and house elections . Chelsea, Michigan: BookCrafters.
  • Levitt, S.D. (1994). Using Repeat challengers to estimate the effect of campaign spending on election outcomes in the U.S. House. The Journal of Political Economy, 102(4), 777-798.
  • Magelby, D.B., & Nelson, C.J. (1990). The Money chase: congressional campaign finance reform. Washington, D.C.: The Brookings Institution.
  • Overton, S.A. (2004). The Donor class: campaign finance, democracy, and participation. University of Pennsylvania Law Review, 152.
  • Patterson, T.E., & McClure, R.D. (1976). Television and the less-interested voter: the costs of an informed electorate. Annals of the American Academy of Political and Social Science, 425, 88-97.
  • Smith, B.A. (1996). Faulty assumptions and undemocratic consequences of campaign finance reform. The Yale Law Journal, 105(4), 1049-1091.
  • Snyder, J.M. (1990). Campaign contributions as investments: the u.s. house of representatives, 1980-1986. The Journal of Political Economy, 98(6), 1195-1227.
  • Stratmann, T. (2006). Contribution limits and the effectiveness of campaign spending. Public Choice, 129, 461-474.
  • Sullivan, K. (1996). Political money and freedom of speech. University of California Davis Law Review, 30(3).