In 17 District Courts, Removed Cases Are Remanded at Least 20% of the Time
Removal to federal court is a powerful tool of the parallel judicial systems in the U.S. to oust a case from state court jurisdiction. A defendant can automatically remove a case from state court, filing the proper notices and supporting documents, if the case could have originally been brought in federal court, usually based on diversity or federal questions jurisdiction. 28 U.S.C. § 1441 et seq. After being served with a state court complaint, defendants must decide how to respond in a relatively short amount of time, deciding whether to answer, challenge the sufficiency of the claims or jurisdiction, or remove the case to federal court, if appropriate. However, defendants may not know at the outset whether their decision to remove to federal court will stick, so to speak. This is where historical data can help.
Removal: Nuts and Bolts
To effect removal, a defendant must file a notice of removal in the district court, in the district and division within which the state action is pending, within 30 days from receipt of the first pleading in the state action that sets forth a removable claim. 28 U.S.C. § 1446(b)(1), (2)(B), (3), (c)(1). This process can involve significant effort in providing the proper notices, complying with the varying local rules, obtaining joinder from other defendants where needed, and compiling supporting documentation to support removal jurisdiction. See 28 U.S.C. § 1446. Moreover, if the defendant has not responded to the complaint, it generally has a short window—potentially as short as a week—within which to file its responsive pleading. Fed. R. Civ. P. 81(c)(2). This can mean that a defendant is preparing its response concurrently with the removal, adding to the cost and effort needed at the start of the case.
The original purpose of removal jurisdiction in diversity cases was to protect non-resident defendants from local bias in state court, as well as to put federal question cases before federal—rather than state—judges. In diversity cases, this bias rationale underpins the rule that local defendants, i.e. defendants who are citizens of the forum state, cannot remove. 28 U.S.C. § 1441(b)(2). However, in modern practice, these rationales have been replaced by strategic or practical considerations by defendants who prefer litigating in federal court rather than state court. These considerations could include a preference for federal pleading standards, more favorable jury rules (e.g. Fed. R. Civ. P. 48), early disclosures under Fed. R. Civ. P. 26(a)(1), general familiarity with federal rules, or faster time to bring a case to trial—to name a few. Also, research has shown that removal eliminates the plaintiff’s forum advantage and decreases its win rate. This is reason enough for an out-of-state defendant to remove and decrease the plaintiff’s chances of success.
To start, how common is removal? According to data collected by the Federal Judicial Center, since 2000, 702,095 cases have been removed from state court to federal district court, making up 13% of all civil cases initiated in the federal system (not counting multi-district litigation). Each year, 33,433 cases are removed on average.
Of the cases removed over the last two decades, 57% were based on diversity jurisdiction and 40% were based on federal question jurisdiction, with the remainder being made up of cases where the U.S. government is a party. The biggest share of cases removed to federal court involve insurance litigation (13%); other personal injury claims (12%); other contract actions (11%); civil rights jobs (8%); product liability claims (8%) and other civil rights claims (6%).
Across the country, we see variation across the districts on the number of removals filed in each district. Not surprisingly, some of the busiest districts see the most number of raw removals, e.g. the Central District of California, the Southern District of Texas, the Middle District of Florida, and the District of New Jersey. However, taking the number of removals as a percentage of overall filings in each district, we see that the Southern District of Mississippi, Middle District of Louisiana, Southern District of West Virginia, and the District of New Mexico have the highest percentage of cases filed based on removal jurisdiction. The Southern and Eastern Districts of New York appear to have a lower than expected number of removals in comparison to the general volume of cases filed in those districts.
Examining removals over time since 2000, we see some variation in the number of cases removed to federal court by subject matter and district. In 2001, we see a large jump in asbestos cases, as well as a large number of removals in the Northern District of Ohio and the Eastern District of Virginia. We also can see variation in the number of cases removed for product liability claims, as well as an increase in foreclosure cases being removed in the wake of the Great Recession. Other types of cases seem to be removed at a relatively constant rate over time.
U Turn: Remand Back to State Court
Yet, sometimes after all the effort to remove a case (potentially requiring the compilation of supporting documentation under short timelines to support jurisdiction), a case is remanded right back to the state court. In fact, district courts are obligated to examine jurisdiction—even if no objection is made or the parties stipulate to federal jurisdiction. See, e.g., University of South Alabama v. American Tobacco Co., 168 F.3d 405, 410–411 (11th Cir. 1999) (holding error to rule on plaintiff’s right to dismiss removed case before determining existence of removal jurisdiction). That means, after expending the resources to remove and potentially file a motion to dismiss, a defendant may end up back where it started.
After settlement, remand has been the second most common outcome for cases removed since 2000. Remanded cases settled 33% of the time, were remanded 18% of the time, voluntarily dismissed 11% of the time, otherwise dismissed 10% of the time, and disposed of by judgment on motion before trial 8% of the time.
As a percent of the cases removed to a district, the District of the Northern Mariana Islands and the District of Guam have the highest average remand rate, based largely on the small volume of cases. Those districts are followed by the Southern District of Alabama, the Central District of California, the Northern District of Georgia, the Middle District of Alabama, and the Southern District of West Virginia. In the top 20 districts, approximately 20%, or one in five, of removed cases are remanded on average. For example, in the Central District of California, 28% of removed cases were sent back to state court. Compare that to the District of Massachusetts where 10% of removed cases were remanded and the District of the District of Columbia where 8% of removed cases were remanded. The Western District of Tennessee and Middle District of North Carolina have the lowest remand rates at 6.4% and 5.9%, respectively.
Across subjects, there is substantial variation in remand rate as well. For example, rent, lease, and ejectment cases have an average remand rate of 55.6%. In other words, over half of rental cases were sent back to state court. On average, 44% of removed civil rights voting claims were remanded. The district courts remanded 60%, on average, of other immigration actions. While some types of cases are never remanded, some of the most common cases to be removed, are remanded on average with less frequency, including civil rights jobs claims (10.3%); insurance claims (13.6%); other contract actions (13.7%); other personal injury claims (15.8%).
Although an aggregate high-level view of the success of removal to federal court, we can see that the rate of remand varies by district and subject matter, sometimes in surprising ways. Federal courts seem more willing to keep certain types of cases compared to others. Before a defendant spends significant time and money deciding whether to remove a case to federal court, it would be worth spending time investigating whether the likelihood of remand is high and worth the risk.