Ex Ante and Ex Post Control over Courts in the US States: Court Curbing and Political Party Influence

Abstract

Court curbing, proposed policy that attempts to “restrict, remove or otherwise limit” the power of the judiciary, occurs regularly and with considerable variation throughout the US states. I deviate from past studies, which consider court curbing as an ex post control mechanism, by focusing on ex ante controls of state courts – judicial selection rules and processes. I argue that levels of political party influence over judicial selection (before a judge is seated) alter the motivation to engage in court curbing after a judge is seated. I test this novel ex ante explanation with an original data set of court curbing and judicial selection from 2015 to 2018. Mixed results offer some support for my theory while opening new opportunities for study, particularly among partisan election systems.

Keywords:

• Judicial independence
• state courts
• court curbing
• judicial selection

Acknowledgements

I offer my thanks to the participants at the Midwest Political Science Association 2021 Annual Meeting, the faculty and graduate students at the Political Science Department workshop at SUNY Binghamton, and the thoughtful anonymous reviewers of this manuscript for their beneficial feedback. Special thanks goes to Dr. Wendy Martinek for her tremendous support and encouragement throughout this project and to Dr. Matthew Walz and Andrew Foote. Finally, I must thank my muses – my children (Jenna, Theodore, and Luca) and my wife Olivia, who provided me with daily inspiration on this project.

Declarations of interest statement

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Notes

1 Bills proposed:

HRJ1037-2016 http://www.oklegislature.gov/BillInfo.aspx?Bill=hjr1037&Session=1600

SJR 23-2014 http://www.oklegislature.gov/BillInfo.aspx?Bill=SJR23&Session=1400

HJR 1006-2016 http://www.oklegislature.gov/BillInfo.aspx?Bill=HJR1006&Session=1600

HB 3162-2016 http://www.oklegislature.gov/BILLINFO.ASPX?SESSION=1600&BILL=HB3162

SJR 50-2016 http://www.oklegislature.gov/BillInfo.aspx?Bill=sjr50&Session=1600

2 This paper considers only state courts of last resort, in line with work by Leonard (2016), Blackley (2019), and Hack (2022).

3 See Appendix Table 8 for the states the have the lowest and highest amounts of court-curbing efforts.

4 A potential implication of my theory may extend to how legislatures structure-court curbing bills to consolidate influence over ex ante control mechanisms in the future (judicial selection process) through current ex post control mechanisms. This may occur in the use of selection-type court-curbing bills. We see this occur in both Tennessee and North Carolina between 2015 and 2018. In Tennessee in 2015–2016, Republicans in control of state government transitioned the judicial selection method for their state court of last resort from a merit selection system (with the least amount of political party influence) to more of a gubernatorial appointment system (with the most amount of political party influence). A year later in North Carolina, the Republicans in control of state government also moved the judicial selection system for their state court of last resort from system with less political party influence (nonpartisan election) to one with more political party influence (partisan election).

In my analysis testing this, there seems to be some support that Republicans, when in the majority of the state upper chamber, do appear to propose more selection-type court-curbing bills that offer more political party influence over the judicial selection process, compared to Democrats in the majority. This does not appear to be the case in the lower chamber. In terms of the percentage of selection-type court-curbing bills that increase political party influence over judicial selection, again Republicans in the state upper chamber propose a greater percentage of selection-type court-curbing bills that increase political party influence over judicial selection, compared to Democrats in the majority. And again, this dynamic does not seem to occur in the lower chamber.

This implication on the content and directionality of court-curbing proposals warrants further study by scholars.

5 It may be the case that where the out-party in state legislatures has little chance of molding the judiciary through ex ante control mechanisms due to low competitiveness, they may engage in proposing more court curbing to attempt to control the judiciary ex post. However, this does not appear to be the case as nearly two-thirds of all court-curbing proposals are sponsored by members of the “in-party” (the legislative majority).

6 Two states, Oklahoma and Texas, have two state courts of last resort. All other states have one. This means that the maximum number of observations in a year is 52, one for each state court of last resort.

7 The GaveltoGavel blog’s link is at http://gaveltogavel.us/database/. I used LegiScan (https://legiscan.com/) to cross reference text and information on each bill. LegiScan is a real-time legislative tracking service, which also possessed links to the bill’s page and text on state legislature websites.

8 I cross-referenced my coding of judicial selection with the categorization used by the Book of the States, which is published by the Council of State Governments. The book of the states categorized Ohio and Michigan as partisan elections due to the partisan nature of the candidate selection process (even though both states use nonpartisan ballots for judicial races in the general election). I coded Ohio and Michigan as nonpartisan but estimated the model with both coded as partisan selection as a check on robustness (Nelson, Caufield, and Martin 2013).

The Book of the States categorized West Virginia’s judicial selection type as partisan elections. While, the state did use partisan election prior to 2015, HB2010 was signed into law in the March of 2015 and became effective prior to the general election that year, meaning nonpartisan elections were used for the first time in 2015. I relied on the method of selection that was used by the end of the year, rather than what was law at the beginning of the year. Cite for West Virginia House Bill 2010 (2015): http://www.wvlegislature.gov/bill_status/bills_history.cfm?INPUT=2010&year=2015&sessiontype=RS.

In addition, I used McLeod (2009) and the National Center for State Courts’ Judicial Selection page – http://www.judicialselection.us/ for further guidance and verification.

9 I do not include observations for state court-years when the state legislature did not meet. Montana, Nevada, North Dakota, and Texas (with two state courts of last resort) state legislatures did not meet in even years of this study (2016 and 2018).

10 The last year in the Bonica and Woodruff data set for state judge ideal points is 2012. Due to this limitation, I hold the state court median justice’s ideal point constant as the value in their final year of data – 2012.

11 I include robustness checks for my legislature ideology measures, which include the pivotal legislator ideology score and legislative party median ideology score. Using the Shor and McCarty ([2011] 2020) individual state legislator ideology scores, I constructed the “Pivot” Legislature Ideology Score by determining the state-specific pivot points for each state’s upper chamber and the median for each state’s lower chamber. I defined pivot as the legislator score for the legislator who could break a filibuster in the state’s upper chamber. I also constructed two “Party Medians” Legislature Ideology Scores (one for Democrats and one for Republicans) by taking the mean score of the median Democrat in both chambers and the mean score of the median Republican in both chambers.

Findings remain consistent across different legislative ideology measures and model specifications for gubernatorial appointment systems, offering robustness to my findings (see Appendix Tables 11–15). Partisan election systems appear more sensitive to how legislature ideology is measured. In one of the three legislature ideology measures among the “base” models, partisan election systems remain significant and consistent. However, when introducing more variables in the “full” models, partisan election systems drops out of significance when using two of the three legislature ideology measures.

12 I removed observations for state court-years when the state legislature did not meet. This includes Montana, Nevada, North Dakota, and Texas in 2016 and 2018 as these state legislatures only meet in odd years. I removed observations for state court-years that lacked state legislature ideology scores. This includes Hawaii (all 4 years), Iowa 2018, Kentucky 2015 and 2016.

13 As Nelson, Caufield, and Martin (2013) note, Michigan and Ohio could be considered nonpartisan or partisan election systems. I estimated the model with Michigan and Ohio as partisan election states as a check on robustness, with results similar to the base model. See Appendix Table 10.

14 As a robustness check, I estimated “full” models that included Governor’s Party, Percent of Democrats in the Lower Chamber, Percent of Democrats in the Upper Chamber, Majority Party in Lower Chamber, and Majority Party in Upper Chamber (Appendix Tables 13–15). Gubernatorial appointment remains statistically significant across all model specifications. Partisan elections are more sensitive to model specifications in terms of legislature ideology score measures but estimates remain consistent in significance and direction in the “full” model that uses my primary legislature ideology score measure – mean of the legislature medians.

15 Since 1980, only three governors were elected without being nominated by a political party: Angus King (Maine), Lincoln Chafee (Rhode Island), and Bill Walker (Alaska). Additionally, three other governors won election having been nominated by minor parties: Wally Hickel (Alaska – Alaskan Independence), Lowell Weicker (Connecticut – A Connecticut Party), and Jesse Ventura (Minnesota – Reform Party).