While driving a Greyhound bus from St. Louis to Columbus, Ohio on February 25, 1993, Fred Hurley was involved in a serious accident with a jack-knifed double trailer. Hurley himself suffered severe lower body injuries in the collision, which collapsed the front wall of the bus. The bus that Hurley was driving was equipped only with a standard two point seat belt. It had neither an airbag nor any structural enhancements that would provide additional protection to the driver in the event of a high speed crash like the one in which he was involved. Hurley believes that these flaws make the bus he was driving unreasonably dangerous and, consequently, render the bus’s manufacturer, Motor Coach Industries, liable for his injuries.
He and his wife (whose loss of consortium claims are entirely dependent on his success in this litigation) thus decided to sue Motor Coach, as well as Consolidated Freightways (the operators of the trailers that his bus hit), Road Systems, Inc. (the makers of the trailers that Hurley struck), and Pines Trailer Corporation (another trailer manufacturer). Hurley began in Illinois state court, but Consolidated Freightways removed the action to the District Court for the Southern District of Illinois on diversity grounds. After the removal, Hurley settled his claims against Consolidated Freightways and Road Systems and dismissed Pines under Fed. R.Civ.P. 41(a). Except as they relate to a jurisdictional problem that we address below, Hurley’s claims against these parties are not at issue in this appeal. The parties agreed to submit their case to a magistrate judge, as 28 U.S.C. § 636(c) permits. Finding that Hurley’s case was preempted by the federal crashworthiness standard, Magistrate Judge Proud dismissed the action; this appeal followed.
I
Before we discuss the merits of Hurley’s appeal, we must address a potential jurisdictional problem. Hurley’s original suit involved four defendants — Motor Coach, Consolidated Freightways, Road Systems, and Pines Trailer. For jurisdictional purposes, the Hurleys are citizens of Arkansas. Motor Coach is a Delaware corporation with its principal place of business in Arizona. Consolidated Freight-ways is also a Delaware corporation, but its principal place of business is Oregon. Road Systems is a California corporation and its principal place of business is there, too. Finally, Pines Trailer is an Illinois corporation with its principal place of business in Illinois. Complete diversity, which is required for jurisdiction under 28 U.S.C. § 1332, therefore exists between the Hurleys and all defendants. (Given the severity of Hurley’s injuries, nobody is contesting the adequacy of the amount in controversy.)
Not every diversity case qualifies for removal, however. For cases that start out in state court, where it is the defendant who wants the federal forum, there is an additional hurdle to clear before successfully reaching federal court. Under 28 U.S.C. § 1441(b), a non-federal question case “shall be removable only if none of the parties in interest properly joined and served as defendants is a citizen of the State in which such action is brought.” This rule, sometimes called the “forum defendant” rule, means that Pines’s presence in the case, if proper, would normally keep the case in state court. We say “normally” because in the typical case, a plaintiff who is about to see her chosen court slip away will object to removal if there is an in-state defendant, and thereby secure a speedy return to state court.
That did not happen in this case. After the defendants filed their removal
*379
petition, Hurley did nothing. Since 28 U.S.C. § 1447(c) gives plaintiffs only 30 days to object to removal, Motor Coach argues that Hurley has waived any objection to removal. Of course, waiver is possible only if the forum defendant rule is nonjurisdictional; true jurisdictional flaws are nonwaivable and can be raised at any time. See,
e.g., Ruhrgas AG v. Marathon Oil Co.,
The distinction between subject-matter jurisdiction and waivable defenses is not a mere nicety of legal metaphysics. It rests instead on the central principle of a free society that courts have finite bounds of authority, some of constitutional origin, which exist to protect citizens from the very wrong asserted here, the excessive use of judicial power. The courts, no less than the political branches of the government, must respect the limits of their authority.
Id.
at 77,
We must decide, therefore, whether the forum defendant rule is jurisdictional, in the sense we have been using the term, or if it is of a lesser status. That question has been bouncing around the federal courts of appeals for more than 75 years, yet oddly enough it remains unresolved in this circuit.
LaMotte v. Roundy’s, Inc.,
A number of reasons persuade us that the majority rule is the correct one. First, the theme of several recent Supreme Court decisions that have considered curable defects in diversity jurisdiction has been to find that as long as the court’s jurisdiction is proper at the time of trial and judgment, the case need not be dismissed because of an earlier jurisdictional problem. This was the precise holding of
Caterpillar, Inc. v. Lewis,
Newman-Green, Inc. v. Alfonzo-Larrain,
The situation we face is different, and in many ways less troublesome, than the problem the Court encountered in Caterpillar and Nevmanr-Green. If Hurley’s case had been filed in the first instance in federal court, jurisdiction under § 1332 would have been clear. Or, if Hurley had originally filed a state court action against Motor Coach, Consolidated, and Road Systems, those defendants could have removed the case to federal court since none was an in-state defendant. Once in federal court, Hurley could have filed an amended complaint adding Pines, if it turned out that Pines might have been responsible for his injuries. In short, the case as it arrived in federal court met every requirement for federal jurisdiction: it simply took the wrong path, in a sense, because there was an in-state defendant. This, we think, is more a matter of removal procedure, and hence waivable, than a matter of jurisdiction.
The policy behind the forum defendant rule also supports the nonjurisdictional characterization. The rule is designed to preserve the plaintiffs choice of a (state) forum, under circumstances where it is arguably less urgent to provide a federal forum to prevent prejudice against an out-of-state party. After removal, if the plaintiff wants to remain in state court, she can file a timely motion for remand. Under the circumstances, it seems that the only purpose that would be served by declaring the forum defendant rule jurisdictional would be to preserve for plaintiffs rights that the plaintiffs themselves failed to assert. For these reasons, we join the longstanding line of authority that holds that the forum defendant rule is nonjuris-dictional. Hurley thus had only 30 days to object to removal based on the presence of Pines in the case (which did not last long anyway, as it turned out that Pines had not built the trailers in question). Since there were no other potential flaws with the district court’s diversity jurisdiction, we now turn to the merits of the appeal.
II
Hurley’s claim against Motor Coach is an Illinois products liability action. In order to show that Motor Coach’s product was unreasonably dangerous, he had to present an alternative design that is “economical, practical, and effective.”
Baltus v. Weaver Division of Kidde & Co., Inc.,
Motor Coach argued to the court below that there are three reasons that Hurley’s design is insufficient to get his claims to a jury. First, it maintained that his design is not economical, practical, or effective
*381
because it will be too expensive and, in any event, the steel cage will subject the driver to crushing deceleration in a front-end collision. Next, it contested the qualifications of Hurley’s experts under the standard articulated in
Daubert v. Merrell Dow Pharmaceuticals,
509
U.S.
579,
Specifically, the district court found that Hurley’s claim is preempted by FMVSS 208, which is the federal crash-worthiness standard. The Illinois products liability cases dealing with alternative designs suggest that there is more than one way of showing that a product as designed by the defendant was unreasonably dangerous, although one possibility is to prove that an “economical, practical, and effective” alternative design would be safer.
Baltus,
Hurley’s theory is remarkably close to the one the Supreme Court rejected in
Geier v. American Honda Motor Co.,
— U.S. -,
Aside from the fact that Hurley’s claim concerns buses and Geier’s claim concerned cars, it is hard to see any significant difference between the two situations. The applicable standard for buses like the one that Hurley was driving is found at 49 C.F.R. § 571.208 S4.4.3.1; see also National Transportation Safety Board Factual Report of Highway Accident Investigation, CRH-93-F-H018 at 4 (Hurley’s bus was manufactured in 1992; its gross vehicle weight was approximately 28,200 pounds.). The regulatory section provides that “[e]ach bus with a gross vehicle weight rating of more than 10,000 pounds shall comply with the requirements of S4.4.2.1 or S4.4.2.2.” That “or” is critical, since S4.4.2.1 calls for a “complete passenger protection system” for the driver; it must comply with crash protection requirements found in S5 of FMVSS 208. On the other hand, S4.4.2.2 requires only a seat belt assembly; it, too, is subject to guidelines, but what is important for our case is that it does not require an airbag or a three point seat belt, as Hurley’s design does. Nobody doubts, and indeed Hurley argued in his brief, that under his theory the manufacturer loses the choice between two point seat belts and the comprehensive system that his engineers have designed. Motor Coach’s argument is thus a fairly straightforward one — because federal law gives bus manufacturers a choice as to the driver protection systems installed in a particular bus, an Illinois tort suit that rests on a theory that forecloses that choice is preempted.
*382
Geier
confirms Motor Coach’s theory that a state lawsuit that forecloses an option left open by FMVSS 208 is in fact preempted. In
Geier,
the Court traced the history of FMVSS 208 and concluded that Congress and the Department of Transportation expressed a preference for leaving to manufacturers the choice between airbags and other types of safety devices (at least for the period in time during which the plaintiffs car was manufactured).
Geier
noted the controversy over the efficacy and utilization of airbags and seat belts and concluded that FMVSS 208 “deliberately sought variety” by leaving the choice of passenger protection system up to manufacturers.
Geier,
One particular aspect of Geier’s discussion of FMVSS 208 is especially relevant here. One of the major problems with which the Department of Transportation needed to deal when formulating standards for passenger cars is a strong aversion among the driving public to seat belts. Id. at 1923-24 (recalling the “fiasco” that resulted when auto-makers installed systems that prevented drivers from starting their cars unless seat belts were buckled). NHTSA found that this problem was even more acute among heavy vehicle operators. See 53 Fed. Reg. 25336, 25338 (July 6, 1988) (noting that seat belt use among heavy truck drivers is far below the national average for all vehicles). The commentary accompanying the final rules indicates that even the details of seat belt design — such as the retractor mechanism, mounting position, and buckle releases— were hotly debated during the rulemaking process. See id. at 25339-41. Consequently, NHTSA’s decision to leave the manufacturers of heavy buses with the option of using a two point seat belt that met regulatory standards is quite understandable as a way of promoting safety by encouraging drivers to use the safety equipment that manufacturers install. But Hurley’s proposal unequivocally requires a three point seat belt, a requirement that could reduce utilization and thereby undermine FMVSS 208’s safety objectives. Perhaps the design would not in fact reduce seat belt utilization, but that is not the point. The point is that, as in Geier, the decision to leave options open to bus manufacturers was made with specific policy objectives in mind. Hurley’s suit, if successful, would undermine that policy objective and is therefore preempted.
Even before
Geier,
our own case law would have led us to the same result. See
Gracia v. Volvo Europa Truck, N.V.,
Ill
We conclude that the Supreme Court’s opinion in Geier governs Hurley’s case and compels the conclusion that when a Federal Motor Vehicle Safety Standard leaves a manufacturer with a choice of safety device options, a state suit that depends on foreclosing one or more of those options is preempted. Hurley’s proposed design, with three point seat belts and airbags, would preclude a bus manufacturer from exercising the “[sjecond option” contained in S4.4.2.2. The history of the safety requirements for large buses like the one Hurley was driving indicates that these options were important in the overall safety regime created by FMVSS 208. His suit interferes with these federal policy choices. Under Geier, it is preempted, and we thus AffiRm the judgment of the district court.
