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

Nos. 07-3323 & 07-3338
PATRICK L. BAUDE, et al.,
                                             Plaintiffs-Appellees,
                               v.

DAVID L. HEATH, Chairman of the Indiana
Alcohol and Tobacco Commission,
                                           Defendant-Appellant,
                              and

WINE AND SPIRITS WHOLESALERS OF INDIANA,
                            Intervening Defendant-Appellant.
                        ____________
       Appeals from the United States District Court for the
        Southern District of Indiana, Indianapolis Division.
      No. 1:05-CV-0735-JDT-TAB—John Daniel Tinder, Judge.
                        ____________
    ARGUED FEBRUARY 22, 2008—DECIDED AUGUST 7, 2008
                        ____________


  Before EASTERBROOK, Chief Judge, and BAUER and POSNER,
Circuit Judges.
  EASTERBROOK, Chief Judge. After Granholm v. Heald,

(2005), held that states that allow wineries to
ship direct to consumers may not discriminate against out-
of-state vintners, Indiana revised its statutes. We had
2                                    Nos. 07-3323 & 07-3338

held in Bridenbaugh v. Freeman-Wilson, 

(7th
Cir. 2000), that the portions of Indiana’s laws there
under challenge were non-discriminatory but had flagged
other questionable provisions. Indiana eliminated them
and revamped the way in which it regulates direct ship-
ments.
  Today wineries inside and outside Indiana may ship
to customers, if (a) there is one face-to-face meeting
at which the buyer’s age and other particulars can be
verified; and (b) the vintner is not allowed to sell to
retailers in any state as its own wholesaler. Indiana
also requires wineries to obtain licenses and remit taxes,
and it limits each customer to 24 cases per winery per
year, but these elements of the state’s system have not
been challenged. The district court enjoined enforcement
of the two contested provisions because they have a
disparate impact on out-of-state sellers. 2007 U.S. Dist.
LEXIS 64444 (S.D. Ind. Aug. 29, 2007).
  A state law that discriminates explicitly (“on its face,”
lawyers are fond of saying) is almost always invalid
under the Supreme Court’s commerce jurisprudence,
which the Justices recapped this spring in Department of
Revenue of Kentucky v. Davis, 

, 1808–11 (2008).
(That recent decision makes it unnecessary for us to
rehearse the standards.) Plaintiffs, oenophiles who want
easier access to wine from small vineyards in other
states, do not contend that either of the two challenged
provisions discriminates in terms. Every rule applies to
every winery, no matter where it is located. The argu-
ment instead is that the rules impose higher costs on
interstate commerce as a practical matter.
  That brings into play the norm that, “[w]here the statute
regulates even-handedly to effectuate a legitimate local
Nos. 07-3323 & 07-3338                                     3

public interest, and its effects on interstate commerce are
only incidental, it will be upheld unless the burden im-
posed on such commerce is clearly excessive in relation to
the putative local benefits.” Pike v. Bruce Church, Inc., 

, 142 (1970). State laws regularly pass this test,
see 

–09, for the Justices are wary
of reviewing the wisdom of legislation (after the fashion
of Lochner) under the aegis of the commerce clause. For
recent cases in which this circuit has held that Pike
tolerates state laws of dubious benefit, see, e.g., Cavel
International, Inc. v. Madigan, 

(7th Cir. 2007);
National Paint & Coatings Ass’n v. Chicago, 

(7th
Cir. 1995).
   One of the two provisions challenged here is indeed a
needless and disproportionate burden on interstate com-
merce. The wholesale clause in Ind. Code §7.1-3-26-7(a)(6)
provides that a winery may sell direct to consumers only
if it “does not hold a permit or license to wholesale alco-
holic beverages issued by any authority” and is not
owned by an entity that holds such a permit. Indiana
says that this clause is designed to protect the state’s
“three-tier system” under which retailers may buy their
inventory only from wholesalers. If a wholesaler in another
state could sell wine direct to consumers, the state insists,
the winery-to-wholesaler-to-retailer-to-consumer model
would collapse.
  State laws that regulate the distribution chain, as this
one does, have been sustained against other challenges
under the commerce clause. See Exxon Corp. v. Governor of
Maryland, 

(1978). But the Court concluded in
Exxon that Maryland’s separation of the retail and whole-
sale functions did not affect interstate commerce in petro-
leum, all of which came from out of state no matter how
4                                    Nos. 07-3323 & 07-3338

the distribution system was organized. Indiana’s whole-
saler clause, by contrast, prevents direct shipment of
almost all out-of-state wine while allowing all wineries
in Indiana to sell direct. That happens because states
organize their distribution systems differently. Although
Indiana forbids any winery to sell to a retailer, many
other states either forbid wholesaling or are indifferent
to where retailers get their inventory. California, Oregon,
and Washington, which produce 93% of this nation’s
wine, have two-tier systems in which retailers buy from
producers without a middleman. All wineries in those
states lawfully may sell to retailers—which means that
Indiana classifies them as wholesalers and will not allow
them to ship wine to customers in Indiana. The statute
is neutral in terms, but in effect it forbids interstate ship-
ments direct to Indiana’s consumers, while allowing intra-
state shipments.
  Indiana does not defend the wholesale clause, though
a trade association, which intervened to protect its eco-
nomic interest, insists that the clause is valid. Pike asks
whether the putative local benefits could possibly
justify the burden on interstate commerce. All the whole-
salers can muster in support of the statute is that the three-
tier system may help a state collect taxes and monitor
the distribution of alcoholic beverages, because there are
fewer wholesalers than there are retailers, so state en-
forcement efforts can focus on the middle layer. That may
be so, see 

(stating in dictum that
the three-tier system is compatible with the dormant
commerce clause), but once a state allows any direct
shipment it has agreed that the wholesaler may be by-
passed. It is no harder to collect Indiana’s taxes from a
California winery that sells to California retailers than
Nos. 07-3323 & 07-3338                                      5

from one that does not. The wholesale clause protects
Indiana’s wholesalers at the expense of Indiana’s con-
sumers and out-of-state wineries.
   Analysis of the law’s other requirement is more com-
plex. Indiana requires any consumer who wants to
receive direct shipments of wine—from any winery, in
or out of Indiana—to visit the winery once and supply
proof of name, age, address, and phone number, plus a
verified statement that the wine is intended for personal
consumption. See Ind. Code §§ 7.1-3-26-6(4), 7.1-3-26-
9(1)(A). The parties call this the face-to-face clause. Plain-
tiffs say that a face-to-face meeting is more expensive,
the farther away is the winery (so the law has a disparate
impact on interstate commerce), and that local benefits
are negligible because people under 21 are bound to find
some way to get hold of wine no matter what the law
provides (they could, for example, present forged creden-
tials or bribe sellers to overlook their youth).
  Any balancing approach, of which Pike is an example,
requires evidence. See Minnesota v. Clover Leaf Creamery
Co., 

(1981). It is impossible to tell whether
a burden on interstate commerce is “clearly excessive in
relation to the putative local benefits” without under-
standing the magnitude of both burdens and benefits. See
Cherry Hill Vineyard, LLC v. Baldacci, 

(1st Cir.
2007). Exact figures are not essential (no more than esti-
mates may be possible) and the evidence need not be in
the record if it is subject to judicial notice, but it takes
more than lawyers’ talk to condemn a statute under Pike.
  The vital bit of information for the wholesale clause is
that 93% of all wine comes from states that have two-tier
systems. Indiana concedes as much and does not proffer
any local benefit to offset the exclusionary effect. But
6                                   Nos. 07-3323 & 07-3338

Indiana has not conceded that it is particularly costly
for consumers to visit wineries on the west coast, or that
an effort to verify buyers’ ages is worthless. Plaintiffs
have waged the suit as a “facial” challenge to the stat-
ute—which means that Indiana receives the benefit of
any plausible factual suppositions, for a statute is not
unconstitutional “on its face” if there is any substantial
possibility that it will be valid in operation. See, e.g.,
Washington State Grange v. Washington State Republican
Party, 

(2008). When some form of height-
ened scrutiny applies—as it does if a law’s own terms treat
in-state and out-of-state producers differently—then the
burdens of production and persuasion rest on the state.
But when challenging a law that treats in-state and out-of-
state entities identically, whoever wants to upset the law
bears these burdens.
  The costs of a face-to-face meeting depend on distance,
not on borders, and many consumers in Indiana are
closer to some wineries in Michigan or Illinois than to most
wineries in Indiana. But then plaintiffs aren’t interested
in wine from Illinois, Michigan, Kentucky, or Ohio. They
have their hearts set on the boutique wineries of Cali-
fornia, Oregon, and Washington, which are materially
farther away.
  Plaintiffs invite us to think of a trip to California for
the sole purpose of signing up at a single vintner. Yet one
winery per trip is not the only, or apt to be the usual,
way to satisfy the face-to-face requirement. Many oeno-
philes vacation in wine country, and on a tour through
Napa Valley to sample the vintners’ wares a person
could sign up for direct shipments from dozens of winer-
ies. Wine tourism in Indiana is less common, and the
state’s vineyards—which altogether have fewer than 350
Nos. 07-3323 & 07-3338                                    7

acres under cultivation—are scattered around the state,
making it hard for anyone to sign up at more than a few
of Indiana’s wineries. Wineries of Indiana, a trade associa-
tion, has a map showing its 40 members’ locations. See
http://www.indianawines.org/wineries/?loc=map. These
wineries are all over the map. A connoisseur might
well find it easier to visit and sign up at 30 California
wineries than at 30 Indiana wineries. So although it may
be more costly for a person living in Indianapolis to
satisfy the face-to-face requirement at five Oregon win-
eries than at five Indiana wineries, it is not necessarily
substantially more expensive (per winery) to sign up at
a larger number of west-coast wineries than at an equiva-
lent number of Indiana wine producers.
  If it turns out to be more expensive (per winery) to
sign up in California than in Indiana, is the extra cost
justified by the wineries’ ability to check the credentials
of potential buyers? Plaintiffs and several amici curiae
supporting them maintain that age verification when the
wine is delivered is enough. But we know from Rowe v.
New Hampshire Motor Transport Ass’n, 

(2008),
that states cannot require interstate carriers to verify the
recipients’ age. Even if that case had come out the other
way—or if some carriers offer an age-checking service
without the need for legal compulsion—a rushed driver
is unlikely to take as much care in checking credentials,
and testing for forgery with ultraviolet light and other
methods, as a winery’s desk clerk. Some drivers treat
anyone 18 and over as an “adult”, see Staff of the Federal
Trade Commission, Possible Anticompetitive Barriers to
E-Commerce: Wine 36 (2003); no winery would do so. The
FTC’s staff concluded that data do not reveal “how
often couriers obtain a valid adult signature.” 

& 07-3338

  Plaintiffs concede that keeping alcohol out of minors’
hands is a legitimate, indeed a powerful, interest. Still, they
want us to take judicial notice that minors who are deter-
mined to drink will find a way to beat any system, so that
there is no point in having a “system” in the first place.
That’s not at all clear. How well any given system of
screening works is an empirical subject on which we
lack reliable information. As we observed in National
Paint, a legal system need not be foolproof in order to
have benefits. The face-to-face requirement makes it
harder for minors to get wine. Anything that raises the
cost of an activity will diminish the quantity—not to zero,
but no law is or need be fully effective.
  According to plaintiffs, Internet-based age-verification
services are as effective as verification in person. The main
support offered for this proposition is an assertion on one
provider’s web site that it achieves 94% accura-
cy in matching data to people of known ages. See http://
www.choicepoint.com/products/age_verification.html?
l2=verification_authentication&bc=bva&sb=b. Yet neither
the record in this case nor any third-party testing of the
web site’s accuracy shows whether its assertion is correct
or how easy it is for teenagers to supply data that pro-
duces a spurious match to an adult.
   Plaintiffs also point to two reports that, they say, estab-
lish the ineffectiveness of in-person age verification. See
the FTC’s Staff Report (above) and National Research
Council, Institute of Medicine, Reducing Underage Drinking:
A Collective Responsibility (2004). These reports do not
support plaintiffs’ contention. What they show instead
is that state officials “report few problems” and the like.
That subjective, unquantified reaction (perhaps it shows
that the officials haven’t searched for problems, or that no
Nos. 07-3323 & 07-3338                                      9

adverse stories have appeared in local newspapers) is not
enough to override a state legislature’s assessment. The
FTC’s staff also reported that, in tests of the verification
system in liquor stores, minors were able to buy alcoholic
beverages between 15% to 30% of the time. Possible
Anticompetitive Barriers 35. That’s a far cry from proof
that face-to-face verification at a winery would be inef-
fective or unimportant. Even though it does imply that
minors who visit enough stores (or enough wineries) are
likely to be accepted eventually at one or more of them,
the need to visit multiple outlets raises the cost and so
reduces sales to minors. Remove the verification require-
ment from direct shipments, and more minors would turn
to that source. It is important to remember that we are
dealing with effects on the margin; make it easier for
minors to get wine by phone or Internet, and sales to
minors will increase.
   Indiana thinks that in-person verification with photo ID
helps to reduce cheating on legal rules, for both buying
wine and voting (and perhaps other subjects). After
the Supreme Court held in Crawford v. Marion County
Election Board, 

(2008), that a belief that in-
person verification with photo ID reduces vote fraud
has enough support to withstand a challenge under the
first amendment, it would be awfully hard to take judicial
notice that in-person verification with photo ID has no
effect on wine fraud and therefore flunks the interstate
commerce clause.
  Given the state of this record, and the state of the empiri-
cal literature, we know very little. What we can guess
at implies that face-to-face verification will reduce the
fraction of all wine shipments that go to minors, though
the size of this effect is hard to estimate. Minors who
10                                   Nos. 07-3323 & 07-3338

can get beer locally may not want to pay for costly, up-
market wine plus shipping charges; if so (and we don’t
know whether it is so), then Indiana may come to con-
clude that age verification for direct shipments is not
vital. The cost of verification per winery rises with dis-
tance, if consumers sign up at only one winery per trip;
but when traveling through wine country consumers
may be able to sign up at many wineries at small incre-
mental cost. So both the marginal cost and the marginal
benefit of Indiana’s face-to-face system may be modest.
That is not enough to declare a law unconstitutional—not
when the effect on interstate commerce is negligible.
   Indiana has not tried to keep wine from crossing its
border. Go to a liquor outlet in Indiana, and you will find
wines from California, Oregon, Washington, France,
Germany, Italy, Australia, South Africa, and Chile—but
little if any wine from Indiana. It is possible that the face-
to-face clause benefits small Indiana wineries near the
state’s population centers but lacking wholesale distribu-
tors, vis-à-vis small California wineries that lack whole-
sale distributors in Indiana, but Indiana’s system does
not disadvantage California (or other) wineries in gen-
eral. The law’s principal effect may be to boost larger Cali-
fornia (Oregon, etc.) wineries, which have established
distribution systems, over smaller wineries from any
state, including Indiana, that do not have wholesale
distributors.
  None of the plaintiffs contends that Indiana’s law has
led him to buy more wine from Indiana and less from other
states. The law simply shifts sales from smaller wineries
(in all states, including Indiana) to larger wineries (all of
which are located outside Indiana). The Indiana Wine-
growers Guild has filed a brief as amicus curiae opposing
Nos. 07-3323 & 07-3338                                  11

the face-to-face clause, which the Guild maintains has
made it unduly difficult for its members to ship their
wine direct to consumers. But if what the Guild says is
true, then the statute—although bad economically for
Indiana’s wineries—must be sustained against a chal-
lenge under the commerce clause. Favoritism for large
wineries over small wineries does not pose a constitutional
problem, and the fact that all Indiana wineries are small
does more to show that this law’s disparate impact cuts
against in-state product than to show that Indiana has
fenced out wine from other jurisdictions.
  The judgment of the district court with respect to the
wholesale clause is affirmed, and with respect to the face-
to-face clause is reversed. The case is remanded for the
entry of a judgment consistent with this opinion.




                           8-7-08