Representation, Competing Principals, and Waffling on Bills in US Legislatures

• Published date: 01 August 2016
• Author: Jeffrey J. Harden,Justin H. Kirkland
• Date: 01 August 2016
• DOI: http://doi.org/10.1111/lsq.12132

JUSTIN H. KIRKLAND

University of Houston

JEFFREY J. HARDEN

University of Colorado Boulder

Representation, Competing

Principals, and Waffling on Bills

in US Legislatures

Legislators are often placed in the position of representing the interests of their

constituents against the preferences of their own party leaders. We develop a theoretical

framework indicating that these cross-pressured legislators are more likely to initially

support legislation and subsequently change their minds than are legislators whose con-

stituents and leaders share similar preferences. Moreover, we expect this pattern to be

most pronounced among members of majority parties than minority-party members. We

test our expectations using data on bill cosponsorship and final passage votes from 46

lower state legislative chambers and the US House, finding considerable support for our

theory.

On May 22, 2014, the US House passed H.R. 3361, commonly

referred to as the USA Freedom Act, by a margin of 303 to 121. The

f‌inal vote was strongly bipartisan with 179 Democratic supporters and

124 Republican supporters. Despite robust support for the bill, 76 of

the bill’s 152 cosponsors voted against the passage of the legislation

at f‌inal roll call. This outcome poses an important question for legisla-

tive scholars. Why did some members of Congress remain supportive

of the legislation while others reversed course, or “waff‌led” on the

bill? More generally, why do cosponsors choose to vote against their

own bills? Answering this question is critical to both scholarly and

normative discussions of legislative politics. Waff‌ling affects cohe-

sion and unity within parties, which makes it an important part of

understanding legislative organization. Additionally, waff‌ling impacts

the process of representation because it signals unreliability in legisla-

tors’ decision making, which may impact constituents’ attitudes

toward their representatives.

While voting against one’s own cosponsored legislation is rare,

examples do ref‌lect some consistencies. Democratic Representative

LEGISLATIVE STUDIES QUARTERLY, 41, 3, August 2016 657

DOI: 10.1111/lsq.12132

V

C2016 Washington University in St. Louis

Dutch Ruppersberger (MD) voted against his own cybersecurity bill in

2012 after his party leaders and the White House voiced strong opposi-

tion to it (Sasso 2012). Additionally, during the 2010 Senate vote on

President Obama’s proposed National Commission on Fiscal Responsi-

bility and Reform, six Republican cosponsors opposed establishing the

commission during the roll-call vote after Senate Republican Leader

Mitch McConnell (KY) voiced his opposition to the bill (Rushing

2010). Finally, eight Democratic cosponsors of the Federal Reserve and

Transparency Act of 2012 voted against their own bill along with their

party leaders, despite the bill passing by a wide margin (Nelson 2012).

In each of these examples, the legislators who waff‌led in support of their

own legislation faced some conf‌lict between their initial commitment

and the wishes of their party leaders. Sometimes, their leaders wished to

dramatically alter the bill and subsequently vote on a weakened version.

Other times, their leaders openly opposed and voted against a piece of

legislation. In both circumstances, waff‌ling was strongly connected to

conf‌licts between what a legislator initially wished to do and what his or

her party leaders desired as a legislative outcome.

In this research, we develop and test a theory of legislative waff‌ling

on bills. We def‌ine waff‌ling as legislators initially sponsoring or cospon-

soring a bill and then voting against that bill on f‌inal passage. We

suggest that waff‌ling is most likely to occur when legislators are cross-

pressured by the preferences of their constituents and the preferences of

their party leaders. When these two primary legislative principals are at

odds with one another, legislators’ support strategies for legislation have

reasons to change over the life of a bill. Using data on cosponsorship

and roll-call voting across 46 lower state legislative chambers, our

empirical analysis provides strong support for our expectations and de-

monstrates that legislators with moderate districts and extreme leaders,

and legislators with extreme districts and moderate leaders, are the most

likely legislators to waff‌le on legislation. Our analysis also suggests that

these effects are stronger for majority party legislators than minority

party legislators and are strongest for majority parties with considerable

agenda-setting powers.

Finally, we use the model results from our state legislative data to

make predictions about patterns in legislative waff‌ling in Congress.

We then test those predictions with da ta on waff‌ling in the 112th US

House. Our results from the House m irror our expectations based both

on our theory and our state-level model. Specif‌ically, we demonstrate

that minority-party Democrats’ propensity to waff‌le increased in their

districts’ ideological extremity, while major ity-party Republicans

evidenced no relationship between district ideology and waff‌ling.

658 Justin H. Kirkland and Jeffrey J. Harden