THIS PAPER EXPLORES recent developments and changes in the law of removal and remand. Recent is, of course, a relative term. Although most attorneys regularly practice in federal court, it is not unusual to go for an extended period of time, perhaps even years, between removing a case from state court or seeking to remand one. In this context, therefore, recent encompasses the last few years, with emphasis on the changes made in removal and remand procedure by the Judicial Improvements and Access to Justice Act of 1988 (the Judicial Improvements Act). [FN1] In addition, some common removal and remand traps are discussed.

Sections 1441 through 1452 of the Judicial Code [FN2] contain the great majority of removal and remand law. In addition to those sections of the Judicial Code, certain *396 specialized statutory schemes, such as the Financial Institutions Reform, Recovery and Enforcement Act of 1989 (FIRREA), [FN3] the Securities Act of 1933, [FN4] and the Federal Employer's Liability Act (FELA) [FN5] also provide removal and remand provisions.

Section 1441 contains the four general categories of removable actions. [FN6] First, section 1441(a) provides for the removal of an action where the federal district court would have diversity jurisdiction. Second, section 1441(b) provides for removal when the federal district court has federal question jurisdiction. Third, section 1441(c) provides that “separate and independent” claims or causes of action which, if sued upon alone would be removable, may be removed even if joined with non-removable claims or causes of actions. Fourth, Section 1441(d) provides that a foreign state which qualifies as such under the Foreign Sovereign Immunities Act may remove an action from state to federal court. In addition to Section 1441 which sets out the types of removable actions, Section 1445 makes certain actions non-removable.

Diversity of citizenship must exist both at the time the original action was filed and at the time of removal. [FN7] When a corporation is a party to a diversity suit, the question of corporate citizenship can sometimes be perplexing. The general citizenship rule is easily stated: a *397 corporation is a citizen of its state of incorporation and the state wherein its principal place of business is located. However, identifying the principal place of business is not always simple. The Fifth Circuit embraced a test for determining a corporation's principal place of business characterized as the total activity test. [FN8] The Fifth Circuit test is really a combination of the two other principal place of business tests which courts employ: nerve center test and place of operations test. In the Fifth Circuit's view, the total activity test considers both “the general rules of the ‘place of activities' test and the ‘nerve center’ test in the light of the particular circumstances of a corporation's organization” in order to “balance the facts to determine the ultimate question: the location of the corporation's principal place of business.” [FN9]

Generally speaking, the nerve center test is a back-up test used when a corporation's operations are so far-flung that it is extremely difficult to pick one particular place of operation which can be said to be the primary one. [FN10] Thus, in a situation where there is a single place of operation and a headquarters located in another state, the principal place of business is under the total activity test will likely be the place of operations. The Fifth Circuit court also examined other situations where more significant activity at the corporate headquarters might outweigh actual operations. [FN11] The analysis operates on a sliding scale, which is often difficult to gauge with any degree of certainty, and without significant bright-line rules.

An interesting Ninth Circuit case [FN12] recently examined the nerve center and place of operations tests, and held that the place of operations test prevails over the nerve center test when the two conflict. [FN13] In Industrial Tectonics, the Ninth Circuit was faced with a case where a company *398 had the majority (around 60%) of its operations in California, yet had a very substantial minority of its operations (about 40%) and its corporate headquarters, its nerve center, in Michigan. This scenario is the most difficult one in deciding where a corporation's principal place of business is located. If a corporation's operations were evenly split between two states, yet one state also contained the corporation's “brain,” it seems that under any test the balance should be struck where half of the operations plus the headquarters are located. Thus, the question put to the Ninth Circuit was, essentially, just how much of the operations must be located in one state such that it outweighs a substantial minority of operations plus the existence of the nerve center located in a different state.

Apparently, the Ninth Circuit found that a simple majority of operations was sufficient to outweigh the existence of the nerve center in another state, even if that nerve center was coupled with a substantial minority of the corporation's operations. The Ninth Circuit specifically held “where a majority of a corporation's business activity takes place in one state, that state is the corporation's principal place of business, even if the corporate headquarters are located in a different state.” [FN14] The Ninth Circuit's approach seems to embrace at least this limited bright-line rule. In this area, the more bright-lines the better. What a terrible waste it would have been if J.A. Olson Co. and Industrial Tectonics had been post-trial cases which the appellate courts felt bound to dismiss for lack of subject matter jurisdiction.

Fraudulent joinder is a means of destroying diversity of citizenship requirements in removal actions. [FN15] Until recently,*399 the Fifth Circuit's seminal case on fraudulent joinder has been B., Inc. v. Miller Brewing Co.. [FN16] Now, however, the Fifth Circuit may be taking a different position on the fraudulent joinder question with a recent decision out of the Eastern District of Louisiana. [FN17]

B., Inc. stands for the proposition that the removing parties claiming fraudulent joinder have a heavy burden to show that there is no possibility that a valid state law claim can be stated against defendants allegedly fraudulently joined. [FN18] B., Inc. also states that current state law is not necessarily determinative of the fraudulent joinder issue, especially where the applicable state law is uncertain. [FN19] The Court found that even an arguably reasonable basis for predicting that state law might impose liability on the facts involved suffices to make joinder non-fraudulent. [FN20] For determining whether joinder is fraudulent, the B., Inc. court approved a procedure which allows the district court to consider not only the plaintiff's pleadings, but the affidavits and depositions of all parties. [FN21] In fact, the B., Inc. court analogized the scope of the record which could be considered in resolving the fraudulent joinder question to summary judgment practice under Federal Rule of Civil Procedure 56(b) . [FN22]

The Fifth Circuit's recent Carriere case, while relying upon B., Inc., seems to have gone far beyond the procedural limitations that B., Inc. imposed. The Carriere court has taken the summary judgment allusion in B., Inc. and applied it quite literally. [FN23] Where the B., Inc. court allowed*400 evidence outside of the pleadings to be considered in deciding the fraudulent joinder question, the panel limited the use of that evidence to deciding whether the plaintiff succeeded in stating some possible claim under state law against the supposedly fraudulent defendant. [FN24] In other words, the B., Inc court contemplated consideration by the district court of the pleadings and the other fraudulent joinder evidence merely to determine whether a cause of action has been or could possibly be stated on those facts. The B., Inc. court did not endorse a procedure where pleadings and fraudulent joinder evidence are examined for the very different purpose of weighing whether the plaintiff could ultimately recover against the defendant on the facts presented.

That the Carriere court has gone beyond B., Inc. and has blended the propriety of summary judgment on the merits with the fraudulent joinder question is indisputable. This represents a major alteration, and a singular anomaly, in fraudulent joinder jurisprudence. The Carriere court specifically stated that “[b]ecause we have already concluded that Sizeler was fraudulently joined, we need not consider appellant's argument on this point [the propriety of summary judgment] further. Summary judgment will always be appropriate in favor of a defendant against whom there is no possibility of recovery.” [FN25]

The Carriere court's acknowledgment that the state court petition alleged just what it needed to allege under Louisiana law to state a cause of action against one defendant bolsters the fact that the approach has changed. [FN26] The court, although acknowledging the petition, considered only the affidavits and found that the plaintiffs “could not possibly recover” from that defendant. [FN27] As to another supposedly fraudulently joined defendant, the plaintiff apparently had an argument for the good faith *401 extension of current Louisiana premises liability law to allow a cause of action against this other defendant. [FN28] Yet, the Carriere court considered only the current state of Louisiana law on this point and found fraudulent joinder. [FN29]

The difference in procedure between piercing the pleadings and considering other evidence to determine whether a colorable claim has been stated (the proper analysis), versus a full-blown summary judgment type inquiry (the improper analysis), was recognized by a recent Third Circuit case. [FN30] The Third Circuit's position is truer to B., Inc., than is the Fifth Circuit's own Carriere case. It will be interesting to note whether future Fifth Circuit cases will expand upon Carriere and make fraudulent joinder removal easier or whether it will retreat to the B., Inc. position.

The Judicial Improvement Act worked several changes in removal procedure. Section 1016(a) of the Judicial Improvements Act added a new sentence to 28 U.S.C. § 1441(a) , instructing that for purposes of removal, citizenship of defendants sued “under fictitious names” are to be ignored. [FN31] This section, of course, is a congressional reversal of Bryant v. Ford Motor Co. [FN32] and Coker v. Amoco Oil Co. [FN33] which made the presence of “Doe” defendants destructive of diversity.

The affirmative defense of federal preemption of the plaintiff's state cause of action does not constitute a federal question for the purpose of federal jurisdiction and removal. [FN35] However, since the Supreme Court's decision in Metropolitan Life Insurance Co. v. Taylor, [FN36] an exception known as the complete preemption doctrine holds that if certain federal statutory schemes completely fill a field of law, there is no longer any “room” for state causes of action. [FN37] Therefore, any attempt to plead a state cause of action in such an area is merely a disguised way of pleading a federal cause of action. As thus recharacterized, the case becomes removable to federal court as a federal question. It is most common for the complete preemption doctrine to arise in either the Employee Retirement Income Security Act (ERISA) [FN38] context, [FN39] or where the action involves a matter covered by a collective bargaining agreement under federal labor law. [FN40]

*403 A recent development in this area, one significant to aviation law, was the Fifth Circuit's decision in Trans World Airlines, Inc. v. Mattox. [FN41] In TWA, the Texas attorney general filed suit in Texas state court against Pan American World Airways alleging deceptive trade practices under state law in connection with airfare advertising. The Fifth Circuit held, after a careful analysis of Section 105(a)(1) of the Deregulation Act [FN42] and case law thereunder, [FN43] that Congress intended to occupy the field of alleged deceptive airline fare advertising so that state laws regulating that subject matter are entirely excluded. [FN44] Therefore, a facially state law cause of action alleging deception in airline pricing can be removed to federal court. [FN45]

The derivative removal concept, which held sway for decades, stated that when a plaintiff brought a claim in *404 state court over which the federal courts had exclusive jurisdiction, that case could not be removed, as the federal courts' removal jurisdiction was said to be “derivative” of that of the state court. Therefore, the argument went, because the state court did not have jurisdiction over the case, the federal court could acquire none. [FN46] In 1986, Congress abrogated the concept of “derivative removal” by enacting 28 U.S.C. § 1441(e) .

Section 1441(c) [FN47] allows removal of “separate and independent” causes of action which are within the jurisdiction of the federal district court, even if those causes of action are joined with other causes of action outside the federal court's jurisdiction. [FN48] Federal courts have a strict view of what types of claims are “separate and independent” and removals on this ground are seldom successful. [FN49]

Recently, section 1441(c) was significantly amended by Title III of the Judicial Improvements Act of 1990. [FN50] Section 312 of this Act eliminated the removal of separate and independent diversity claims, only allowing removal of separate and independent federal question claims. Section 1441(c) also empowers the federal court to hear the entire case or remand matters in which state law predominates.

The most interesting question surrounding removal *405 under the foreign sovereign removal provision, section 1441(d) , [FN51] involves the treatment of co-defendants who are not covered foreign sovereigns. In 1980, the Fifth Circuit held that removal by a foreign sovereign removed the entire case against all defendants, including domestic co-defendants. [FN52] The Fifth Circuit's decision in Arango occurred, however, prior to the Supreme Court's recent landmark decision on pendent-party jurisdiction. [FN53] After Finley, the Ninth Circuit held that a multiple defendant removal under the Foreign Sovereign Immunities Act and section 1441(d) was proper. [FN54] Although the Ninth Circuit did not discuss Finley by name, the court's analysis certainly parallels its analysis of the propriety of pendent-party jurisdiction. [FN55] The Ninth Circuit opinion clearly implies that section 1441(d) and the Foreign Sovereign Immunities Act were intended by Congress to provide for pendent-party jurisdiction. [FN56]

A thorough discussion of the complex statutory and constitutional questions surrounding pendent-party jurisdiction is beyond the scope of this paper. [FN57] Pendent-party jurisdiction is, however, relevant to questions of removal and remand not only in the context of the Foreign Sovereign Immunities Act, [FN58] but also in the context of separate *406 and independent claim removal under section 1441(c) . After Finley, two federal district courts have discussed whether section 1441(c) is itself a statutory basis for the exercise of post-removal pendent-party jurisdiction. Furthermore, section 310 of the 1990 Act has both expanded and contracted the scope of pendant-party jurisdiction.

The northern district of Ohio has held that section 1441(c) is not a statutory basis for allowing pendent-party jurisdiction. [FN59] In Perkins, a non-diverse defendant had pendent state law claims made against him in a case involving a Title VII claim against other defendants. After removal, the court held that even if the claims against the non-diverse pendent-claim defendant were “separate and independent,” jurisdiction could not be exercised because there was no independent basis for federal jurisdiction over that defendant. [FN60] Jurisdiction was improper since Congress, under section 1441(c) , did not grant pendent-jurisdiction to the federal courts. [FN61]

The northern district of Illinois, however, allowed an exercise of pendent-party jurisdiction under section 1441(c) . [FN62] The pendent party in Carter, as a co-plaintiff rather than a co-defendant, brought a section 1983 claim for wrongful arrest. The claimants were husband and wife. The wife, as co-plaintiff, also sued the same defendants under state law for loss of consortium. [FN63] The Carter court recognized the existence of Finley, but followed a previous Seventh Circuit analysis that held pendent-party jurisdiction is proper, unless Congress expressly prohibited*407 the action. [FN64] Perkins justifiably criticizes Carter for adopting authority which was superseded by Finley. [FN65] The Ohio court probably has the better argument.

Although not cited in support of its position, the Perkins court may have relied on section 1447(e) of the Judicial Code, recently added by the Judicial Improvements Act. [FN66] Instead of supporting pendent-party jurisdiction in the removal context, section 1447(e) shows a clear congressional intent to forbid pendent-party jurisdiction in the removal context. [FN67]

The amended section 1447(e) addressed situations where, after removal, “the plaintiff seeks to join additional defendants whose joinder prior to removal would have destroyed subject matter jurisdiction.” [FN68] Congress has yet to take a specific position on “pendent party” jurisdiction in the removal context. Section 1447(e) provides that, in the post-removal pendent party situation, the “court may deny joinder, or permit joinder and remand the action to the State court.” [FN69] Note, however, that Congress forbids joinder of these “pendent parties” unless the case is then remanded to state court. [FN70]

The 1990 Act established a new basis for federal subject-matter jurisdiction, referred to as Supplemental Jurisdiction. [FN71] By enacting section 1367, Congress responded to the Supreme Court's invitation in Finley to establish some boundaries for the exercise of pendent-party jurisdiction. Section 1367(a) grants the federal courts original jurisdiction over pendent parties (and claims), in federal *409 question cases, to the constitutional limits of Article III. [FN72] Section 1367(b) disallows pendent-party jurisdiction in cases founded solely on diversity jurisdiction. [FN73]

Section 1445(c) provides that any action in state court under the worker's compensation laws may not be removed. [FN74] Certain other federal statutes also provide for a similar “plaintiff's privilege” to choose which court system will hear the case. Securities cases brought under the 1933 Act and FELA cases are good examples of this genre of cases. [FN75]

The most interesting question under section 1445(c) is how to characterize a wrongful discharge cause of action brought after an alleged firing or filing of a worker's compensation claim. Section 1445(c) has been held by federal district courts in Texas to make such claims non-removable. [FN76] One court has held that a direct suit against a worker's compensation insurance carrier for breach of the duty of good faith and fair dealing is non-removable under section 1445(c). [FN77] Another court held that where the worker's compensation claim is joined in the state court with a claim presenting a federal question, the case can be removed because the worker's compensation claim *410 is pendent to the federal question. [FN78]

Questions still remain concerning just who may remove a case to federal court. Defendants can clearly remove, unless they are diverse defendants sued in their own state's courts. [FN80] The Fifth Circuit, however, embraces the minority position which also allows third-party defendants to remove. [FN81] At the same time, a federal court in Dallas held that an intervenor may not remove a case from state court. [FN82]

In addition, interesting questions about the proper state court from which a case may be removed have recently been addressed under certain special statutes. Particularly, under FIRREA, the Federal Deposit Insurance Corporation (FDIC) may remove a case to federal district court even when the case is pending in a state court of appeals. [FN83] Moreover, once the case arrives in federal court, all orders entered by the state court remain in effect unless the federal court chooses to vacate them. [FN84] After removal the federal court may then rely completely on the state court record to justify the order in question, so that de novo examination of the evidence or facts is not required. [FN85] The propriety of the now “federalized” state court order, however, must be judged under the closest analogous federal, not state, rules. [FN86]

Section 1016(b)(1) of the Judicial Improvements Act changed the name of the document which affects removal. “Notice of Removal” has replaced “Petition for Removal.” [FN87] Another formal requirement, that of verifying the Petition of Removal, now the Notice of Removal, has been eliminated. The Notice, however, is subject to rule 11 of the Federal Rules of Civil Procedure . [FN88] Furthermore, a removal bond is no longer required. [FN89]

When removing a case, certain formal modes about which “the law demands strict adherence” apply. [FN90] One is *412 the necessary allegations of corporate citizenship. [FN91] A corporation must state both its place of incorporation and its principal place of business to affirmatively allege citizenship. [FN92] It is not sufficient to aver that a corporation does not have its principal place of business in a state in which the opposite party claims its citizenship, so that diversity is established “argumentatively or by mere inference.” [FN93] On the contrary, jurisdictional allegations must be “clear, distinct and precise.” [FN94] Also, when alleging citizenship, diversity is not established by averring “residency,” since “residency” is not equivalent to “citizenship” for purposes of diversity jurisdiction. [FN95]

An interesting timing question is presented where there are multiple defendants in a removable case and they are served at different times. Several logical options exist. First, removal might be allowed if each defendant files a Notice of Removal, or joins in one, [FN96] within thirty days of being served. Second, removal may only be allowed within the thirty day period beginning when the first defendant*413 is served. Each served defendant must have either filed a Notice of Removal, or have joined in one, by the expiration of thirty days after service on the first-served defendant. The Fifth Circuit has opted for the latter approach. [FN97] To alleviate some of the potential abuse inherent in this rule, the Fifth Circuit allows, in “exceptional circumstances,” removal “when a later-joined defendant petitions more than precisely thirty days after the first defendant is served.” [FN98]

When a case is not initially removable, but later becomes so because of some procedural act of the plaintiff, questions can arise about precisely when the case became removable. The answer to this timing question is a matter of state law. For example, in Recognition Equipment, [FN99] a federal court faced the question of precisely when a case became removable. The court held that because parties were added to a state court amended petition without leave of court, in violation of Texas Rules of Civil Procedure Rule 41 , the amended petition did not suffice to begin the removal time-clock running. [FN100] The time did not begin to run until leave was granted by the state court. [FN101]

In a significant new statute, section 1016(b) of the Judicial Improvements Act imposes new and additional time limitations on removals. [FN102] Under prior law, any diversity or federal question case could be removed from state court at virtually any time if the case was removed within thirty days after first becoming removable. [FN103] This was *414 true even if the case first becomes removable years after it was filed. [FN104] Section 1446(b) as amended now contains a “statute of limitations” for removal, but for diversity actions only. [FN105] Section 1446(b) now provides that a diversity action may not be removed more than one year after the action is commenced. [FN106]

Another significant procedural change was made by section 1016(c) of the Judicial Improvements Act. [FN107] Section 1447(c) was amended to limit the time within which certain Motions to Remand can be filed. As amended, section 1447(c) provides that a Motion to Remand based on “any defect in removal procedure” [FN108] must be made within thirty days after the filing of the Notice of Removal. [FN109]

An interesting question arises under the new thirty day limitation on Motions to Remand and involves whether, in light of the amendment, a court retains the power to remand a case sua sponte for defects in removal procedure. *415 The resolution to this question pits the Third Circuit against the Northern District of Texas. [FN110]

One issue in this area is whether a sua sponte remand is, or is not, equivalent to a motion to remand, albeit the court's own motion. The Third Circuit treated the two as substantially similar and held that after the thirty day limit, the court had no power to remand the case because of removal defects. [FN111] The northern district of Texas reached the opposite conclusion. [FN112] Yet, this whole discussion of whether a “motion” is at issue seems to miss the point. The focus should not be on the first sentence of section 1447(c): “A motion to remand the case on the basis of any defect in removal procedure must be made within thirty days after the filing of the notice of removal under section 1446(a)” [FN113] ; but upon the second sentence, which states: “If at any time before final judgment it appears that the district court lacks subject matter jurisdiction, the case shall be remanded.” [FN114]

Given the present statutory scheme and the controlling case law, [FN115] the analysis must focus on whether specific statutory authorization exists to remand a case. Certainly, one ground for removal could have been an untimely removal. Removal, however, is untimely only if the complaint is not made within the statutorily authorized time limits. [FN116] After the time limit passes, the statute no longer authorizes remand on the basis of that defect. When authorization expires, the only ground for remand (whether by motion or sua sponte) contemplated by section 1447(c) *416 is lack of subject matter jurisdiction. [FN117] Therefore, because the court only has jurisdiction when a procedural defect is at issue, the court cannot remand the case on its own motion. [FN118]

A recent Fifth Circuit case bolsters this decision when it succinctly states that “a court may consider remand only if the parties raise the issue; conversely, a court must consider the existence of subject matter jurisdiction on its own motion.” [FN119] Therefore, the result reached by the Third Circuit in Air Shields is correct; after the thirty day period for filing procedural remand motions, a court cannot remand a case to state court sua sponte for defects in removal procedure. This conclusion is surely the correct one; although FDIC v. Lloyd remains pending on appeal, [FN120] the Fifth Circuit has clearly telegraphed its opinion on the propriety of a sua sponte post-thirty day remand in the recent significant decisions styled In re Shell Oil Co. [FN121]

In Shell Oil I the district court granted an untimely motion to remand based upon the presence in the action of a forum state defendant. On writ of mandamus, the Fifth Circuit first addressed whether the presence of a forum state defendant in a removed action was a defect in “removal procedure.” [FN122] The court defined “procedural” defects to encompass every sort of defect which was not a defect in subject matter jurisdiction. Thus, because Article III jurisdiction exists even in the absence of complete *417 diversity, the presence of the forum state defendant was found to be a procedural defect. [FN123]

Significantly in the Shell Oil I discussion of the procedural/jurisdictional dichotomy, the Third Circuit's holding in Air-Shields, [FN124] was discussed and cited with approval, with the Fifth Circuit specifically stating that, by moving to remand past the thirty-day limit, “plaintiffs have waived any non-jurisdictional grounds for remand existing at the time of removal.” [FN125] This is significant because once the right to remand is waived, it is inconsistent with removal and remand statutes for the district court to unilaterally revive the right and exercise it on behalf of a party. Therefore, in Shell Oil I, the Fifth Circuit issued a writ of mandamus to vacate the remand order entered by the district court. Certainly, if the district court had the power to remand for procedural defects after the 30-day limit, the fifth circuit would not have issued the writ of mandamus merely because the party's remand motion was tardy. Tardiness under the circumstances would be a wholly harmless error if the district court had the power to act on its own and remand the action on the same procedural basis at any time.

Fascinating questions arise when the new statutes of limitations on removal and remand collide. The problem is as follows: a defendant removes the case more than one year after filing, but the plaintiff fails to move to remand the case within the thirty day limit. The statutory question is, of course, whether removal outside of the one-year limit is a procedural defect. The few courts which have addressed this situation have reached diametrically opposite results.

Without discussion, the district court for the northern *418 district of California held that a plaintiff waives any complaint about a removal outside of the one-year limit if a Motion to Remand is not made within thirty days of the Notice of Removal. [FN126] The eastern district of Wisconsin, after a thorough analysis, reached the same result. [FN127] The district court for the northern district of Illinois ruled the opposite way after analyzing the language of the statute and its legislative history. The Illinois court held that the one-year limitation is jurisdictional and is not waived by failure to move for remand within thirty days. [FN128]

The Foiles opinion appears to be incorrect while Gray and Leidolf are correct. Although the Foiles court is right when it points out that section 1446(b) states clearly and emphatically that a diversity case may not be removed after one year, [FN129] the significance of this language vis-a-vis a waiver argument is problematic. A statute of limitations, when pleaded and proven, completely extinguishes a cause of action on the merits. Yet even this draconian advantage may be waived by someone entitled to assert it.

The Federal Rules of Civil Procedure specifically mention statutes of limitations in rule 8(c) as an affirmative defense which must be pleaded. [FN130] Certainly, a time limitation on removal should be entitled to no more sanctity than are statutes of limitations on the merits of claims. Furthermore, the Foiles court's alternative, making the one-year limitation on removal jurisdictional and not subject to waiver, is wholly unjustified given that the similar thirty day limitation on removing any then-removable action [FN131] is clearly non-jurisdictional and subject to waiver. [FN132]

*419 The Foiles court also ignored the fact that the one-year removal limitation was placed in section 1446 which is solely concerned with removal procedure. Although this “location” was noted, the court stated that “Congress is not required ... to write and amend its statutes with the kind of semantic exactitude valued by lawyers.” [FN133] Perhaps if section 1446 were the only removal statute, the court would have a point. The court simply ignored the fact that if Congress indeed meant for the one-year removal limitation to be jurisdictional, the limitation would have been placed in section 1445 which discusses non-removable actions, or at least in section 1441(a) which covers the removability of diversity actions.

Ordinarily, appellate relief from an order of remand is unavailable. [FN134] The exceptions to this general rule are: (1) removals under the civil rights removal provision of section 1443; [FN135] (2) where the remand is made on grounds not authorized by section 1447(c); [FN136] and (3) where the FDIC is appealing under the FIRREA. [FN137] In addition, even when appellate relief of the merits of the remand are unavailable, the court of appeals may still entertain an appeal from any order awarding sanctions in connection with the remand of the case. [FN138]

*420 In what is probably the most important decision in the area of appellate relief from remand orders since the Supreme Court's Thermtron decision in 1976, the Fifth Circuit recently expanded reviewability in Shell Oil I. [FN139] Prior to its modification by the 1988 Act, section 1447(d), as interpreted by the Thermtron Court, prohibited appellate review of any remand order based upon section 1447(c). Old section 1447(c) mandated remand in cases of “improvident” removal or lack of jurisdiction. [FN140] The 1988 Act changed section 1447(c) by dropping the “improvident removal” language. The Fifth Circuit held that this amendment opened the door to appellate review of certain orders of remand which are based upon defects in removal procedure, as opposed to those based upon lack of subject matter jurisdiction. [FN141]

The writs of mandamus sought in the Shell Oil cases were from remand orders which were filed outside the thirty-day limit. The Fifth Circuit expressly reserved the question of whether appellate relief was available in situations where a non-jurisdiction based remand order followed a timely remand motion. [FN142] The court did say, however, that it is “arguable” that appellate relief would be unavailable in such situations. [FN143] Given the strong language limiting the reviewability of remand orders in section 1447(d), and the limited nature of the original Thermton exception thereto, it is likely that remand orders based upon procedural defects, which follow timely motions to remand, will be non-reviewable.

Although one can question the fairness of section 1447(d)'s limitations on review, interpreting that section to allow reviewability of remand orders based upon procedural defect, following timely remand motions, would be improper. The prior version of section 1447(c) specified*421 two types of remands which were beyond appellate jurisdiction: orders based upon “improvident” removals and orders based upon lack of jurisdiction. Amended section 1447(c) retains the bar against review of jurisdictionally based orders, but replaces improvident removal with procedurally defective removals complained of within thirty days of removal. Thus, if the ground for remand is an alleged procedural defect brought to the court's attention within thirty days of removal, an order of remand based thereon is not reviewable. The only alternative is to allow appellate review of any non-jurisdictionally based remand order. It is inconceivable, however, that such a radical alteration of the availability of appellate review was intended by the amendment. This is especially true given the Shell Oil cases, wherein the Fifth Circuit created a world in which only two types of removal defects exist. There are defects in subject matter jurisdictions and all other defects. The court deemed “all others” to be “procedural defects.”

Combining such an expansive definition of procedural defects with an interpretation of section 1447(c), allowing review of all remands based upon defects in removal procedure, would render section 1447(d)'s prohibitions on review almost meaningless. Such a view would allow appellate review of a remand order in every case where Article III jurisdiction existed. Thus, it is obvious that review of remand orders which are based upon defects in removal procedure, but which follow timely motions to remand, are not subject to appellate review. If, however, the court orders a remand based upon a procedural defect, but a remand motion was not filed within thirty days, appellate relief would be available.

[FNa] . J.D., magna cum laude, 1984, University of Illinois, College of Law, Order of the Coif. Junior Shareholder, Litigation Section, general trial and appellate law, Johnson & Gibbs, P.C., Dallas, Texas. Member: Dallas Bar Association, American Bar Association, Fifth Circuit Bar Association, Supreme Court Civil Rules Revision Task Force Consultive Group. Other publications include: The Judicial Improvements and Access to Justice Act: New Patent Venue, Mandatory Arbitration and More, 5 HIGH TECH.L.J. 311 (1990) . Sanctions Under Texas Rule 13, to be published in 11 REV.LITIG. (1991).

[FN43] . See Illinois Corporate Travel v. American Airlines, 889 F.2d 751 (7th Cir.1989) (Illinois Consumer Fraud act claim based upon alleged anti-competitive activity consisting of refusing to honor tickets issued by travel agents who engage in discount pricing or discount price advertising is preempted); New England Legal Foundation v. Massachusetts Port Auth., 833 F.2d 157 (1st Cir.1989) (state imposed landing-fee scheme preempted); O'Carroll v. American Airlines, 863 F.2d 11 (5th Cir.1989) (state law claims by passenger removed from aircraft after causing a disturbance preempted); Hingson v. Pacific Southwest Airlines, 743 F.2d 1408 (9th Cir.1984) (state laws concerning handicap discrimination preempted), Anderson v. US Air, 818 F.2d 49 (D.C.1987) (same). The Fifth Circuit in TWA also rejected several lower court holdings contrary to its decision. See New York v. Trans World Airlines, 728 F.Supp. 162 (S.D.N.Y.1989) ; California v. Trans World Airlines, 720 F.Supp. 826 (S.D.Cal.1989) ; Kansas v. Trans World Airlines, 730 F.Supp. 366 (D.Kan.1990) ; People v. Western Airlines, 155 Cal.App.3d 597, 202 Cal.Rptr. 237 (1984) , cert. denied, 469 U.S. 1132 (1985) . Also, the court distinguished Nader v. Allegheny Airlines, 426 U.S. 290 (1976) , which allowed fraud claims to proceed regardless of the CAB's authority over “deceptive practices” because Nader was decided prior to the addition of the express preemption provision.

[FN45] . Note that although state laws and causes of regulating airline rates, routes and services are completely preempted by the Deregulation Act, state courts still have jurisdiction over such claims. Federal Express. v. Pub. Serv. Comm., 925 F.2d 962, 967-68 (6th Cir.1991) . Therefore such claims may validly be litigated in state court if the parties do nothing to remove the case or remove it improperly.

[FN49] . Such removals have occasionally been successful. See Recognition Equip., Inc. v. Mohawk Data Sciences Corp., CA3-88-0761-T (N.D.Tex. Sept. 30, 1988). An in-depth discussion of what constitutes a “separate and independent” claim or cause of action is beyond the scope of this paper.

[FN53] . Finley v. United States, 490 U.S. 545 (1989) . In brief, Finley held that, assuming that pendent party jurisdiction is constitutional under Article 333, such jurisdiction is still unavailable to the federal courts unless Congress authorizes its exercise by statutory means. Id. at 549.

[FN67] . See id. There is tension between section 1441(d) added in 1976 by the Foreign Sovereign Immunities Act and section 1447(e) regarding pendent party jurisdiction. Compare 28 U.S.C. § 1441(d) (1988) with 28 U.S.C. 1447(e) , as amended by Pub.L. No. 100-702 , tit. X, § 1016(c), 102 Stat. 4642, 4670 (1988). Perhaps the ninth circuit decided Chuidian was incorrect and that Birkinshaw was the correct decision. The unique legislative history of section 1441(d) , whoever, supports the Chuidan decision. Chuidian, 912 F.2d at 1098-99.

[FN70] . See, e.g., Righetti v. Shell Oil Co., 711 F.Supp. 531 (N.D.Cal.1989) (after defendants removed case to federal court, plaintiffs sought leave to amend complaint by substituting one defendant for another, which destroyed diversity jurisdiction; thus, the case was remanded back to state court).

(a) Except as provided in subsections (b) and (c) or as expressly provided otherwise by Federal statute, in any civil action of which the district courts have original jurisdiction, the district courts shall have supplemental jurisdiction over all other claims that are so related to claims in the action within such original jurisdiction that they form part of the same case or controversy under Article III of the United States Constitution. Such supplemental jurisdiction shall include claims that involve the joinder or intervention of additional parties.

(b) In any civil action of which the district courts have original jurisdiction founded solely on section 1332 of this title, the district courts shall not have supplemental jurisdiction under subsection (a) over claims by plaintiffs against persons made parties under rule 15 , 19 , 20 , or 24 of the Federal Rules of Civil Procedure , or over claims by persons proposed to be joined as plaintiffs under rule 19 of such rules, or seeking to intervene as plaintiffs under rule 24 of such rules, when exercising supplemental jurisdiction over such claims would be inconsistent with the jurisdictional requirements of section 1332 .

(d) the period of limitations for any claim asserted under subsection (a), and for any other claim in the same action that is voluntarily dismissed at the same time as or after the dismissal of the claim under subsection (a), shall be tolled while the claim is pending and for a period of 30 days after it is dismissed unless State law provides for a longer tolling period.

[FN76] . See Soto v. Tonka Corp., 716 F.Supp. 977 (W.D.Tex.1989) (court held that § 1445(c) makes cases alleging discharge in retaliation for filing a worker's compensation claim non-removable); Chavez v. Farah Mfg. Co., 715 F.Supp. 177 (W.D.Tex.1989) (when plaintiff claims wrongful discharge for filing a worker's compensation claim defendant employer cannot remove to federal court); Watson v. Liberty Mut. Fire Ins. Co., 715 F.Supp. 797 (W.D.Tex.1989) (plaintiff's action to receive full amount of benefits awarded to her by the Texas Industrial Accident Board is non-removable to federal court); Wallace v. Ryan-Walsh Stevedoring Co., 708 F.Supp. 144 (E.D.Tex.1989) (wrongful discharge case for filing worker's compensation question not removable to federal court).

[FN78] . Nabors v. City of Arlington, 688 F.Supp. 1165 (E.D.Tex.1988) (statute prohibiting removal of action under worker's compensation laws did not preclude the district court from exercising pendent jurisdiction over employee's retaliatory discharge claim following removal of civil rights claim with which it was joined).

[FN82] . J. Baxter Brinkman Oil and Gas Corp. v. Thomas, 682 F.Supp. 898, 899-900 (N.D.Tex.1988) . In this case, the intervenor voluntarily injected itself into the suit and thus the situation was analogous to voluntary-involuntary rule situations. It is difficult to align this analogy with the Carl Heck case, which allows third-party defendant removals. Certainly, a plaintiff cannot be said to voluntarily make an action removable if it is the action of a defendant, an impleader, which makes the case removable. For the same reason, it is also hard to square Carl Heck itself with the voluntary-involuntary rule.

[FN83] . See In re Meyerland Co. (FDIC v. Meyerland Co.), 910 F.2d 1257, 1258-63 (5th Cir.) , reh'g granted, 921 F.2d 55 (5th Cir.1990) (Congress clearly intended federal court removal jurisdiction to reach cases not merely in the post-judgment phase, but also after appeal begins in the state court).

[FN95] . See, e.g., Keller v. Honeywell Protective Serv., 742 F.Supp. 425, 427 (N.D.Ohio 1990) (residency is not equivalent to citizenship); Stiegler v. McQuesten, 198 U.S. 141, 143 (1905) (residency and citizenship are wholly different concepts). Formal defects can sometimes be cured under 28 U.S.C. § 1653 . See 28 U.S.C. § 1653 (1988) . Section 1653 allows technical defects in the notice of removal, including defective allegations of citizenship, to be cured at any time, so long as the party seeking to cure the defective allegations has not engaged in bad faith or undue delay. Getty Oil, 841 F.2d at 1258 n. 5.

[FN96] . In multiple defendant removal situations, not all defendants must actually file a joint Notice of Removal. Instead, all that is required is for every served defendant to either file a Notice of Removal, or file some written instrument with the federal court which formally concurs with the existing Notice of Removal. Getty Oil, 841 F.2d at 1262 n. 11. However, it must be clear that every served defendant has actually filed a Notice or a joinder. Unsupported averments are insufficient to show consent to removal. Id.

[FN98] . Id. at 1263 n. 12 (citing Brown v. Demco, Inc., 792 F.2d 478, 481 (5th Cir.1981) ). Conversely, if a defendant is joined after removal has been accomplished, that defendant (who, for example may not be diverse) may move to remand the case back to state court. See Getty Oil, 841 F.2d at 1263; 28 U.S.C. § 1448 (1988) .

[FN103] . Id. Depending upon when the case first becomes removable, 30 days is not always allowed for removal. In some circumstances, waiver of the right to remove occurs when a defendant waits for a half-hour during a break in a state court trial to file a removal petition. See Walker v. A.T. & T. Co., 684 F.Supp. 475, 478 (S.D.Tex.1988) (diverse defendant's fifteen minute to one hour delay in notifying plaintiff it wanted to remove the case to federal court during a settlement conference held while the trial was ongoing and following voluntary dismissal of a non-diverse defendant estopped the diverse defendant from removing the case).

[FN104] . For instance, if the only non-diverse party is dropped by the plaintiff, thus creating diversity jurisdiction, the case could be removed at that time. 28 U.S.C. § 1446(b) (1988) . The elimination of the non-diverse party must be the voluntary act of the plaintiff or diversity jurisdiction still will not attach. See Weems v. Louis Dreyfus Corp., 380 F.2d 545, 549 (5th Cir.1967) ; Walker, 684 F.Supp. at 477; Higgens v. Pittsburgh-Des Moines Co., 635 F.Supp. 1182, 1184 (S.D.Tex.1986) .

[FN106] . Id. For the purposes of determining when the one-year removal limitation period begins, the fact that other defendants are added later in the case is irrelevant. The one-year period begins when the action was filed against the initial defendant. See Royer v. Harris Well Serv., Inc., 741 F.Supp. 1247, 1248-49 (M.D.La.1990) (applying state law to determine when the action was commenced, the court found the action commenced upon filing suit).