Position-taking in House debate and the issue of immigration

Abstract

Traditional studies of legislative behavior often rely on analyses of roll-call votes. However, this approach ignores variation in legislator position that may not be expressed through a recorded vote. To better examine legislative position-taking, I rely on legislative speech and explore what factors motivate a member of Congress to take a position on immigration. I apply Wordfish, a text-scaling algorithm, to House debate from the 104th and 109th sessions of Congress in order to estimate each legislator's position on immigration. This approach facilitates a more nuanced understanding of legislative position-taking enabling us to speak to the degree of support or opposition to a particular policy – something we are unable to extract from studies relying on vote data alone. Results indicate first, the factors influencing legislative speech have changed over time, with institutional motivations playing a greater role than constituent preferences in determining positions; and second, the issue has become increasingly polarized with Republican members becoming more conservative in their speech over time, while Democrats less so.

Keywords:

• Latino politics
• immigration
• legislative behavior
• political speech

Acknowledgements

The author would like to thank Bradford S. Jones, Kenneth Benoit, Sven-Oliver Proksch, and three anonymous reviewers for helpful comments and suggestions.

Notes

1. I use the terms “Latino” and “Hispanic” interchangeably to refer to individuals who trace their origin or ancestry to the Spanish-speaking countries of Latin America.

2. See, for example, Poole and Rosenthal's (1985) NOMINATE.

3. Restrictive positions are understood to mean any policy aiming to limit, reduce, or prevent immigration.

4. See, for example, the Comparative Manifestos Project (Budge et al. 2001; Budge, Robertson, and Hearl 1987; Laver and Budge 1992) and the Policy Agendas Project.

5. Some studies argue the issue of immigration should not be measured along a single dimensional space because of the variety of different policy dimensions that accompany this issue. Jeong et al. (2011) argue in the USA, the debate over immigration has been related to two major dimensions: “admission,” pertaining to border control and the number and type of immigrants in the USA, and “rights,” which pertain to access to public benefits. As this paper is concerned with debate on the issue of immigration relating to two bills aiming to restrict immigration (or the “admission” dimension), I avoid the potential pitfalls of measuring multi-dimensional speech in one dimension.

6. It is important to mention that this time period may be unique due to the aforementioned millions of individuals that participated in marches and other events nationwide during the spring of 2006 to protest the immigration reforms proposed by Congress. However, as legislative proposals to restrict immigration remain on the agenda, it is reasonable to expect interest in this issue to remain high.

7. The printed copy of the Congressional Record, allows members to revise their remarks and extend their remarks after the conclusion of the legislative business day. The ability to revise and extend remarks offers members of Congress the opportunity to complete an argument that may have only seem limited time on the floor, presenting a more complete picture of a member's position. Additionally, it also offers members who were not given the opportunity to speak on the floor to the ability to record their voice. It is possible that members may use the opportunity to revise their remarks and change a position a later time (usually within five legislative days) to better reflect their district's preferences, however, since there were only 21 insertions relating to the Immigration Control and Financial Responsibility Act of 1996 and 13 insertions relating to the REAL ID Act (compared to well over 100 individual speeches occurring on both pieces of legislation) it is highly unlikely that members drastically changed their position to reflect a more favored viewpoint.

8. See Laver, Benoit, and Garry (2003) for a discussion of the problems in extracting policy positions from political speech rather than texts such as party manifestos.

9. I use Will Lowe's (2011) JFreq program, available at http://www.williamlowe.net/software/.

10. See Slapin and Proksch (2008) for a detailed discussion of implementing Wordfish.

11. The subcommittee of the Judiciary committee is currently known as the Subcommittee on Immigration, Citizenship, Refugees, Border Security, and International Law, which has jurisdiction over immigration and naturalization, border security, admission of refugees, treaties, conventions and international agreements, claims against the USA, federal charters of incorporation, private immigration and claims bills, and non-border enforcement.

12. See Appendix 1 for plots from the position estimates of each speaker on each bill.

13. Further discussion of the politically relevant words is provided in Appendix 2.

14. Rocca (2007) finds that these seemingly disadvantaged legislators opt for more unconventional methods for taking positions, such as one-minute speeches, in the House.

15. While the results presented in Table 2 appear to indicate the effect is conditioned by party membership, this interaction is not significant for the substantively possible range of data.

16. There is some evidence the effect of ideology is moderated by representing a border state district, similar to what was observed in the 104th Congress. This result would indicate that conservative members from Border States (those most familiar with the issue of immigration on a day-to-day basis) are less likely to conservative positions, however the interaction is only significant at the .10 level, meaning the effect is marginal at best.

17. See Branton et al. (2011) for a discussion of the nationalization of the immigration issue during this time period.

18. See the work of Pantoja, Ramirez, and Segura (2001), as well as Lopez and Minushkin (2008) and Barreto (2013) for discussion of the repercussions and potential repercussions of anti-immigrant posturing.

19. There is continuing uncertainty over whether President Bush received 40% of the Hispanic vote in 2004, as indicated by exit polls in the 50 states and the District of Columbia conducted on Election Day, or 44%, as indicated by the nationwide National Election Pool exit poll (see Suro, Fry, and Passel 2005 for a discussion of the varying methodological debates regarding Hispanic turnout in 2004).