In the
 United States Court of Appeals
              For the Seventh Circuit
                         ____________

No. 07-2128
BRIAN ASHER, et al.,
                                                         Plaintiffs,
                                v.

BAXTER INTERNATIONAL INCORPORATED, et al.,
                                          Defendants-Appellees.

APPEAL OF:
   CITY OF FAYETTEVILLE        FIREMEN’S PENSION
   AND RELIEF FUND
                         ____________
           Appeal from the United States District Court
      for the Northern District of Illinois, Eastern Division.
          No. 02 C 5608—Blanche M. Manning, Judge.
                         ____________
 SUBMITTED SEPTEMBER 26, 2007—DECIDED OCTOBER 17, 2007
                         ____________


 Before EASTERBROOK, Chief Judge, and MANION and
ROVNER, Circuit Judges.
  EASTERBROOK, Chief Judge. More than three years ago
we held that the complaint in these six consolidated
securities actions could not be summarily dismissed under
the safe harbor, 15 U.S.C. §78u–5(c), created by the
Private Securities Litigation Reform Act of 1995 for
forecasts and other forward-looking statements. Asher v.
2                                                No. 07-2128

Baxter International Inc., 

(7th Cir. 2004). We
expected that discovery sufficient to make a prompt
decision about the safe harbor would follow our opinion,
for the safe harbor is supposed to be applied at an
early stage. What happened instead was extended wran-
gling about who should be the “lead plaintiff ” under the
1995 Act, and thus which law firm would control the
plaintiffs’ side of the litigation. See 15 U.S.C. §78u–4(a)(3).
   “Lead plaintiffs” are supposed to counteract the domi-
nance of lawyers over class-action suits; the district judge
should select a representative with a financial stake
large enough to make monitoring of counsel worthwhile,
and with the time and skills needed to make monitoring
productive. The idea is that securities suits then will
proceed in the interest of investors rather than the
lawyers who appoint themselves to prosecute these
actions. In this case, however, the district court eventu-
ally held that none of the persons proposed as lead plain-
tiffs is satisfactory and that the suit therefore cannot
proceed as a class action. A motions panel authorized
an interlocutory appeal under Fed. R. Civ. P. 23(f ).
   The principal substantive questions on appeal are (a)
whether the City of Fayetteville Firemen’s Pension and
Relief Fund (“the Fund”) is unsuitable as a lead plaintiff
because it learned about Baxter International’s sup-
posed wrongs from a securities lawyer rather than from
a business executive, and (b) whether “no one” can be
the answer to the question “who is the best representa-
tive of investors”? Perhaps, when all potential lead plain-
tiffs have shortcomings, the district judge must choose
the least bad of a mediocre lot; after all, the 1995 statute
refers to “the most adequate plaintiff ” among many,
without setting a floor. Minimum standards of adequacy
are the domain of Fed. R. Civ. P. 23(a)(4), which permits
a class action only if “the representative parties will
fairly and adequately protect the interests of the class.” In
No. 07-2128                                             3

this case, though, the district court never asked whether
the Fayetteville Pension Fund would be adequate under
Rule 23(a)(4)—or for that matter whether its financial
interest is great enough to make it an appropriate cham-
pion under §78u–4(a)(3)(B)(iii).
  We may consider these issues, however, only if the
appeal is timely, and Baxter insists that it is more than
a year late. The district court denied a motion for class
certification in November 2005 after concluding that
James and Heidi Hill, who had been selected as lead
plaintiffs in 2002, had misrepresented their ownership of
Baxter’s stock. Their motion for selection said that they
owned 2,663 shares; discovery revealed that they owned
only 2.663 shares. (Their lawyer, who has since been
indicted for fraud in conducting other class-action suits,
called the misrepresentation an “administrative error.”)
The Hills also disclosed that they had not sought lead-
plaintiffs status (they thought the form they returned
was an application to share in any recovery) and had
never attempted to monitor counsel’s work on behalf of
the class.
   Instead of seeking to pursue an interlocutory appeal
under Rule 23(f ), counsel proposed two other lead plain-
tiffs. Cauley Bowman Carney & Williams PLLC proposed
Tommy Newman, who had purchased 900 shares, and
Milberg Weiss LLP (as it is now known) proposed Eliza-
beth G. Sherry, who owned 300 shares of Baxter’s stock.
Discovery revealed that neither Newman nor Sherry
wants to supervise the work of counsel in a complex
securities case, or would be good at that task. Newman
had been recruited by the law firm and, the district
judge determined, is its tool; Sherry, like the Hills, had
returned a form thinking it essential to receive any
recovery and had no desire to play an active role. The
district court concluded that both Newman and Sherry
would be “totally inadequate” as class representative
4                                              No. 07-2128

and in September 2006 denied a renewed motion to
certify a class.
  Once again no appeal was taken; once again counsel
proposed new candidates for lead plaintiff. Cauley Bow-
man proposed the City of Fayetteville Firemen’s Pension
and Relief Fund, which purchased 600 shares during the
time when Baxter’s price was said to have been artificially
high, and the law firm now known as Coughlin Stoia
Geller Rudman & Robbins LLP proposed the Alaska
Laborers Employers Retirement Fund, which bought
10,500 shares during that window. The district court
deemed both the Alaska and the Fayetteville funds
inadequate because their investments are much smaller
than those of other mutual or pension funds. One can’t
help thinking that the unwillingness of any substan-
tial shareholder to step forward as a representative
suggests that the suit may not be in investors’ interest. To
the district judge, the fact that two modestly sized pools
with modest stakes in Baxter had been recruited by the
lawyers already trying to represent a plaintiff class
implied that they would be subservient to counsel. This
ruling was made in January 2007.
  At this point the Fund asked the judge to deny its own
motion for class certification—not because the Fund had
decided to support Baxter, but to set up the possibility of
interlocutory appeal. Evidently the judge, having al-
ready twice denied a motion to certify a class of investors,
had not seen a reason to enter a third such order; the
judge had thought it sufficient to rebuff another attempt
to designate lead plaintiffs. But Rule 23(f ) does not allow
interlocutory appeals from orders designating (or not
designating) lead plaintiffs—and although 28 U.S.C.
§1292(b) could in principle allow an interlocutory
appeal from such an order with the approval of both the
district court and this court, the Fund did not ask for
permission to appeal under §1292(b). On March 29, 2007,
No. 07-2128                                                 5

the district judge granted the Fund’s motion to deny its
own motion for class certification, and an application for
leave to appeal under Rule 23(f ) was made and granted.
   Rule 23(f ) says: “A court of appeals may in its discretion
permit an appeal from an order of a district court grant-
ing or denying class action certification under this rule
if application is made to it within ten days after entry of
the order.” Baxter maintains that the 10 days expired
in November 2005, when the district court first denied a
motion for class certification. The Fund replies that
each new “order of a district court granting or denying
class action certification” starts a new 10-day period. (The
Fund also argues that the decision of this court’s mo-
tions panel not to dismiss the application as untimely
is conclusive, but that’s a makeweight. The motions
panel did not hold that the appeal is timely; it simply
denied the motion to dismiss. This allowed briefing to
proceed, so that the question could be resolved after a
full adversarial presentation. A one-line order denying a
motion to dismiss does not block full consideration by
the merits panel.)
  This circuit has held that the 10 days under Rule 23(f )
cannot be extended by making another motion for class
certification. See Gary v. Sheahan, 

, 893 (7th
Cir. 1999). Cf. Blair v. Equifax Check Services, Inc., 

, 837 (7th Cir. 1999) (a timely motion for recon-
sideration of the original order granting or denying
class certification extends the period for appeal, but an
untimely or successive motion does not). Other circuits
have come to the same conclusion. E.g., Carpenter v.
Boeing Co., 

(10th Cir. 2006); Jenkins v.
BellSouth Corp., 

(11th Cir. 2007). The
time limit would not be worth anything if it restarted
with each new motion. Then the rule might as well say “at
any time” instead of “within ten days”. A short limit
would be turned into an indefinite one. Rule 23(f ) sets a
6                                              No. 07-2128

brief limit because the appeal is interlocutory; if the
disposition is not reversed swiftly, the case should proceed
in the district court. Arguments pro and con about class
certification then can be made on appeal from the final
decision. There’s no reason to drag out the prospect of
interlocutory review when a new window of appellate
review will open once a final decision has been reached.
  According to the Fund, however, this presumes that
there was only one decision in the district court. The Fund
allows that successive motions for the same relief do not
reopen the time for appeal. But here, as the Fund sees
things, there have been three distinct decisions: first
whether a class could be certified with the Hills as lead
plaintiffs, second whether a class could be certified with
Newman and Sherry as lead plaintiffs, and third whether
a class could be certified with itself as lead plaintiff.
Doubtless the motions were different enough that the
district court could not have invoked the law of the case to
reject any of them. But the ability to extend the debate
about certification in the district court does not mean
that the window of opportunity for appellate review must
be open indefinitely. The longer this process takes in
the district court, the less appropriate is interlocutory
review that will prolong the litigation even further.
  Behrens v. Pelletier, 

(1996), holds that
defendants may have more than one chance to take an
interlocutory appeal to argue for official immunity. The
reason for that decision—that each order rejecting an
immunity defense is independently “final” under 28
U.S.C. §1291 as that term was understood in Mitchell v.
Forsyth, 

(1985)—does not carry over to
appeals under Rule 23(f ). As we stressed in Fairley v.
Fermaint, 

(7th Cir. 2007), Behrens and
Mitchell allow appeals without regard to the district
court’s reasons; it is enough that a given order prolongs
the litigation and thus (further) impinges on a defendant’s
No. 07-2128                                                7

potential right not to be sued. Rule 23(f ) does not create
a right not to be sued, or for that matter a right to inter-
locutory appeal. It creates a (brief ) opportunity for expe-
dited review, and if that opportunity passes the entitle-
ment to review at the end of the case remains.
  Still, the Fund insists, the fact that the district judge
made three decisions denying class certification shows
that the first and second were not “definitive,” and that
the time to appeal did not begin until the third decision.
Yet no interlocutory decision is “definitive”; classes may be
certified, modified, or decertified as the case progresses.
The Fund’s approach would allow lengthy deferral of
appeal, and we would become embroiled in questions
such as whether the district judge’s ruling was tentative,
definitive, or something in between; that would be a
formula for paralysis. The Fund cannot get traction from
its further argument that the district judge failed to
supply “adequate legal analysis” for the first two deci-
sions; that amounts to saying that the time for appeal
does not begin until the judge has cured all errors! The
thing being appealed is the order; a paucity of legal
analysis may be a reason to reverse an order but is not a
reason to pretend that the judge never entered an order.
  Having said all of this, we recognize that a district court
could keep open the prospect of interlocutory review under
Rule 23(f ) by denying one class member’s motion for
appointment as lead counsel and inviting a new motion
from some other member, all the while leaving the
motion for class certification undecided. Only after run-
ning through all potential lead plaintiffs would the dis-
trict judge act on the motion to certify a class—and an
appeal within 10 days then could bring up the decisions
about lead-plaintiff status. (Baxter is wrong to think
that an appeal under Rule 23(f ) does not present anteced-
ent questions such as lead-plaintiff dispositions. Any
appeal presents the order for appellate decision, and a
8                                              No. 07-2128

court of appeals is free to address all considerations that
make the order sound or erroneous. See Yamaha Motor
Corp. v. Calhoun, 

, 204–05 (1996).)
  As it happens, however, the district judge did not
proceed that way—at least not the first two times. The
court denied lead-plaintiff status to the Hills and entered
a contemporaneous order denying a motion for class
certification, repeating that pattern with Newman and
Sherry. Only when it denied the third set of applications
for lead-plaintiff status did the district court leave the
class question open. Perhaps class counsel missed the
boat by not asking our leave to appeal from the first
decision, not on the ground that the Hills should have
been selected as lead plaintiffs, but on the ground that
rejecting one potential lead plaintiff does not warrant an
immediate denial of class certification. The opportunity to
make such an argument passed by in November 2005,
however, and cannot be reopened now. Prudently or not,
the district court declined to certify this suit as a class
action in November 2005, and prudently or not no one
invoked Rule 23(f ) at the time. That ended the opportunity
for interlocutory review under Rule 23(f ), leaving §1292(b)
and an appeal from a final decision. The Fund did not
request certification of the January 2007 lead-plaintiff
decision under §1292(b), and it is too late to do this now.
But appeal from a final decision remains possible.
  Waiting for a final decision will expose class counsel to
additional financial risks. Without a certified class, the
case has lower settlement value, and if counsel litigates
to final decision on the merits Baxter may well prevail,
and then there would be no point in arguing for class
certification. But the difficulties in pursuing a suit to a
final decision do not justify taking liberties with Rule
23(f ). The final-decision rule of §1291 is the norm, and
Rule 23(f ) is an exception that, like §1292(b), must be
used sparingly lest interlocutory review increase the
No. 07-2128                                               9

time and expense required for litigation. Counsel repre-
senting the plaintiffs are in the business of litigating
securities actions; they spread risk across a portfolio of
suits and cannot demand that this court bend the rules
to relieve them of risk voluntarily assumed in a particular
case.
  Both the judicial system and the investors gain from
dispatch. Class actions are unwieldy and often dominated
by lawyers. (That’s a major reason why the lead-plaintiff
statute was enacted.) Investors are poorly situated to
protect their own interests, while lawyers are tempted to
drag out the case to increase their fees (or, by fighting
to have their clients named as lead plaintiffs, to engross
larger portions of available fees). Judges should insist
that these cases proceed to decision rather than linger on
the docket. Preventing the window for Rule 23(f ) review
from remaining open for years (as the Fund proposes)
promotes the public interest.
  In concluding that the appeal comes too late, we have
not needed to answer the question whether the 10-day
limit in Rule 23(f ) is “jurisdictional.” Bowles v. Russell,

(2007), holds that statutory deadlines
for appeal are jurisdictional, but read in conjunction with
decisions such as Eberhardt v. United States, 

(2005), holds out the possibility that deadlines in the
federal rules are just claim-processing norms. See also
Scarborough v. Ryan, 

(2004); Kontrick v.
Ryan, 

(2004). The 10-day limit in Rule 23(f )
lacks a statutory counterpart. Rule 23(f ) was adopted
in 1998 as an exercise of the Supreme Court’s power under
28 U.S.C. §1292(e) to authorize interlocutory appeals
by promulgating rules under the Rules Enabling Act, 28
U.S.C. §2072. How much time litigants have to take
interlocutory appeals is a question for the rulemaking
process, which implies that the deadline is not juris-
10                                            No. 07-2128

dictional. See Coco v. Belle Terre, 

, 491 (2d
Cir. 2006).
  But jurisdictional or not, the time limit is manda-
tory—which means that it must be enforced if the litigant
that receives its benefit so insists. See El Paso Natural
Gas Co. v. Neztsosie, 

(1999). There will be
time enough to choose between “jurisdictional” and “claim-
processing norm” characterizations if, in some future case,
the appellee either consents to a belated appeal or fails
to object. Then the “jurisdictional” characterization
would matter; today it does not, because Baxter has asked
us to enforce the 10-day limit.
 The appeal is dismissed as untimely.

A true Copy:
      Teste:

                       ________________________________
                       Clerk of the United States Court of
                         Appeals for the Seventh Circuit




                  USCA-02-C-0072—10-17-07