2012 Trends in Patent Case Filings and Venue: Eastern District of Texas Most Popular for Plaintiffs (Again) But 11 Percent Fewer Defendants Named Nationwide
By James C. Pistorino
The year 2011 closed with the U.S. District Court for the District of Delaware becoming the most popular place for plaintiffs to file patent litigation after the effective date of the America Invents Act. 1 From Sept. 16, 2011, when the AIA went into effect, through the end of 2011, more cases by more plaintiffs naming more defendants were filed in the District of Delaware than in any other district in the nation. The Eastern District of Texas dropped to being only the second most popular place for plaintiffs to bring patent infringement suits. Thus, to the extent that the AIA was intended to reduce the volume of cases in the Eastern District of Texas, it appeared to be working.
A full calendar year with the AIA in effect has now passed, and the AIA’s longer term effects on patent litigation have begun to emerge. The 2012 case filing data lead to four important takeaways:
• The districts have returned to their pre-AIA rankings, with the Eastern District of Texas regaining the crown as the most popular place for plaintiffs to file patent suits and the District of Delaware somewhat behind in second place.
• The concentration of patent cases in the Eastern District of Texas and the District of Delaware remains very high, with approximately 40 percent of all patent cases brought and 37 percent of all defendants named in those two districts alone.
• Compared to 2011, there was an 11 percent decrease in the number of defendants named in patent cases nationwide, with over 15 percent fewer defendants named in the Eastern District of Texas and District of Delaware.
• Probably as a result of the joinder provisions of the AIA, patent plaintiffs filed 52 percent more cases in 2012, and the average number of defendants in each case dropped to less than three in nearly all the top 10 districts.
Overall, the AIA does not appear to have substantially reduced patent case concentration in the two small districts that are most popular with plaintiffs. Nevertheless, the AIA has reduced the number of defendants named in each case and may have reduced the total number of defendants somewhat. This article begins by presenting the 2012 patent case filing statistics. It then compares those numbers to case filings in 2011 and explores possible reasons for the changes, including how districts and district judges have reacted to continued venue challenges.
A. The Data Sources
Docket data on patent lawsuit filings across the United States from Jan. 1, 2012, to Dec. 31, 2012, were retrieved from PACER and compiled into a database. Pacer classifies each party in a suit as a plaintiff, defendant, or something else (e.g., intervenor, third-party plaintiff/defendant and counter/cross-claimant). The secondary classifications were filtered out and only the primary plaintiff(s) and defendant(s) were examined. For the year 2012, the dataset included information on 21,541 parties involved in 5,584 cases. Combined with previous analyses of cases between 1999 and 2011, the dataset includes information on 179,638 parties involved in 41,747 cases. 2
2 Some aspects of the Pacer dataset bear noting. First, it includes both information from the original case filing and any later added parties. Thus, the reported numbers are not a perfect reflection of the actions at the time of filing. Second, the data used in this article were retrieved from Pacer on Jan. 20, 2013, and represent a complete retrieval of all data for the year 2012. By contrast, earlier data reported for 2012 were based on data sampled periodically throughout the year. Third, the dataset only classifies parties as plaintiffs or defendants; it does not classify the parties as patentees or accused infringers. Thus, in actions seeking a declaratory judgment of noninfringement or invalidity, the parties bringing seeking declaratory relief, the accused infringers will be classified as the plaintiffs and the patent owners will be classified as the defendants. Fourth, named plaintiffs and defendants are counted each time they appear as a party. Thus, a single party that brings four lawsuits alleging patent infringement is counted as four plaintiffs in the reported data. Fifth, only cases coded as patent cases (code 830) were pulled. Cases that began as, say, copyright, trade secret, or breach of contract actions may not have been coded as patent cases. Finally, neither the Court of Federal Claims nor the International Trade Commission fully participate in Pacer, so patent cases filed there are not included.
Because of the passage of the AIA, no effort was made to identify false marking cases filed in 2012. The AIA effectively limited private false marking cases to those brought by competitors seeking only actual damages and rendered marking with an expired patent number non-actionable. See 35 U.S.C. §292(b) and (c). As a result, false marking cases have been all but eliminated, and they should not materially affect analysis of the larger trends. Slightly more than 1,000 false marking cases were filed in 2010 and 2011 and they have been excluded from the datasets for those years so that trends in traditional patent infringement cases can be analyzed.
B. 2012 Patent Filing Statistics
As shown in Table 1, the Eastern District of Texas led the nation in patent infringement case filings in 2012 based on: (a) the number of cases; (b) the number of plaintiffs; and (c) the number of defendants. In 2012, the average case in the Eastern District of Texas involved a single plaintiff suing two defendants (often corporate relatives). By comparison, in 2010, the average non-false-marking case in the Eastern District of Texas involved 13 defendants and the average case for the pre-AIA period in 2011 had 10 defendants. Noteworthy in Table 1 is the general decrease in the number of defendants per case across all districts. In 2011 (even including the period after the enactment of the AIA), 8 of the top 10 districts averaged more than 3 defendants per case. In 2012, all but one of the top 10 districts averaged fewer than 3 defendants per case. Table 2 shows the data of Table 1 as a percentage of the national totals. As an absolute number, nearly 21 percent of all the defendants named in patent cases filed in 2012 were sued in the Eastern District of Texas. By comparison, the District of Delaware accounted for 16 percent, the Central District of California accounted for nearly 12 percent, and no other district accounted for 6 percent or more. For ease of comparison, Table 3 shows the 2012 and 2011 data and calculates the percentage change year over year. Compared to 2011, the number of cases non-false-marking cases increased from 3,660 to 5,584 (52 percent), the number of plaintiffs increased from 6,119 to 8,894 (45 percent), while the number of defendants dropped from 14,201 to 12,647 (decrease of 11 percent). The decrease in the number of defendants named in the Eastern District of Texas and the District of Delaware was significant, with both districts seeing a drop of over 15 percent. The 2012 filings are generally in line with the 2011 filings in that seven of the top 10 districts in 2011 were again in the top 10 in 2012, with the Eastern District of Texas and the District of Delaware occupying the top two spots. Figure 2 depicts the combined Eastern District of Texas and District of Delaware case filings from 2000-2012 as a percentage of the national totals.
II. What Explains the 2012 Filings
A. AIA Joinder Provision
One important provision of the AIA that affects case filings relates to joinder. Before passage of the AIA, numerous defendants (including direct competitors) were frequently named in a single case based on the fact that the same patent was allegedly infringed—regardless of differences in the products accused of infringing. This practice was most pronounced in the Eastern District of Texas, where, on average, 13 defendants were named in each non-false-marking patent case filed in 2010. That figure decreased slightly in 2011 to 10 defendants per case through Sept. 16, 2011, when the president signed the AIA and the new joinder statute went into effect. The AIA generally precludes joining multiple defendants in the same lawsuit unless the same product or process is accused of infringement. Before the AIA, tens of defendants with different accused products could be joined in a single suit alleging infringement of the same patent, and that practice was especially prevalent in the Eastern District of Texas. After the AIA, the same plaintiff wishing to bring suit against the same defendants would need to bring numerous separate suits. Thus, it is not surprising that AIA joinder provisions would result in an increase in the absolute number of cases and plaintiffs, and that is what the data indicate: the number of cases increased by 52 percent, and the number of plaintiffs increased by 45 percent over the 2011 figures. In 2011, including the period after the AIA went into effect, there were 790 same-day, same-district plaintiffs in non-false-marking cases. That is, 790 times the same plaintiff had already filed a patent infringement case on that same day in that same district. By contrast, in 2012, there were 3,266 same-day, same-district plaintiffs. The difference (2,476) closely matches the increase in the number of plaintiffs between 2012 and 2011 (2,775). Thus, it appears that nearly 90 percent of the increase in plaintiffs can be attributed to multiple, same-day, same-district filings presumably resulting from the AIA’s stricter joinder provisions. Conversely, the AIA joinder provisions predictably resulted in a decrease in the number of defendants. In the pre-AIA period, the marginal cost of adding another defendant was low. Post-AIA, however, adding a new defendant requires filing another suit. Moreover, it is more likely (although not certain) that the case will be treated as a separate matter. The data suggest that the plaintiffs found that benefits from suing marginal defendants did not justify the increased burdens. As indicated above, there was a decrease of approximately 11 percent in the number of defendants named in 2012 over 2011. Of course, it is possible that there were simply fewer defendants to name in 2012 than in 2011. That seems unlikely, but 2013 may provide more information. .
B. The Eastern District of Texas
Figures 3 and 4 depict filings in the Eastern District of Texas from 2000-2012. Figure 3 reports the actual number of cases, plaintiffs and defendants, while Figure 4 reports the parties (combined plaintiffs and defendants) and cases as a percentage of national totals. As the figures show, there has actually been a reduction in both the absolute number and percentage of defendants named in the Eastern District of Texas from the 2010 peak. Nevertheless, the district is still the most popular with plaintiffs, and nearly 21 percent of all defendants were named in this single district in 2012. Courts in the Eastern District of Texas continue to issue rulings that have the effect of keeping cases in the district even in the face of legislative changes and previous mandamus rulings by the U.S. Court of Appeals for the Federal Circuit.
(i) Joinder in Pre-AIA Cases.
In the week between passage of the AIA by Congress and the President’s signature, more than 1,500 defendants were named in patent cases in an apparent effort to avoid the non-retroactive AIA joinder provisions that would take effect. Indeed, more than 700 defendants were named in the Eastern District of Texas during that week with an average 13 defendants per case. 3 Thus, while the AIA joinder provisions will guide cases going forward, a substantial volume of pre-AIA cases with large numbers of defendants remain and those cases are not governed by the AIA.
The Federal Circuit addressed joinder standards in pre-AIA cases in a mandamus decision, In re EMC Corp. 4 The plaintiff, Oasis Research L.L.C., filed suit in August 2010 naming 16 unrelated defendants and asserting that Oasis’s principal place of business was in Marshall, Texas. In November 2010, several defendants moved to dismiss, sever, and/or transfer to various districts alleging, among other things, that Oasis alleged presence in Marshall was a sham created solely to manufacture venue. In May 2011, Magistrate Judge Amos L. Mazzant issued reports and recommendations denying all the motions to dismiss. Finding that the actions were “not dramatically different,” the magistrate judge recommended denying all the severance motions and, as a result, denying transfer.
In July 2011, District Judge Michael H. Schneider adopted the magistrate’s recommendations and refused to sever or transfer. The day before Congress passed the AIA, three defendants sought a writ of mandamus, arguing “not dramatically different” was not the appropriate standard for evaluating misjoinder under Fed. R. Civ. P. 20 and that their cases should be severed and transferred to other districts. On review, the Federal Circuit confirmed that the AIA joinder provisions were not retroactive, but nevertheless held that in pre-AIA cases, independent defendants may be joined in a single case only where the accused products or processes “are the same in respects relevant to the patent” and that merely practicing the same alleged patent is not sufficient. 5 “Unless there is an actual link between the facts underlying each claim of infringement,” the Federal Circuit held, “independently developed products using differently sourced parts are not part of the same transaction, even if they are otherwise coincidentally identical.” 6
Although that standard did not exactly track the AIA’s requirement that the “same product or process” be accused, it was quite similar and far removed from the “not dramatically different” standard used by the district court. Rather than addressing severance and transfer in the first instance, the Federal Circuit remanded for the district court to reanalyze under the proper standard. 7
In some cases, Eastern District of Texas judges have simply not ruled on transfer motions for extended periods of time. For example, in Software Rights Archive L.L.C. v. Google Inc., Google filed a motion to transfer in February 2009. 8 Having received no ruling on the motion by July 2010, Google filed a petition for a writ of mandamus with the Federal Circuit seeking an order compelling the district court to rule on the motion. Within days, the district court took up the motion and granted transfer before the Federal Circuit ruled.
In SimpleAir Inc. v. AWS Convergence Technologies Inc., the defendants filed a motion to transfer in March 2010. Magistrate Judge Chad Everingham did not rule until June 2011 and then denied the motion. The defendants petitioned for a writ of mandamus, but the Federal Circuit denied relief, noting, among other things, that they “failed to employ any strategy to pressure the district court to act, such as seeking mandamus to direct the district court to rule on the motion” and also waited over three months after the transfer motion to seek writ relief. 9
Of course, defendants may be reluctant to seek mandamus directing district courts to rule on severance and venue issues out of concerns that they might antagonize the district judge who may still need to rule on the merits, and/or that the Federal Circuit may treat a petition as premature if the defendants do not wait long enough for the district court to rule. On the other hand, defendants may be encouraged by the Federal Circuit’s recent (albeit nonprecedential) decision in the second EMC mandamus petition in which the Court denied mandamus, but stressed “the importance of addressing motions to transfer at the outset of litigation,” that such motions should have “top priority,” and that “Congress’ intent ’to prevent the waste of time, energy and money and to protect litigants, witnesses and the public against unnecessary inconvenience and expense,’ … may be thwarted where … defendants must partake in years of litigation prior to a determination on a transfer motion.” 10 It is too early to tell whether that decision will result in more prompt rulings.
(iii) Consolidation of Cases for Pre-Trial Purposes and Deferring Transfer Until After Claim Construction.
One of the most remarkable decisions of 2012 occurred in Norman IP Holdings L.L.C. v. Lexmark International Inc. 11 The day before the AIA was signed into law (Sept. 15, 2011), Norman IP filed one lawsuit in the Eastern District of Texas naming two defendants. Norman then filed a series of amended complaints in that suit naming additional, unrelated defendants so that by January 2012, 18 defendants had been named.
In August 2012, Chief District Judge Leonard E. Davis issued an order addressing numerous motions to dismiss or transfer claims against various defendants for misjoinder and convenience. In light of EMC, the court severed the cases against the various defendants. Nevertheless, the court further ruled that the cases would be consolidated for all pretrial purposes other than venue determination and that any transfer of venue would not take effect until after the court issued its claim constructions. Further the court stated: This Court has limited resources and constantly strives to employ efficient and cost-saving case-management procedures for the benefit of the parties, counsel, and the Court. In response to the AIA’s joinder provision, plaintiffs now serially file multiple single-defendant (or defendant group) cases involving the same underlying patents. This presents administrative challenges for the Court and, left unchecked, wastes judicial resources by requiring common issues to be addressed individually for each case. For example, what was once a single motion to substitute parties (or join a plaintiff) becomes multiple motions. These must each be processed by the Court and staff, including review of the underlying motions and docketing individual orders addressing each motion. More substantive motions, particularly where the same arguments are used in each individual case, present even more difficulties. There, the Court is required to waste time digesting duplicate arguments to ensure that new arguments are not hidden among the plethora of common arguments. … Thus, to permit efficient case management, the Court ORDERS these newly severed actions consolidated with the original filed case as to all issues, except venue, through pretrial only.
However, this case management approach should not be perceived as an invitation to file motions to transfer venue. In recent years, this Court has expended considerable time addressing venue. This Court has many, many, issues before it—both criminal and civil—and it carries one of the heaviest patent dockets in the country; yet, venue in patent cases has increasingly become an extremely expensive and time-consuming matter, not only for the Court but for the parties as well. For instance, in a recent set of serially filed cases involving only seven defendant groups, the parties had already expended over $700,000 on venue-related discovery and briefing before the cases were even ready for status conference (i.e., before all defendants had answered the complaint). This Court currently has approximately forty pending motions to transfer venue. If the average costs of discovery and briefing for each of these transfer motions is only $300,000, then approximately $12 million is being spent by the parties on an issue that does not move the ball down the field, but only seeks a new field upon which to play. Finally, some parties have even called courts in this district to essentially “threaten” mandamus if a venue ruling is not issued within the timeframe desired by the parties. This Court manages a very busy docket—as do all courts in this district—with pending motions of varying levels of priority. Criminal cases take first priority because individuals’ freedom is at stake. In the patent context, trials and Markman hearings are a high priority. Venue motions are important, but not any more important than everything else this court has to do. The court rules on these motions as soon as it can. The Civil Justice Reform Act of 1990 was instituted to provide checks on long-pending motions and cases. Courts in this district take their cases seriously and strive to timely address pending motions in an effort to resolve cases promptly. This Court will address motions to transfer venue as timely as possible, while balancing the many other issues unrelated to venue requiring the court’s attention. 12
Whether other district judges will follow this lead, and whether the Federal Circuit will endorse this approach, remains to be seen. 13
iv) Denials of Writ Relief by the Federal Circuit.
Although the Federal Circuit has granted mandamus to force transfers of venue in some cases, it has made clear that such relief will be granted only in extreme cases where it considers a district court’s ruling to be beyond the pale of reasonableness. For example, in In re Amazon.com Inc., the Federal Circuit refused to order transfer to a district (the Western District of Texas) where none of the defendants was headquartered. According to the panel, the fact that none of the defendants was headquartered in the proposed forum made the situation significantly different from other cases where mandamus was granted. 14
Mandamus to transfer was also denied in In re Vicor Corp.. There, both the district court and the Federal Circuit focused on the fact that an earlier case that involved the defendants’ products (albeit different defendants) had been litigated in the Eastern District of Texas (albeit before a different judge). The newly assigned judge (Schneider) retained jurisdiction over that case for post-injunction damages and contempt issues. In the Federal Circuit’s view, the petitioner did not make a compelling showing that there was insufficient overlap between the earlier case and the current one such that judicial economy would not be gained. 15
Writ relief was also denied in In re Fusion-Io Inc. There, the plaintiff filed suit nine days before the AIA was signed, naming nine unrelated defendants. One of the defendants (Fusion-Io) filed a motion to sever and transfer in January 2012, before answering. In September 2012, Judge James Rodney Gilstrap severed the parties and denied the motions to transfer without prejudice, indicating that the defendants could file new motions to transfer that only addressed their severed case. Fusion-Io first sought reconsideration of denial of its transfer motion in light of the fact that its original motion filed in January had addressed only the circumstances of its case and re-filing the same motion again would merely lead to delay. After the district court denied reconsideration, Fusion-Io sought a writ. The Federal Circuit denied relief, indicating that it would not weigh the facts on transfer before the district court. Nevertheless, the Federal Circuit suggested that Fusion-Io promptly re-file its motion to transfer with the district court along with a motion to stay and indicated that the district court should take up those motions before proceeding to any motion on the merits of the case. 16
The Federal Circuit also refused to order transfer in In re HTC Corp. There again, the district court and the Federal Circuit focused on the fact that the proposed transfer district (the Northern District of California) was not the headquarters district of either of the parties (although it did contain the headquarters for a third party, Google, whose relevance to the case was disputed). HTC’s relatively recent change of incorporation from Texas to Washington and refusal to waive privilege to disclose the reasons for that move were also noted. 17
v) Judge Selections by Plaintiffs.
As noted in a previous article, the Eastern District of Texas allows plaintiffs to influence which district judge is assigned to their case because case assignments are allocated based on the division where the case was filed. 18 The Eastern District has several divisions, six of which are relevant to civil cases: Beaumont (division 1); Marshall (division 2); Sherman (division 4); Texarkana (division 5); Tyler (division 6) and Lufkin (division 9). With the confirmation of Judge Gilstrap to replace Judge T. John Ward, case assignment practices were returned to those in place before Ward’s resignation. After a further general order dated Jan. 17, 2012, cases were assigned as follows:
• Beaumont—100 percent of patent cases to Judge Ron Clark;
• Marshall—75 percent of civil cases to Judge Gilstrap and 25 percent to Judge Schneider;
• Sherman—50 percent of civil cases to Judge Richard A. Schell and 50 percent to Judge Marcia A. Crone;
• Texarkana—90 percent of all cases to Schneider and 10 percent to Gilstrap;
• Tyler—95 percent of patent cases to Chief Judge Davis and 5 percent of patent cases to Schneider; and
Table 4 shows the breakdown of cases and parties by division filing. As shown in the table, patent plaintiffs appear to have particular preferences for Davis and Gilstrap.
C. The District of Delaware
Figures 5 and 6 show patent case filings in the District of Delaware from 2000-2012. Figure 5 reports the actual number of cases, plaintiffs and defendants, while Figure 6 reports the parties (combined plaintiffs and defendants) and cases as a percentage of the national totals. As the figures show, both the absolute number and percentage of defendants named in Delaware declined in 2012 from the 2011 peak. Nevertheless, the district is still the second popular district with plaintiffs, and more than 16 percent of all defendants were named in this single district in 2012. . As in the Eastern District of Texas, courts in Delaware continue to issue rulings that have the affect of keeping cases in the district. At the close of 2011, in In re Link_A_Media Devices Corp., 19 the Federal Circuit issued a writ of mandamus ordering a case transferred from the District of Delaware. At issue in that case was the focus of District Court Judge Sue L. Robinson on the fact that the defendants were incorporated in Delaware. Although the dispute otherwise had little or no connection with Delaware, the district court denied a motion to transfer holding: “… because all [the] defendants are incorporated in Delaware, they have no reason to complain about being sued in Delaware.”
On the writ of mandamus, the Federal Circuit ordered the case transferred, noting that place of incorporation is not a factor listed for consideration in either 28 U.S.C. §1404 or Third Circuit case law. In the Federal Circuit’s view, the place of incorporation should not have been given “dispositive” weight, and placing “heavy reliance” on place of incorporation was inappropriate. 20
If place of incorporation were removed as a consideration in transfer motions, it would be expected that many more cases should be subject to transfer from the District of Delaware because few companies have significant operations in Delaware. Nevertheless, Delaware district courts have distinguished Link_A_Media and the Federal Circuit has since refused to overturn a ruling denying transfer in a case where the district court continued to rely, in part, on the defendants’ Delaware incorporation. In Intellectual Ventures I L.L.C. v. Altera Corp, all the plaintiffs and the defendants were Delaware corporations with headquarters on the west coast (Washington and California) and none had offices or employees in Delaware. Further, the named inventors, prosecuting attorneys, and a number of potential witnesses resided in California. The defendants sought to transfer the case to the Northern District of California, where most of the defendants’ accused products were developed and where the plaintiff also had an office. Instead of the “heavy reliance” disapproved of in Link_A_Media, Delaware District Court Judge Leonard P. Stark placed “substantial weight” on place of incorporation and found that the convenience of the parties favored transfer, “but only slightly.” With regard to the convenience for the witnesses, the court first found that the location of party witnesses carried no weight in the “balance of convenience” because parties are required to produce witnesses wherever the case is pending. With regard to non-party witnesses (e,g., the inventors, prosecuting attorneys, or likely third-party witnesses), none of whom resided in Delaware or were within the subpoena power of the Delaware court, the court held that the relevant inquiry was only whether the witnesses were within the subpoena power of the court for trial. Thereafter, the court held that the fact that the non-party witnesses were beyond the subpoena power of the court in Delaware for trial (but many would be within the subpoena power of the Northern District of California) favored transfer but was entitled to “little weight” because, in the court’s view, it would be “statistically rare” for a case to go to trial and witnesses could be presented by deposition in any event. The fact that none of the relevant evidence was in Delaware also favored transfer but was given “little weight” because of alleged technological advances. Weighing all the factors, the court found that the defendants had not shown that the factors strongly favored transfer. 21
On a petition for mandamus, the Federal Circuit agreed that the place of incorporation could be considered and distinguished Link_A_Media, stating: Although the circumstances here are in certain respects similar to those in the petition we granted in Link_A_Media, that precedent did not go so far as to limit the trial court’s discretion to deny transfer in this case. The trial court in Link_A_Media disposed of the transfer motion based on the plaintiff’s forum preference and the fact that the defendant had incorporated in Delaware. In doing so, the court failed to give any consideration to the fact that transfer would significantly minimize the travel and cost to the identified witnesses and move trial to where the accused products were developed. Because the court viewed those considerations as entirely superfluous, its error could not have been more clear. On that view, this case is clearly distinguishable; in its thorough opinion, the district court endeavored to evaluate each of the forum non conveniens factors in light of the same arguments raised in the petition, and there is no clear indication that the court failed to meaningfully consider the merits of the transfer motion. Moreover, … in this case, unlike Link_A_Media, there are rational grounds for denying transfer given that all of the parties (not just a single defendant) had incorporated in Delaware and some witnesses would potentially find Delaware more convenient. 22
With this road map, transfer out of Delaware may be difficult in all but extreme cases.