Upstream influence: The positive impact of PAC contributions on Marcellus Shale roll call votes in Pennsylvania

Research output: Contribution to journalArticlepeer-review

8 Scopus citations


Political scientists have long tried to explain how interest group lobbying and political action committee campaign donations affect election outcomes and public policy debates. Unfortunately, they are often met with null or conflicting results. Consequently, scholars have tested the influence of campaign donations on Congressional outcomes in different parts of the policy-making process, such as roll call votes and committee hearings. Building on these findings, I use the adoption of a Marcellus Shale impact fee in Pennsylvania to test whether campaign contributions have a different effect on bill amendment roll call votes than final floor votes. Given the differing political contexts of these two votes, it stands to reason that there is variation in the relationship between campaign donations and member voting behavior. Specifically, I expect that legislators have more flexibility in voting on amendments than on final bills, and thus factors other than party, including campaign funding, are also relevant. I find that while party, ideology and tenure are the only significant factors associated with roll call votes on final bills, campaign contributions and local-level salience are positively associated with voting on amendments. This shows that while moneyed interests may not be successful in stopping undesirable legislation, they can achieve legislative victory by shaping the bill as it travels through the legislature.

Original languageEnglish (US)
Pages (from-to)293-314
Number of pages22
JournalInterest Groups and Advocacy
Issue number3
StatePublished - Oct 1 2014

All Science Journal Classification (ASJC) codes

  • Sociology and Political Science


Dive into the research topics of 'Upstream influence: The positive impact of PAC contributions on Marcellus Shale roll call votes in Pennsylvania'. Together they form a unique fingerprint.

Cite this