Transparency and the Federal Election Commission

Abby Wood and Christian Grose

Research update

Bedrosian Faculty Research Award: Government Transparency Laws: Why Do Some Legislators Over Comply?, Awarded January 2015

__________________________

Transparency and the Federal Election Commission: A Boon for Citizens, Candidates, and Democracy?

Overview

Does transparency enhance the ability for voters to evaluate candidates for office, and reward those candidates who are transparent and honest? Did the random audit policy formerly used by the Federal Election Commission (FEC) have positive benefits for both citizens and candidates?  We theorize that increased transparency creates a stronger democratic process by providing more information to voters about candidates. From a public policy perspective, our research also suggests that re-instituting a more robust random audit procedure in the FEC may result in beneficial outcomes for citizens and some candidates for office (those least likely to violate campaign finance laws).  In short, greater transparency via FEC audits could yield better opportunities for high-quality candidates to win office while deterring corrupt and lower-quality candidates from running and winning.

Our research examines legislative over-compliance with transparency requirements.  We have used the Bedrosian funds to target an area in which voluntary audits of compelled disclosures – a type of over-compliance – can provide electoral benefits to candidates for office.  We use data from a set of random, compelled, FEC audits of campaign finance disclosures to form preliminary findings.  Our early work shows that candidates who are audited, regardless of the outcome of the audits, do better than those who are not audited. Audits that reveal no serious campaign finance violations give an additional electoral boost to candidates.  The first finding is particularly surprising.  The second finding indicates that voters are “getting the message” on transparency.  If the results hold up as we continue to gather and analyze data, an important policy implication of our work is that the FEC should allow candidates to submit to voluntary audits, and candidates should embrace voluntary audits as a way to increase voter support.  Our preliminary research indicates that a new policy of voluntary audits by the FEC would have the simultaneous benefit of enhancing transparency on campaign finance for voters, enhance the reelection of high-quality and transparent public officials, and lower the likelihood of reelection for candidates more likely to be corrupt and in violation of campaign finance laws.

Theory

Since the 1970s, most academic, judicial and policy discussions of campaign finance disclosures center on two perceived benefits of campaign finance disclosures:  (1) the ex post informational benefit to voters of knowing which individuals and groups support the candidate, thereby allowing voters to understand candidates’ positions better than candidates’ stump speeches and advertising can convey; and (2) the ex ante anticorruption benefit of having to disclose the people or entities funding a candidate or making expenditures on her behalf.  (See Buckley v. Valeo and academic articles citing it.)  According to an FEC auditor with whom we interviewed for this project, the audit process helps to maximize the two benefits.

We are most interested in the so-called informational benefit and how audits can provide an additional layer of information.  We theorize that transparency has informational benefits to voters.  Specifically, audits not only help verify that the information the candidate provided to regulators is correct; they also provide voters with information on the candidates’ traits such as honesty, integrity, and willingness to comply with the law.  In the theoretical literature in political science, these traits are often called valence characteristics.  Our argument is that voters use campaign finance disclosure information as cues to evaluate the valence of each candidate – specifically which candidates are most competent or trustworthy (Stokes 1963, 1992; Bianco 1994).  Scholars consider “character valence” and “campaign valence” (Stone & Simas 2010), and FEC audits can partially help separate the two. For example, a clean audit showing no violations reveals either personal integrity, the ability to run a well-managed campaign, or both.  An audit that reveals missing transactions might not call personal integrity into question, but it would call into question the ability to run a well-managed campaign.  And an audit that reveals illegal contributions or other substantive violations calls into question personal integrity and the ability to run a well-managed campaign.

We theorize that greater transparency will provide useful information to voters about the type of candidate running.  Those candidates who have few or no violations in campaign finance law will be rewarded by voters, while candidates with significant violations will be harmed electorally.  Thus, transparency is useful to voters in learning about the traits of candidates.

For purposes of this report, we collapse our hypotheses into two main expectations.  (1) Where audits reveal no violations, candidates will receive an electoral boost (from the increase in valence) over both non-audited candidates and audited candidates whose audits reveal violations; (2) Where incumbent audits reveal violations, we will see a higher retirement rate among audited incumbents than non-audited incumbents and than audited incumbents whose audits reveal no violations (from the decrease in valence).

Data and Method

After the federal elections of 1976, the FEC conducted a random audit of 10% of all races for the U.S. House and Senate, and both major party presidential candidates.  They decided to conduct the audits in November 1976, meaning that candidates couldn’t change their behavior during the 1976 campaign on the knowledge that they had a 10% chance of being audited.[1]

The audits came at a heady time for the FEC and for efforts to combat corruption in the federal government more generally.  The Watergate scandal had just ended, and the FEC was a newly created independent federal agency. The FEC, as a new organization, had not placed regulations, processes, and procedures in place, and the auditors were, according to an auditor involved, “very optimistic” as to what they could do with what they had.

The audits took longer than anticipated, but most were concluded in 1978, and were all concluded by the end of 1979.[2]  As a result, during the 1978 and 1980 elections, some opponents had the ability to raise the results of the audit during the campaign, providing additional information to voters about their opponent’s type.  Candidates in severe violation of campaign finance laws were found out via the FEC random audit.  Due to unpopularity from some senior members of Congress who had campaign finance violations, the FEC random audit process ended after this time period.  However, the re-institution of this administrative procedure has been proposed frequently by reformers in Congress and was nearly though ultimately not included in the McCain-Feingold campaign finance reforms.

While substantively interesting as a public policy worthy of evaluating its impact and theoretically important to assess the role of transparency on citizens, candidates and elections, the fact that the FEC conducted random audits enhances the methodological rigor of the study.  Randomization of incumbent members of Congress for a FEC audit means we can causally assess the role of transparency and audits on candidate behavior and electoral outcomes.

In summer 2015, two Bedrosian-funded RAs from the Price School helped us to code the 242 audits started in 1976 as well as gather information on 6334 candidates and committees from the 1976, 1978, and 1980 elections.  They are still gathering additional variables for which we want to ensure balance between the audit- and non-audit groups, but the data collection is substantially complete though continuing.

Early results

We have conducted very early analysis on the data we have so far.  FEC audits show that many members of Congress violated disclosure, reporting, and campaign donation regulations established in the 1970s.  Interestingly, approximately half of members of Congress had no campaign finance violations. Preliminary evidence shows that when the FEC found no campaign finance violations, these members of Congress did better in their reelection bids.  Those whose audits reveal regulatory violations are more likely to do worse in subsequent elections compared to the audited candidates who had no violations.

These results are very preliminary.  We plan to include additional variables in our models as well and carefully account for the time it took the FEC to audit each candidate.

Conclusion and policy implications

Transparency and audits regarding campaign finance are policies that can increase information to voters.  This information results in an increased likelihood that candidates in violation of campaign finance law will not win reelection.

Our future work in this project will proceed along two avenues.  First, we will examine how the relative severity of violations reported affects competitiveness in future elections.  Second, we will search for examples of candidates using the results of their audits to either boast of their clean records or publicize the poor audit outcomes of their competitors, in order to explain how voters actually learn about their candidates.

A strong FEC with audit powers can help inform the public, who in turn rewards elected officials who embrace transparency.  Policy implications of our study suggest that the FEC’s audit powers should be strengthened in order to increase accountability between voters, federal agencies, and elected officials; and that candidates should voluntarily submit to audits, to help reveal their type and earn votes based on valence traits.

__________________________

References

Bianco, William T. 1994.Trust: Representatives and Constituents. Ann Arbor: University of Michigan Press

Buckley v. Valeo, 424 U.S. 1 (1976).

Stokes, Donald E. 1963. “Spatial Models of Party Competition.” American Political Science Review 57(2): 368–77.

Stokes, Donald E. 1992. “Valence Politics.” In Electoral Politics, ed. D. Kavanaugh. New York: Oxford University Press.

Stone, Walter & Simas, Elizabeth. 2010. “Candidate Valence and Ideological Positions in U.S. House Elections.”  American Journal of Political Science 54(2): 371-388.

[1] See p. 19 of the 1978 FEC Annual report (“In December 1978, the Commission reaffirmed the Audit Division’s current audit policy, adopted in November 1976 and revised in April 1978. This policy covered audit activity to be conducted during the remainder of Fiscal Year 1979.” http://www.fec.gov/pdf/ar78.pdf .  Candidates could probably work any adjustments into their final, post-election, disclosures, once they learned they had a chance of being audited.  This is something we will learn more about in conversations with FEC officials.

[2] The available FEC reports indicate that all audits were wrapped up by 1979 (see appendix 4, p. 58 of http://www.fec.gov/pdf/ar79.pdf), and that 83 of the 106 random congressional audits were released in 1978 (presumably before the election, but we will double check this).  This affects the time frame of our analysis.  Temporally, the audits should affect the 1980 election and may affect the 1978 election.

 

 

Comments

One thought on “Transparency and the Federal Election Commission

  1. Are there any other forms or methods by which elected representatives embrace transparency? For example, by pushing for a transparency policy (e.g. open data or the FOIA), or by publishing their campaign donors, or by publishing their log of office visitors? I know these are much harder to quantify or get data on, but I wonder if a candidate who boasts transparency more generally (and walks the the walk) can benefit from the attractiveness of transparency, even if they don’t get audited.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.