ArticlePDF Available

What Structural Presumption? Reuniting Evidence and Economics on the Role of Market Concentration in Horizontal Merger Analysis

Authors:

Abstract

The “structural presumption” is a proposition in antitrust law standing for the typical illegality of mergers that would combine rival firms with large shares of the same market. Courts and commentators are rarely precise in their use of the word “presumption,” and there is foundational confusion about what kind of presumption this proposition actually entails. It could either be a substantive factual inference based on economic theory, or a procedural device for artificially shifting the burden of production at trial. This paper argues that the substantive inference interpretation is the better reading of caselaw and the sounder application of the laws of antitrust and evidence. By instead interpreting the structural presumption as a formal rebuttable presumption, modern merger analysis needlessly complicates the use of market concentration evidence, and may be systematically undervaluing the probative weight of this evidence. At least in this context, a formal presumption likely confers less evidentiary weight than a simple substantive inference.
What
Structural
Presumption?:
Reuniting
Evidence
and
Economics
on
the
Role
of
Market
Concentration
in
Horizontal Merger
Analysis
Sean
P.
Sullivan*
The
"structural
presumption"
is
a
proposition
in
antitrust
law
standing
for
the
typical
illegality
of
mergers
that
would combine
rival
firms
with
large
shares
of
the
same
market.
Courts
and
commentators
are
rarely
precise
in
their
use
of
the
word
"presumption
"
and
there
is
foundational
confusion
about
what kind
of
presumption
this
proposition
actually
entails.
It
could
either
be a
substantive
factual
inference
based
on
economic
theory
or
a
procedural
device
for
artificially
shifting
the
burden
of
production
at
trial.
This
Article
argues that
the
substantive
inference
interpretation
is
the
better
reading
of
case
law
and
the
sounder
application
of
the
laws
of
antitrust
and
evidence.
By
instead interpreting
the
structural
presumption
as
a
formal
rebuttable
presumption,
modern
merger analysis
needlessly
complicates
the
use
of
market
concentration
evidence
and
may be
systematically
undervaluing
the
probative
weight
of
this
evidence. At
least
in
this
context,
a
formal
presumption
likely
confers
less
evidentiary
weight
than
a
simple
substantive
inference.
1.
INTRODUCTION.............................................................405
II.
THE
QUESTION
POSED.......................
...............................
408
A.
Market
Concentration
Evidence.................................408
B.
The
Structural
Presumption...................................410
C.
Evidentiary
Presumptions....................................411
1.
Substantive
Factual
Inferences..............................412
2.
Rebuttable
(Burden-Shfiting)
Presumptions
.....................
413
D.
Competing
Interpretations...................
................
415
III.
THE
CASE
LAW
HISTORY
...................................................
416
A.
The
1960s
(Philadelphia National
Bank)..............
............
416
B.
The
1970s
(General
Dynamics)................................420
C.
The
1980s
and
1990s
(Baker
Hughes)............................422
IV.
POSITIVE
ANALYSIS.......................................................424
A.
The
Probative
Value
ofMarket
Concentration
.........
....................
425
Bureau
of
Competition,
Federal
Trade
Commission.
This Article
reflects
the
personal
views
of
the
author.
It
does
not
necessarily
represent
the
views
of
the
United
States
Govemnment
or
Federal
Trade
Commission,
which
have
neither approved
nor
disapproved
its
content.
Contact
the
author
at
ssully@gmail.com.
This
Article
has benefitted
from the
thoughtful comments
of
Jon
Baker,
Malcolm
Coate,
Josh
Fischman,
Larry
Fullerton,
Mike Gilbert, Josh Goodman,
Rich Hynes,
Paul
Rothstein,
George
Ruthergien,
Steve
Salop,
and
workshop
participants
at Duke,
Georgetown,
and
Northwestern.
404
The
Journal
of
Corporation
Law
[Vol.
42:2
B.
The
Form
and
Procedure
ofRebuttal.
.......................
.....
429
C.
The
Order
ofHorizontal
Merger
Litigation
..................
......
431
V.
NORMATIVE ANALYSIS
..........................................
.....
434
A.
Confusion
of
Horizontal
Merger
Analysis
...................
......
434
B.
Undervaluation
ofMarket
Concentration
Evidence
.............
.....
438
VI.
CONCLUSION
......................................................
442
What
Structural
Presumption?
1.
INTRODUCTION
The
structural
presumption
is
an
important but
contentious proposition
in
the
antitrust
law
of
horizontal
mergers.
It
stands
for
the
typical
illegality
of
mergers
that
would
combine
rival
firms with
large
shares
of
the
same
relevant
market.
The
structural
presumption
is
so
named
because
the
likely
anticompetitive
effects
of
such
mergers
are
in
some
sense
presumed
to
follow
from
the
change
in
market
structure
involved
in
such
consolidations.
In
one
form
or
another,
the
structural
presumption
has
undergirded
all
antitrust
analysis
of
horizontal
mergers
since at
least
the
early
1960s.
Today,
the
structural
presumption
is a
topic
of
significant
debate.
Though
well
entrenched
in
legal
precedent,
the
validity and normative
desirability
of
this
proposition
are
increasingly
questioned. Proponents
argue
that mergers leading
to
strongly
concentrated
markets
tend
to
lessen
competition
and
harm
consumers.
On
this
basis,
they
support
a
presumption
that
mergers
leading
to
highly
concentrated
markets should
generally
be
prohibited,
even
if
specific
theories
of
competitive
harm
are
not
brought
forward
by
the
party
seeking
relief.
Opponents
of
the
proposition
argue
that market
structure
should not
be
treated
as
determinative
of
the
competitive
consequences
of
a
merger.
They propose
to
eliminate any
presumption
of
harm
based
on
market
concentration
evidence
and
would
generally require
specific
theories
of
harm
to be
raised
as
the
basis for
relief.2
As
thin
s
stand,
consensus
is
not
forthcoming. Across
numerous
academic
papers,3
conferences,
formal
speeches,5
informal commentaries,6
administrative
statements,7
and
1.
See,
e.g.,
Steven
C.
Salop, The
Evolution
and
Vitality
of
Merger
Presumptions:
A
Decision-Theoretic
Approach,
80
ANTITRUST.
L.J.
269,
271
(2015)
(providing
a
decision-theoretic
justification
for
reliance
on
a
structural
presumption
of
competitive
harm
in
certain
cases); Jonathan
B.
Baker
&
Carl
Shapiro,
Reinvigorating
Horizontal
Merger
Enforcement,
in
HOW
THE CHICAGO
SCHOOL
OVERSHOT
THE
MARK:
THE
EFFECT
OF
CONSERVATIVE
ECONOMIC
ANALYSIS
ON
U.S.
ANTITRUST
233
(Robert
Pitofsky
ed.,
2008)
(arguing for
a
strengthening
of
merger
enforcement through
partial
restoration
of
the
structural presumption and
the
need
for
strong evidence
to overcome
a
prima
facie case
based
on
market
concentration
evidence).
Note that these
analyses take
much more
nuanced positions
than
this
simple
summary
suggests.
2. See,
e.g.,
Barry
C.
Harris
&
David
D.
Smith,
The
Merger
Guidelines
v.
Economics:
A
Survey
of
Economic
Studies,
1999
ANTITRUST
REP.
23,
27
(1999)
(arguing
against
the
policy
of
presuming
harm
from
market concentration
evidence);
Ky
P.
Ewing,
Jr.,
The
Soft
Underbelly
ofAntitrust:
Some
Challenging
Thoughts
for
the
New
Millennium,
1999
ANTITRUST
REP.
2,
2-4
(1999)
(criticizing
reliance
on
market concentration
in
merger analysis);
see also
Michael
G.
Cowie
&
Paul
T.
Denis, The
Fall
of
Structural
Evidence
in
FTC
and
DOJ
Merger
Review,
ANTITRUST SOURCE 3
(Feb.
2013)
(interpreting antitrust agencies
to
be
paying
less
attention
to
market concentration
in
their enforcement decisions);
cf
John Harkrider,
Proving
Anticompetitive Impact:
Moving
Past
Merger
Guidelines
Presumptions,
2005
COLUM.
BUS.
L.
REV.
317,
319 (2005)
(arguing
against
the
presumption
of
harm, and
in
favor
of
specific
theories
of
anticompetitive
effects).
3.
E.g.,
Jonathan
B.
Baker,
Market
Concentration
in the
Antitrust
Analysis
of
Horizontal
Mergers,
in
ANTITRUST
LAW
AND
ECONOMICS
234
(Keith
N.
Hylton,
ed.
2010);
Jonathan
B.
Baker
&
Steven
C.
Salop,
Should
Concentration
be
Dropped
from
the
Merger
Guidelines?,
33
U.
WEST.
L.A.
L. REV.
3
(2001);
see
also
supra
notes
1-2
(referencing arguments
for
and
against this structural
presumption).
4.
E.g.,
Conference
on
the
Fiftieth Anniversary
of
United States
v.
Philadelphia
National
Bank:
The
Past,
Present,
and
Future
of
Merger
Law,
N.Y.U.
(Nov.
15,
2013),
http://www.law.nyu.edulconferences/us-vs-
pnb.
5.
E.g.,
Joshua
D.
Wright,
Comm'r,
FED.
TRADE COMM'N,
The
FTC's
Role
in
Shaping
Antitrust
Doctrine: Recent Successes
and
Future
Targets,
Remarks
at
the
2013
Georgetown Global
Antitrust
Symposium
405
2016]
The
Journal
of
Corporation
Law
judicial
opinions,
arguments
have
been
forcefully
articulated
for
and
against
the
structural
presumption
with
little
to
show
for
the
exercise. The
opposing
sides
are
at
an
impasse, leaving
unresolved
the current
and
future
significance
of
the
structural
presumption.
This
is
an
undesirable
state
of
affairs
for
a
proposition
fundamental
to
the
basic legality
of
business
transactions
that
are
often
valued
in
millions
to
billions
of
dollars.
Clarity
on
the
role
and
relevance
of
market concentration evidence
is
attainable,
but
not
by
continuing
to
debate
the
legal
providence
of
using
this
evidence
as
the
basis
for
a
presumption
of
anticompetitive
harm. Instead, this Article
asks
a
more
basic
question:
what
kind
of
presumption
is
it
that market concentration
evidence
supports
in
the
first
place? Courts
and
commentators
are
often
imprecise
in
their
language and
procedure
when
it
comes
to
presumptions,9
and
treatment
of
the
structural
presumption
is
no
exception.
As
this Article
shows,
debate over
the
use
of
market concentration evidence
in
the
antitrust
analysis
of
mergers
can
be
focused-and
largely
resolved-by
simply
addressing
a
latent
ambiguity
in
the
conversation
to
date:
namely,
what
do we
mean
when
we
say
that evidence
of
undue
concentration
supports
a
presumption
of
competitive
harm?
There
are
two
basic
possibilities.
First,
the
structural
presumption
could
be
a
substantive
factual
inference
based
on
evolving economic theory
and
independently
probative
of
the
competitive
consequences
of
a
merger.
In
this
sense,
competitive
harm
is
presumed
to arise
as
a
logical
implication
of
the
increase
in
market
concentration
caused
by
a
merger
of
large rival firms.
Second,
the
structural
presumption
could
be
a
formal
rebuttable
presumption
that
artificially
shifts
the
burden
of
production
to the
defendant
when
undue
concentration
is
shown.
10
In
this
sense,
competitive
harm
is
presumed
by
the
mandate
of
a
procedural
device
of
administrative
convenience, without
direct
reference
to
the
probative
value
of
the
underlying
evidence.
Dinner
(Sept. 24,
2013),
https://www.ftc.gov/sites/default/files/documents/public-statements/ftc%E2%80%99s-
role-shaping-antitrust-doctrine-recent-successes-and-future-targets/130924globalantitmstsymposium.pdf.
6.
E.g.,
Josh
Wright,
The
Guidelines Should
be
Revised
to
Reject
the
PNB
Structural
Presumption,
TRUTH
ON
THE
MKT.
(Oct.
26,
2009),
https://truthonthemarket.com/2009/10/26/the-guidelines-should-be-
revised-to-reject-the-pnb-structural-presumption/;
D.
Daniel
Sokol,
Should
the Philadelphia
National
Bank
Presumption
Be
Abandoned
or
Allowed
to
Evolve?,
ANTITRUST
& COMPETITION
POL'Y
BLOG
(Sept.
30,
2013),
http://lawprofessors.typepad.com/antitrustprof blog/2013/09/should-the-philadelphia-national-bank-
presumption-be-abandoned-or-allowed-to-evolve.html.
7.
FED.
TRADE
COMM'N,
Dissenting
Statement
of
Commissioner
Joshua
D.
Wright
in
the Matter
of
Fidelity
National
Financial,
Inc.
and Lender Processing
Services,
Inc.
1,
3-4
n.8
(Dec.
23,
2013),
https://www.ftc.gov/system/files/documents/cases/140305fidelitywrightstatement.pdf.
8.
Compare
United
States
v.
Baker
Hughes
Inc.,
908
F.2d
981,
984
(D.C.
Cir.
1990)
(giving
little
apparent
weight
to
market concentration evidence, described
as
merely
"a
convenient starting
point"
for
analysis) with
FTC
v.
H.J.
Heinz
Co.,
246
F.3d 708,
717
(D.C.
Cir.
2001)
(giving ostensibly
strong
weight
to
market
concentration
evidence,
with additional commentary
that
"[a]s
far
as
we
can
determine,
no
court
has
ever
approved
a
merger
to
duopoly
under similar
circumstances").
9.
21B
CHARLES
ALAN WRIGHT
ET
AL.,
FEDERAL
PRACTICE
AND
PROCEDURE:
EVIDENCE
§
5122.1
(2d
ed.
2012) (commenting
that
inconsistent
use
of
the
term
presumption
has
by
some
accounts rendered
the word
"all
but
meaningless").
10.
Cf
FED.
R.
EVID.
301
(describing
the
treatment
of
burden-shifting
presumptions
under the
Federal
Rules
of
Evidence).
406
[Vol.
42:2
What
Structural
Presumption?
Intermediate
combinations
of
these
two
extremes
are
also
possible, but
this
Article
supports
no
middle ground.
While
it
is
assumed
today
that
the
structural
presumption
is a
formal
rebuttable presumption,
this Article
argues that
it
is
actually
a
simple
substantive
inference.
The
distinction
is
important, and
more than
academic.
First,
this Article shows
that
the
substantive
inference
interpretation
is
compelled
by
an
honest reading
of
decades
of
controlling
case
law
in
antitrust. That
is,
the
thesis
of
this
Article
is
not
merely
an
economic
or
theoretical
argument.
The need
to
treat
the
structural
presumption
as
a
substantive
factual
inference
is a
practical statement
of
merger law and
should
be
followed
by
courts
and
litigants
when
considering
the
antitrust
legality
of
horizontal mergers.
Second,
even
beyond
the
specific
antitrust
case law on
point,
interpreting
the
structural
presumption
as
a
simple
factual
inference
is
the
better
practice
as
a
matter
of
general legal
principles.
The
substantive inference
interpretation
reflects
economic
theory,
the
substantive
law
of
antitrust, and the
procedural
law
of
evidence
for
a
fact
like
market concentration.
By
contrast,
the
structural
presumption
interpretation distorts
and
confuses
the
relevance
of
market
concentration evidence under both
substantive and
procedural
law.
Third,
the
distinction
is
of
practical
importance.
Interpreting
the
structural
presumption
as
a
rebuttable
legal
presumption complicates
horizontal merger
analysis
without
providing
any
benefit
to
justify
the cost.
This
interpretation
also
perniciously
obscures
the
probative
value
of
market
concentration
evidence,
likely
leading
to the
systematic undervaluation
of
this
evidence
by
courts and
analysts. Paradoxically,
creating
a
formal
rebuttable
presumption
does
not
have
the
expected
effect
of
imbuing
weak
evidence
with
artificial weight
in
this
setting.
Instead, it
actually
weakens
the
natural
evidentiary
weight
of
intrinsically probative
evidence.
Interpreting
the
structural
presumption
as
a
substantive
inference corrects
these
problems.
It
also
narrows and significantly
resolves
the
extant
debate
on the
validity
and
future
viability
of
the
structural
presumption
in
merger
law.
Resolution
of
uncertainty
over
the
role
of
market
concentration evidence
in
merger
analysis
is
immediate:
this
evidence supports
a
substantive
economic inference about
the
competitive consequences
of
a
merger.
Further
economic
inquiry
into the
value
and
limits
of
market
concentration
evidence
as
a
predictor
of
anticompetitive harm
is
needed,
but
all
remaining questions
are
economic
in
nature.
This Article
obviates
future
legal
debates
about
the
calibration
or
utility
of
formal
burden-shifting
devices
on
this
topic.
Market concentration
evidence
deserves
no
more
and
no
less
than
its
own
intrinsic
probative
value
as
evidence.
The
remainder
of
this Article
presents
the
argument
outlined
above.
The
question-
what
structural
presumption?-is
posed
in
Part
II.
Context
on
the
concept
of
market
concentration
is
related
to
the different
presumptions
that
this
evidence
might support.
The
following
sections
then
compare the
substantive
inference
and rebuttable
presumption
interpretations
of
the
structural
presumption, arguing
in
favor
of
the
former.
Part
III
traces
the
history
of
the
structural
presumption
in
case
law,
showing
that
the
presumption
is
more
naturally interpreted
as
a
substantive
inference
than
a
formal
burden-shifting
presumption.
Part
IV
combines
current
economic
thinking
with close
attention
to the
rules
of
evidence
on
presumptions
to
show that interpreting
the
structural
presumption
as
a
substantive
inference
better
fits
the
relevant
procedure
and substantive
goals
of
antitrust
law.
Part
V
compares the
normative
desirability
of
each
approach
to the
2016] 407
The
Journal
of
Corporation
Law
structural
presumption.
The
rebuttable
presumption
interpretation
is
shown
to have
no
advantages
over
the
substantive
inference
interpretation,
but
many
disadvantages.
A
brief
conclusion responds
to
potential criticisms
and
proposes
to
restore
appropriate
weight
to
market concentration evidence
by
reviving
the
historic
substantive
inference
approach
to
evidence
of
undue
concentration.
II.
THE
QUESTION POSED
Under
Section
7
of
the
Clayton
Act-the
principal
statutory basis
for
antitrust
merger
analysis
in the
United
States-a
merger
is
illegal
if
its
effect
"may
be
substantially
to
lessen competition, or
to
tend
to
create
a
mono
oly."II
Since
at
least
the
1963
decision
of
United
States
v.
Philadelphia
National
Bank,
courts and analysts
have
looked
to
market
concentration
evidence
as
a
primary
source
of
information
on
the
likely
competitive consequences
of
a
merger.
This
is
particularly
true
in the
case
of
horizontal
mergers: mergers
of
competitors
in
the
same
relevant
market.
In
contemporary
language,
evidence that
a
horizontal
merger
will
result
in
undue
concentration
is
said
to
support
a
structural
presumption
that
the
merger
is
likely
to
have
anticompetitive
effects.
13
Relief
from
such
a
finding may
include injunction
of
the
merger, divestitures, or
other remedies.
But
the
importance
of
the structural
presumption
in
both
historic
and
contemporary
merger
analysis
belies
lurking
uncertainty
about
what
exactly
is
being presumed
when
undue
concentration
is
proved. There
are
multiple
senses
in
which
market concentration
evidence
could
be
said
to support
a
presumption
of
anticompetitive
effects
and
only
limited
appreciation
of
this
complexity
in
merger
case
law
and
the
academic literature.
Clarity
on
the
role
of
market
concentration
evidence
in
horizontal
merger
analysis
requires
first-order agreement
on
what
kind
of
presumption
this
evidence
is
meant
to
support.
As
this Article
demonstrates,
a
wrong
answer
poses
serious
problems
for
the
antitrust
analysis
of
horizontal mergers.
A.
Market Concentration
Evidence
In
the
abstract,
market
concentration
contemplates
a
classification
of
market
structures
along
a
one-dimensional spectrum
of
competitive
dispersion. At
one
extreme
of
the spectrum
is
a
perfectly competitive
market: for example,
an
infinite number
of
price-taking
wheat
farmers. At
the
other
extreme
lies
a
single
price-setting
monopolist:
the
clearest
example
is a
firm
with
government-granted exclusivity
in
its
market.
Oligopolistic
market
structures
fall
somewhere
between
these
two
extremes,
with
markets
11.
Clayton
Act,
ch.
323,
§
7,
38
Stat. 730,
731-32
(1914)
(current version
at
15
U.S.C.
§
18
(2012)).
This
standard
is
generally
thought
to
differ
from
the
prohibitions
of
Sections
1
and
2
of
the
Sherman
Act,
ch.
647,
§§
1-2,
26 Stat.
209,
209
(1890)
(current
version
at
15
U.S.C.
§§
1-2
(2004)),
in
that
Section
7's
incipiency standard
("may
be
substantially
to
lessen") (emphasis
added) allows
mergers
to
be
prohibited
on
proof
of
probable
anticompetitive effects
as
opposed
to
actual
anticompetitive
effects. Brown Shoe Co.
v.
United
States, 370
U.S.
294,
323 (1962);
see
also
United
States
v.
Koppers
Co.,
202
F.
Supp.
437,
439-40
(W.D.
Pa. 1962)
(discussing
the
legislative
history
of
the
Clayton
Act).
12.
United
States
v.
Phila.
Nat'1
Bank,
374
U.S.
321
(1963).
13.
See Baker,
supra
note
3,
at
2
(discussing market concentration
evidence
and
the
structural
presumption
of
competitive
harm
in
merger analysis).
408
[Vol.
42:2
What
Structural
Presumption?
consisting
of
many
small
firms
falling
closer
to
the
perfectly
competitive
side
and
markets
consisting
of
large
duopolies
falling closer
to
the
monopoly
side.
This
spectral
concept
of
market concentration
is
theoretically
attractive
but
admits
no
perfect
empirical
analog.
In
practice,
courts
and
scholars
rely
on a
wide
array
of
quantifiable measurements to
approximate
the
abstract concept
of
market concentration.
Consider
a
merger
of
the
largest
two
firms
in
a
market
consisting
of
seven
firms with
the
following
shares:
35%,
20%,
15%,
10%,
10%,
5%,
and
5%.
Common
descriptions
of
the
merger's
effect on
market concentration
include
market
shares
in
output
or
sales
revenue
(i.e.,
the
merger
leads
to
a
combined
firm with
a 55%
share
of
the
market),
the
number
of
remaining
firms
(i.e.,
this
is a
seven-to-six
merger),
the
four-firm concentration
ratio
(i.e.,
the
merger
increases
the
C4
index
from
80%
to
90%),
and
the
Herfindahl-Hirschman
index
(HHI)
(i.e.,
the
merger
raises
the
HHI
from 2100
to
3500).14
Each
of
these
empirical
measurements
focuses
on
a
slightly
different aspect
of
the
change
in
market
structure
caused
by
the
merger,
but
all
capture
the basic idea
that
the
post-merger
market
is
more
concentrated than
before.
Economic research
has long
drawn
a
causal
relationship between
market
concentration
and
firm
behavior.
In
the 1950s
and
1960s,
the
predominant
view was
that
even
moderate increases
in
market
concentration
would
tend
to
cause
substantial
reductions
in
competition.
15
Stigler,
for
example,
suggested
in 1955
that sound
economics supported
the
prohibition
of
most
mergers
involving
a
firm
with
more
than
a
20%
share
of
the
relevant
market.16
The
rise
of
Chicago-school
economics and empirical
critiques
of
concentration-profit studies
during
the 1970s
and
1980s
precipitated
an
economic withdrawal
from
confidence
in
the
structure-conduct relationship,
7
but
the
change
was more
in
degree than
conclusion.
The
rapid
introduction
of
game
theory
into
industrial
organization
economics
in
the
1980s
and
1990s
reestablished
the
causal
link
between increased concentration
and
decreased competition
in
many
models.
It
is
safe
to
say
that
mainstream
economic
thinking
currently
holds that substantial
changes
in
concentration
have
at
least
a
modest
causal
relationship
to the
likelihood
of
14.
See
JEAN
TIROLE, THE
THEORY
OF
INDUSTRIAL
ORGANIZATION
§
5.5
(1988)
(discussing
the
abstract
theory
and
measurement
of
market
concentration
and
summarizing important
works
in
this
literature). For
a
historic
perspective,
see
also
Gideon
Rosenbluth,
Measures
of
Concentration,
in
BUSINESS
CONCENTRATION
AND
PRICE
POLICY
57
(1955)
(describing concentration
indexes);
Tibor
Scitovsky,
Economic
Theory
and
the
Measurement
of
Concentration,
in
BUSINESS
CONCENTRATION
AND
PRICE
POLICY
101
(1955)
(discussing
the
effects
of
market
concentration
on
competition).
For
a
gentler
introduction
to
the
idea
of
market concentration,
see
Oz
SHY,
INDUSTRIAL ORGANIZATION:
THEORY
AND
APPLICATIONS
171-73
(1995).
15.
See,
e.g.,
George
J.
Stigler,
Mergers
and
Preventive
Antitrust Policy,
104
U.
PA.
L.
REV.
176,
181-82
(1955)
[hereinafter
Stigler,
Mergers
and
Preventive
Antitrust
Policy]
(providing
simple
concentration-based
rules
for
identifying
potentially
anticompetitive
mergers);
see
generally
George
J.
Stigler,
A
Theory
of
Oligopoly,
72
J.
POL.
ECON.
44
(1964);
George
J.
Stigler,
Introduction
to
BUSINESS CONCENTRATION
AND
PRICE POLICY
(1955);
CARL
KAYSEN
&
DONALD
F.
TURNER,
ANTITRUST
POLICY:
AN ECONOMIC
AND
LEGAL
ANALYSIS
(1959);
Jesse
W.
Markham,
Merger
Policy under
the New
Section
7:
A
Six-Year
Appraisal,
43
VA.
L.
REV.
489
(1957).
16.
Stigler,
Mergers
and
Preventive
Antitrust
Policy,
supra
note
15,
at
182.
17.
See,
e.g.,
Harold
Demsetz,
Two
Systems
of
BeliefAbout
Monopoly, in
INDUSTRIAL
CONCENTRATION:
THE
NEW
LEARNING
164
(Harvey
J.
Goldschmid
et
al.
eds.,
1974)
(explaining
this
critique).
2016]
409
The
Journal
of
Corporation
Law
anticompetitive
effects
in
many
circumstances.
18
The strength
of
this
relationship
is
subject
to
various
influences
such
as
the ease
and
credibility
of
entry and
repositioning,
the
existence
of
factors
conducive
of
coordinated
behavior,
and the
basic nature
of
competition
in the
market.19
But
as
a
general
proposition,
few
would
deny
that
a
sufficiently
large
increase
in
market concentration
is
probative
of
likely
anticompetitive
harm.20
This
inference may
be drawn
circumstantially,
without
reference
to
specific
theories
of
harm,
or
it
may
be
drawn
directly,
by
treating
market concentration
as
a
factor
that strengthens
a
specific
theory
of
competitive
harm.
2 1
B.
The
Structural
Presumption
Economic
thinking
on
the
relevance
of
market
concentration
as
a
predictor
of
the
likely
anticompetitive
effects
of
a
merger
has
long
been
a
part
of
horizontal merger
review
in the
courts.
In
Brown
Shoe Company
v.
United
States,
the
Supreme
Court's
first
merger
case
following
major
amendment
of
the
Clayton
Act
in
1950,22
the Court twice
cited
Stigler's
1955
article
for
the
importance
of
market concentration
as
a
factor
in
merger
review.23
A
year
later
in
United
States
v.
Philadelphia
National
Bank,
the
Court
again
cited
Stigler and
several
other
economists
in
support
of
the
following
proposition:
[A]
merger
which
produces
a
firm
controlling
an
undue
percentage
share
of
the
relevant
market,
and
results
in
a
significant
increase
in
the
concentration
of
18.
See,
e.g.,
TIROLE,
supra
note
14
and
accompanying
text,
§§
5-7;
Louis Kaplow
&
Carl Shapiro,
Antitrust
35-36, 59-74
(Harvard
John
M.
Olin
Discussion
Paper
Series,
Discussion Paper
No.
575,
2007),
http://www.1aw.harvard.edu/programs/olin
center/papers/pdf/KaplowShapiro_575.pdf;
Carl
Shapiro,
Theories
of
Oligopoly
Behavior,
in
HANDBOOK
OF
INDUSTRIAL
ORGANIZATION
329 (Richard Schmalensee
&
Robert
Willig
eds.,
1989);
Baker
&
Shapiro,
supra
note
1;
Baker
&
Salop,
supra
note
3;
SHY,
supra
note
14,
at
171-
206;
see
generally
Raymond
Deneckere
&
Carl
Davidson,
Incentives
to
Form
Coalitions
with
Bertrand
Competition,
16
RAND J.
EcoN.
473
(1985).
19.
See
generally supra
note
18
(concerning
the
connection
between
market
concentration and potential
competitive
harm from
mergers).
20.
See
Salop,
supra
note
1,
at
277
n.45-46
(noting
a
number
of
economic studies which
support
some
form
of
anticompetitive-effect
inference
from
sufficient
market
concentration
evidence).
21.
See,
e.g., Baker
&
Salop,
supra
note
3,
at
8
(noting that market
concentration
is
relevant to
both
coordinate
and unilateral
theories
of
harm under
a
variety
of
models
and
assumptions);
see
generally
Martin
Dufwenberg
&
Uri
Gneezy,
Price
Competition
and
Market Concentration:
An
Experimental
Study,
18
INT'L
J.
INDUS.
ORG. 7
(2000) (relating concentration
to
the
likelihood
of
coordinated
conduct);
cf
John Kwoka,
Professor
Econ.
Northwestern
U.
Some
Thoughts
on
Concentration,
Market Shares,
and
Merger
Enforcement
Policy
at the
FTC/DOJ
Workshop
on
Merger
Enforcement
1
(Feb.
17,
2004),
https://www.justice.gov/sites/default/files/atr/legacy/2007/08/30/202602.pdf
(noting that different measures
of
market
concentration
relate
differently
to
different
specific theories
of
harm).
But
see
generally
Joseph
Farrell
&
Carl
Shapiro,
Antitrust
Evaluation
of
Horizontal
Mergers:
An
Economic
Alternative
to
Market Definition,
10
B.E.
J.
THEORETICAL
ECON.
1
(2010)
(suggesting
that
upward pricing
pressure
is
a
better
indicator
of
anticompetitive
effects than
market concentration
in
most
unilateral effects
models).
22.
Celler-Kefauver
Act,
ch.
1184,
64 Stat.
1125
(1950)
(current version
at
15
U.S.C.
§§
18,
21
(2000));
see generally
Note,
Section
7
of
the
Clayton
Act:
A
Legislative History,
52
COLUM.
L.
REV.
766
(1952)
(examining
and
interpreting
the
legislative history
of
the Clayton
Act).
23.
Brown
Shoe
Co.
v.
United
States, 370
U.S.
294, 332
n.56,
334
n.61
(1962)
(citing
Stigler,
Mergers
and
Preventive Antitrust
Policy,
supra
note
15);
see
also
Brown
Shoe,
370
U.S.
at
322
n.38
(commenting that
market
concentration
is
the
"primary
index
of
market
power"
but
must
be
assessed relative
to
the
market
in
question).
410
[Vol.
42:2
What
Structural
Presumption?
firms
in
that
market
is
so
inherently
likely
to
lessen
competition substantially
that
it
must
be
enjoined
in
the
absence
of
evidence clearly
showing that
the
merger
is
not likely
to
have such
anticompetitive
effects.
24
This
language
in
Philadelphia
National
Bank
is
now
the
well-settled
citation
for
the
structural
presumption
in
merger
analysis.
Parsing
the
exact
meaning
of Philadelphia
National
Bank
is
the
subject
of
much
of
Part
III
of
this Article. For
now,
however,
it
is
important
to
note
that
the
structural
presumption
includes
at
least
one
other
important
source
of
authority.
In
its
1990
disposition
of
United
States
v.
Baker
Hughes,
the
D.C.
Circuit
provided
what
is
now
black-letter
law
on
the
structural
presumption
in
horizontal
merger
analysis:
By
showing that
a
transaction
will
lead
to
undue
concentration
.
..
the
government
establishes
a
presumption
that
the
transaction
will
substantially
lessen
competition.
The
burden
of
producing
evidence
to
rebut this
presumption
then
shifts to the
defendant.
If
the
defendant successfully rebuts
the
presumption,
the
burden
of
producing
-
additional
evidence
of
anticompetitive
effect
shifts
to
the
government,
and
merges
with
the
ultimate
burden
of
persuasion, which
remains
with
the
government
at
all
times.
25
Philadelphia
National
Bank
and
Baker
Hughes
represent fundamentally different
approaches
to
the
use
of
market
concentration
evidence
in
Section
7
litigation. The
difference
is
the
type
of
presumption that
each
entails.
Philadelphia National
Bank
describes
a
substantive
inference
based
on
economic
theory,
while
Baker
Hughes
articulates
a
rebuttable
presumption divorced
from the
specific
probative
value
of
market
concentration
evidence.
C.
Evidentiary
Presumptions
In
the
half-century
since
1963,
courts and
commentators
have
done little
to
clarify
the
type
of
presumption
that
proof
of
undue
concentration
is
meant
to
support.26
Among
the
possible
interpretations
of
presumption
in
legal contexts,27
only
two
are
relevant
to
the
remainder
of
this Article. These
are
(1)
a
substantive
factual
inference, and
(2)
a
burden-shifting
rebuttable
presumption.
The
differences between
these
interpretations
are
subtle
but
important.
24.
United
States
v.
Phila.
Nat'l
Bank,
374
U.S.
321, 363
(1963).
25.
United States
v.
Baker
Hughes
Inc.,
908
F.2d
981,
982-83
(D.C.
Cir.
1990).
26.
Cf
Charles
V.
Laughlin,
In
Support
ofthe
Thayer
Theory
of
Presumptions,
52
MICH.
L.
REV.
195,
195
(1953)
(commenting
that
"the word
[presumption]
has been
so
promiscuously
used
as to
be
devoid
of
much
of
its
utility");
Ronald
J.
Allen,
Presumptions
in
Civil Actions
Reconsidered,
66
IOWA
L. REV.
843, 844
(1981)
(quoting
Learned Hand
on
the
legal
meaning
of
"presumption"
as
saying
"Judges
have
mixed
it
up
until
nobody
can tell
what
on
earth
it
means").
27.
See
Laughlin,
supra
note
26,
at
196-205
(noting eight
senses
in
which
the
word
"presumption"
is
used
by
courts:
(1)
a
general
disposition
of
courts,
(2)
an
authoritative
reasoning principle,
(3)
a
rule
of
substantive
law, (4)
a
rule
fixing
the
burden
of
persuasion,
(5)
a
permissible
inference,
(6)
a
statutory
prima
facie case,
(7)
a
proposition
ofjudicial
notice,
and
(8)
a
rule
shifting
the
burden
of
producing
evidence).
2016]
411
The
Journal
of
Corporation
Law
1.
Substantive
Factual
Inferences
The
plaintiff
in
a
Section
7
case
always
bears
the
initial
burden
of
producing
evidence that
a
merger
may
be
"substantially
to
lessen
competition,
or
to
tend
to
create
a
monopoly."28
In
modem
terminology,
the
plaintiff
must produce
evidence
of
likely
anticompetitive
effects.
In
a
jury
trial,
this
burden
of
production
is
sustained
by
adducing
evidence sufficient
to
allow
a
reasonable
jury
to
infer
the
likely
anticompetitive
effects
of
a
merger;29
in
the
typical
Section
7
bench
trial, the standard
may
be
higher.30
Framed
in
the
negative,
the
burden
of
production
is
sustained
when
the
plaintiff
has
proffered
enough evidence
to
survive
an
adverse
motion for directed
verdict
at
the
close
of
its
case
in
chief,
sometimes referred to
as
making
out
a
prima
facie
case.
31
Such
a
proof
may
be
constructed
with
either
direct
or
circumstantial
evidence.32
In
the
merger
context,
demonstrating
that
a
merger
would
result
in
undue
market
concentration
exemplifies
the
latter mode
of
proof.
As
the
preceding
discussion
of
economic
thinking
explains,
evidence
that
a
merger
will cause
a
substantial
increase
in
market
concentration supports
a
substantive (economic)
inference
that
the
merger
is
likely
to
have
anticompetitive
effects.
33
There
is
no
settled
terminology
to
describe
this
type
of
substantive
factual
inference,
which
will
be
referred
to
in
this
Article
as
simply
a
substantive
inference.
In
other
contexts
and
jurisdictions,
the
same
idea
may be termed
a
circumstantial
inference,
a
permissible
inference,
a
presumed
factual
inference
supported
by
evidence
of
market concentration,
a
presumption
in
the form
of
res
ipsa
loquitur,
34
or
simply
a
presumption
as
the term
is
used
in
common parlance.
Regardless
of
the
language
used,
the
effect
of
the
plaintiff
sustaining
the
burden
of
production
depends
on the
strength
of
the
showing.
If
the
proof
of
anticompetitive
consequences
is
so
strong
that
a
verdict
would
be
directed
in
the
plaintiffs
favor
if
the
defendant
did not
come
forward
with
contrary
evidence,
then
the
demonstration
could
be
said
to shift
the
burden
of
production
to the
defendant.
35
If
the
plaintiffs
proof
permits,
28.
See
supra
note
9
(commenting
that
inconsistent
use
of
the
term
presumption
has
by
some
accounts
rendered
the word
"all
but meaningless").
29. See,
e.g.,
2
MCCORMICK
ON
EVIDENCE
§
338
(Kenneth
S.
Broun
et
al.
ed.,
7th
ed.
2013)
(describing
the
burden
of
production).
30.
In
a
bench
trial,
the
court
is
not
generally
bound to
the "reasonable
jury"
standard
in
making
this
determination. Cf
FED.
R.
Civ.
P.
52(c)
and
advisory
committee's
notes
on
the
1991
amendment.
31. See,
e.g.,
MCCORMICK
ON
EVIDENCE,
supra
note
29,
§
342,
at
676,
n.4
(describing
the
use
of
the
phrase
"prima
facie").
32. E.g.,
id.,
§
338,
at
653-56.
33.
See
supra
notes
15-21
and
accompanying
text
(discussing
the evolution
of
economic
thinking
on
the
relevance
of
changes
in
market concentration
in
predicting
the
competitive
effects
of
horizontal mergers).
34.
MCCORMICK
ON
EVIDENCE,
supra
note
29,
§
342, at
678-79
(distinguishing
this inference
from
a
true
presumption);
KENNETH
S.
ABRAHAM,
THE
FORMS
AND
FUNCTIONS OF
TORT
LAW
95-102
(3d
ed.
2007)
(questioning
the
doctrine
of res
ipsa
loquitur
as little
more
than
a
description
of
circumstantial evidence
sufficient
to
satisfy the burden
of
production);
see
also
Daniel
J.
Pylman,
Res
Ipsa
Loquitur
in
the
Restatement
(Third)
of
Torts:
Liability
Based
Upon
Naked
Statistics
Rather
Than
Real
Evidence,
84
CHI.-KENT
L.
REV.
907,
913
(2010)
(categorizing
the
res
ipsa
loquitur
proposition
as
a
probabilistic
inference);
Wex
S.
Malone,
Res
Ipsa
Loquitur
and
Proof
by
Inference-A
Discussion
of
the
Louisiana
Cases,
4
LA.
L.
REV.
70,
70-72
(1941)
(noting
that
res
ipsa
loquitur
has little to
distinguish
it
from
a
standard circumstantial
inference).
35.
MCCORMICK
ON
EVIDENCE,
supra
note
29,
at
657.
Prior
to
its
amendment
in
1991,
Federal
Rule
of
Civil
Procedure
41(b) required the
court
in
a
non-jury
trial to weigh
the
plaintiffs
evidence
on
the merits
on a
412
[Vol.
42:2
What
Structural
Presumption?
without
compelling,
the
inference
of
likely anticompetitive consequences, then
the
defendant's
obligation
to
produce
contrary evidence
is
merely
the
risk
that
the
plaintiffs
argument
will
be
found
persuasive
if
left
unrefuted.
This
obligation
is
sometimes
referred
to
as
the
burden
ofgoing
forward
with
the evidence.
36
2.
Rebuttable
(Burden-Shifting)
Presumptions
The
evidentiary
role
of
substantive
inferences-logical
conclusions
drawn from
evidence
of
independent
probative
value-contrasts
with that
of
formal
rebuttable
presumptions.
The latter
are
sometimes
referred
to
as
burden-shifting presumptions,
or
"true
presumptions"
in
contrast
to
substantive
inferences.
37
Unlike
substantive
inferences,
where
the
probative
value
of
the
underlying
evidence
is
the
thing
that
shifts
the
burden
to the
defendant,
rebuttable presumptions
are
structures
of
legislative
or
judicial
creation
that
shift the
burden
of
production
by
rule
of
law.
In
a
rebuttable
presumption
framework,
producing
evidence sufficient
to
prove
some
basic
fact
has
the
legal
effect
of
causing
a
presumed
fact
to
be
compulsorily inferred
unless
rebutted
by
the
party against
whom the
presumption
operates.
8
Apart
from
the
probative
value
of
the
basic
fact evidence,
considerations
of
fairness,
access
to
proof,
procedural
ease,
and
social
policy
guide
the
construction
of
rebuttable
presumptions
39
with
the
effect
that basic
facts
sufficient
to
activate
rebuttable
presumftions
may have.-
little
to
no
independent
value
as
substantive
proof
of
the
presumed
facts.
If
the
defendant
fails to
come
forward with
evidence
of
its
own
at the
close
of
the
plaintiffs
case
in
chief,
the
operation
of
a
rebuttable presumption
is
to
establish
a
potentially
compulsory
inference.
Proof
of
the
basic
fact
in
the
presumption
requires
the
fact-finder
to
infer
the
presumed
fact
and anything
following
from
it.41
For
example,
under
an
old
common-law
presumption,
proof
that
a
person disappeared
and
has
been
motion
for
dismissal.
See
Brief
for
Pabst Brewing Company at
*23,
United
States
v.
Pabst
Brewing
Co.,
384
U.S.
546
(1966)
(No.
404),
1966
WL
115448
(summarizing
how the rules
were
amended).
Under
this
standard,
the
plaintiffs
satisfaction
of
the
burden
of
production
would effectively shift
the
burden
of
production
to
the
defendant.
36.
MCCORMICK
ON
EVIDENCE,
supra
note
29,
at
657.
37. Id.
§
342.
38.
The
language
of
basic
facts
and
presumed
facts
is
ostensibly
due to
Edmund
M.
Morgan,
Foreword
to
MODEL
CODE
OF
EVIDENCE
52-54
(AM.
LAW
INST.
1942).
39.
See
MCCORMICK
ON
EVIDENCE,
supra
note
29,
§§
337,
343
(discussing
various
policy objectives that
might
motivate
the
creation
of
a
rebuttable
presumption).
40.
See,
e.g.,
WRIGHT
ET
AL.,
supra
note
9,
at
n.78
("[The]
presumption
has
no
probative
value
but
merely
allows
factfinder
to
reach
conclusion
in
absence
of
proof
to
the
contrary.")
(citing
Jones
v.
LSU/EA
Conway
Med.
Ctr.,
46
So.
3d 205,
211
(La.
Ct.
App.
2010)).
But
ef
Pennzoil
Co.
v.
Fed.
Energy
Regulatory
Comm'n,
789
F.2d
1128, 1137
(5th
Cir.
1986)
(noting that
the
presumed
fact
could
still
be
inferred from
the
basic
fact
if
an
independent probative
basis existed
to
support
the
inference).
41.
See
WRIGHT
ET
AL.,
supra
note
9,
at
422
(citing
CHARLES TILFORD
MCCORMICK,
HANDBOOK
OF THE
LAW
OF
EVIDENCE
§
310
(1954));
EDMUND
MORRIS MORGAN,
BASIC
PROBLEMS
OF
EVIDENCE
41-42
(1961);
JAMES BRADLEY
THAYER,
A
PRELIMINARY
TREATISE
ON
EVIDENCE
AT
THE
COMMON
LAW
336
(1898);
9
WIGMORE
ON
EVIDENCE
§
2490
(3d ed.
1940));
cf
CHRISTOPHER
B.
MUELLER
&
LAIRD
C.
KIRKPATRICK,
1
FEDERAL
EVIDENCE
§
3:7 (3d
ed.
2012)
(noting
a
distinction
between
a
presumption
that affects only the
burden
of
production,
a
presumption
that affects
the
burden
of
persuasion,
and
various intermediate
possibilities).
413
2016]
The
Journal
of
Corporation
Law
absent
and
unheard
of
for seven
years
(the
basic
fact)
would
require
a
finding
that
the
person
has
died
during
this
absence
(the
presumed
fact),
unless this conclusion
is
rebutted
by
some
contravening
explanation
for
the
absence.
42
A
rebuttable
presumption
activates
upon the
production
of
evidence
of
the
basic
fact.
Confusingly,
this
is
often
referred
to
as
a
prima
facie
showing.
43
Activation
is
said
to shift
the
burden
of
production
from
the
plaintiff
to
the
defendant, but
the actual
effect
is
a
bit
more
subtle.
At the
close
of
the
plaintiffs
case
in
chief,
proof
of
the
basic fact
is
still
generally
a
fact
question,
and the
compulsory
inference
never vests
if
the
fact-finder
is
not ultimately
persuaded
of
the
basic
fact.
Unless
proof
of
the
basic
fact
is
overwhelming,
evidence
of
the
basic
fact
in
a
rebuttable
presumption
thus operates more
like
the
shift
of
a
burden
of
going
forward than
as
a
true
shift
of
the
burden
of
production,
at
least
as
the terms
are
used
above.
44
Little
changes
if
the
defendant
seeks
to
rebut
the
presumption
by
disproving
the
basic
fact.
Unless
the
defendant's
rebuttal evidence is
overwhelming,
the
basic
fact
remains
a
fact
question
informed
by
both
the
plaintiff's
and
defendant's
presentations
of
evidence.
45
As
before,
a
finding
of
the
basic
fact
compels
a
finding
of
the
presumed
fact
and
anything
it
implies
under
the
substantive
law.
Also
as
before,
a
finding
against
the
basic
fact
establishes
no
presumption
in
the
first
place, and absent
some
other
evidence
in
demonstration
of
the
presumed
fact,
the
plaintiff
will
likely
suffer
an
adverse directed
verdict.
If
the
defendant
instead
seeks to
rebut
the
presumption
by
disproving
the
presumed
fact,
things
become
more
complicated. Courts
and
scholars
disagree
on
what
effect
evidence
in
rebuttal
of
the
presumed
fact has
on
a
rebuttable
presumption.46
In
the
majority
of
jurisdictions
following
the
'Thayer-Wigmore
school
of
thought,
the
defendant's
satisfaction
of
its
burden
of
production
with
any
evidence
contrary
to
the
presumed
fact
destroys the
presumption
entirely.
47
This
is
sometimes
referred
to
as
the
"bubble-bursting"
theory
of
rebuttable
presumptions.48
42.
See
MCCORMICK
ON
EVIDENCE,
supra
note
29,
§
343,
at
686-87,
n.18-21;
Frances
T.
Freeman Jalet,
Mysterious
Disappearance:
The
Presumption
ofDeath
and
the
Administration
ofthe
Estates
ofMissing
Persons
or
Absentees,
54
IOWA L.
REV.
177,
207
(1968).
43.
BRYAN
A. GARNER,
"prima
facie
case",
GARNER'S
DICTIONARY
OF
LEGAL
USAGE
704
(3d ed.
2011).
44.
See
supra
notes
35-36
and
accompanying
text
(describing
the
burden-shifting
effect
of
producing
basic-fact
evidence
in
a
rebuttable
presumption).
45.
See
WRIGHT
ET
AL.,
supra
note
9,
at
427
("[U]nless
the
defendant's
evidence
is
so
overwhelming
that
the
judge
must
direct
a
verdict
for
him
on
the
issue,
all
contrary
evidence
on
the basic
fact does
is
to
create
a
jury
question
on
the
issue
of
the
existence
of
the
presumption.").
46.
See
Edward
W.
Cleary,
Presuming
and
Pleading:
An
Essay
on
Juristic
Immaturity,
12
STAN.
L. REV.
5,
15-20
(1959)
(exploring
in
detail
the
different
standards
by
which
a
rebuttal
argument
might
be
assessed).
47.
JAMES
B.
THAYER,
A
PRELIMINARY
TREATISE
ON
EVIDENCE
AT
THE
COMMON
LAW
336-39
(1898);
WIGMORE
ON
EVIDENCE,
supra
note
41,
§
2491;
see
MCCORMICK
ON
EVIDENCE,
supra
note
29,
§
344,
at
692-
95.
48.
See
FED.
R.
EVID.
301,
Advisory
Committee Notes
to
the
1972
Proposed
Rules
(citing
Edmund
M.
Morgan
&
John
MacArthur
Maguire,
Looking
Backward
and
Forward
at
Evidence,
50
HARV.
L.
REV.
909,
913
(1937));
MCCORMICK
ON
EVIDENCE,
supra
note
29,
§
344,
at
692;
D.
Craig
Lewis,
Should
the
Bubble
Always
Burst?
The
Need
for
A
DifJerent
Treatment
ofPresumptions
Under
IRE
301,
32
IDAHO
L.
REV. 5,
6 (1995)
("[Sustaining
the
burden
of
production
on
rebuttal
means
the]
'bubble
bursts'-
the
presumption
disappears,
and
the factual issue addressed
by
the
presumption
is
decided
based solely
on
the
evidence presented
414
[Vol.
42:2
What
Structural
Presumption?
In
jurisdictions
following
the
competing
Morgan-McCormick
school
of
thought,
rebuttable
presumptions
are
more
durable and actually shift
the
burden
of
persuasion
to
the
defendant
on
the
issue
of
the
presumed
fact.
49
In
this
setting,
the
defendant's
satisfaction
of
its
burden
of
production
by
producing
evidence
contrary
to the
presumed
fact only
establishes
a
fact
question
on
the
presumed
fact.
50
Finally,
in
federal
courts,
Rule
of
Evidence
301
provides
a
statutory procedure for
the
treatment
of
rebuttable
presumptions
in
civil
suits:
"the
party
against whom
a
presumption
is
directed
has
the
burden
of
producing
evidence
to
rebut
the
presumption.
But this
rule does
not
shift
the
burden
of
persuasion,
which
remains
on the
party
who
had
it
originally."51
Rule
301
is
often
cited
as
a
"bubble-bursting"
rule,52
but
the
legislative
history
of
this rule
leaves
substantial room
for
interpretation.
53
If this
brief
overview
of
the
procedure
of
rebuttable
presumptions
is
dense and
difficult
to
follow, then
it
aptly
conveys
the
reality
of
the
subject matter. The
apparent
simplicity
of
the
rebuttable
presumption
framework belies exhausting
complexity
in the
details.
D.
Competing
Interpretations
McCormick famously complained that
"'presumption'
is
the
slipperiest
member
of
the
family
of
legal
terms, except
its
first cousin,
'burden
of
proof."'
54
This
sentiment
aptly characterizes
the
state
of
the structural
presumption
in
horizontal
merger
analysis.
The
remainder
of
this
Article
addresses
a
latent ambiguity
in
current
applications
of
the
structural
presumption: namely,
uncertainty
about
what kind
of
presumption
the
structural
presumption
entails.
The
modem
structural
presumption
framework,
articulated
by
the
D.C.
Circuit
in
Baker
Hughes,
arguably
fits
the
burden-shifting
model
of
a
rebuttable
presumption
under
Rule
301.
5
One
possibility
is
thus
that
the
structural
presumption
is
a formal
burden-
shifting
rebuttable
presumption.
By
the
language
of
modem
merger
cases,
one
would be
forgiven
for
thinking
this
interpretation
entirely
beyond
dispute.
concerning
the
issue.").
49.
Edmund
M.
Morgan,
Some
Observations Concerning Presumptions,
44
HARV.
L.
REV.
906,
929
(1931);
Morgan
&
Maguire,
supra
note
48, at
912-13.
50.
See
MCCORMICK
ON
EVIDENCE,
supra
note
29,
§
344
(describing
the
operation
of
presumptions
in
civil cases).
51.
FED.R.EvID.301.
52.
See
Pennzoil
Co.
v.
Fed.
Energy
Regulatory
Comm'n,
789
F.2d 1128,
1137
n.24
(5th Cir.
1986)
(collecting
circuit opinions interpreting
Rule
301
as
a
"bubble-bursting"
analogy);
see also
G.
Michael
Fenner,
Presumptions:
350
Years
of
Confusion
and
It
Has
Come
to
This,
25
CREIGHTON
L.
REv.
383,
389
(1992)
(describing
Rule
301
as
the
"bursting bubble"
rule); Lewis,
supra
note
48, at
6
(identifying
Rule
301
as
the
"bubble
bursting"
approach).
53.
As
Congress intervened
to
alter
the
original
Morgan-McCormick
style
of
presumption
adopted
by
the
Supreme Court,
the
meaning
of
Rule
301
is
arguably
best
understood
by
reference
to
the
Conference Committee
Report
that
lead
to
the
altered
rule.
H.R.
CONF.
REP.
No.
1597,
1974
U.S.C.C.A.N.
7098,
7099.
For
careful
discussion
and
interpretation,
see
generally
WRIGHT
ET
AL.,
supra
note
9,
§
5122.2;
MUELLER
&
KIRKPATRICK,
supra
note
41,
§
3:1.
54.
MCCORMICKON
EVIDENCE,
supra
note
29,
§
342.
55.
United
States
v.
Baker Hughes
Inc.,
908
F.2d
981,
982
(D.C.
Cir.
1990).
415
2016]
The
Journal
of
Corporation
Law
However,
this
Article argues
that
the
better
understanding
of
the
structural
presumption
is
as
a
simple substantive
inference. The
next
Part
of
this
Article revisits
the
case
law
history
of
the
structural
presumption
to
argue
that
the
still-controlling
authority
of
Philadelphia
National
Bank
actually
articulates
a
substantive
inference-not
a
rebuttable
presumption.
Parts
IV and
V
defend
the
substance
of
this
interpretation
on
positive
and
normative
grounds,
respectively.
Ill.
THE
CASE
LAW
HISTORY
Context
for
any
review
of
horizontal
merger
case law
is
the
Supreme
Court's
effective
abandonment
of
this
subject
matter after
a
period
of
activism
in
the 1960s and
early
1970s.
While
lower
courts
have
continued
to
evolve
this
area
of
law, old
Supreme
Court
cases
are
still
controlling
authority
on
much
of
the
modern
framework.
Given
the
frequency
with
which
it
is
cited
for
the
proposition,
one
would
thus
suppose
that
Philadelphia
National
Bank
put
forth the
modem burden-shifting
structural
presumption.
But
as
this
Part
shows, the
presumption
that
Philadelphia
National
Bank
articulates
is
actually
a
simple
substantive
inference.
In
reviewing
the
case
law
history
of
the
structural
presumption,
the
burden-shifting
interpretation does
not
appear
until
the
late
1980s
or
early
1990s,
and
its
leading
expression
in
Baker
Hughes
is
essentially
novel
law.
A.
The
1960s
(Philadelphia National Bank)
Like
so
much
of
modem horizontal
merger
analysis,56 the
history
of
the
structural
presumption
starts
with the
1962
case
of
Brown
Shoe
Company
v.
United
States.
57
This
case, the
Court's
first
commentary
on the
role
of
market
concentration
evidence
under
the
revised
Section
7,
is
important
mainly
in
the
negative
sense
of
what
it
doesn't
say.
Brown
Shoe
is
deferential
to
market
concentration
evidence,
but
only
in
the
sense
of
recognizing
it as
one
important
factor
among
many.58
In
terms
of
its
contribution
to
the
structural
presumption,
Brown
Shoe
can
be
read
for
two
propositions.
First,
the
probative
value
of
market
concentration
evidence
is
derived
in
large part
from
economic
reasoning.
The
opinion
cites
Stigler
for
economic
confidence in
the strong
relationship between high market concentration
and the
likely
anticompetitive
consequences
of
a
merger.59
Second,
market concentration
evidence
is
an
important
factor among others
to
be
considered
in
the
review
process.
In
a
footnote,
the
opinion
asserts
that
"[s]tatistics
reflecting
the
shares
of
the
market controlled
by
the
industry
leaders
and the
parties to
the
merger
are,
of
course,
the
primary
index
of
market
power."60 But
the
opinion
immediately qualifies this
statement
by
going
on
to
caution
56.
See,
e.g., Robert
A.
Skitol
&
Kenneth
M.
Vorrasi,
The
Remarkable
50-Year Legacy
of
Brown Shoe
Co.
v.
United
States,
26
ANTITRUST
47,
47
(2012)
(identifying
this
case as
"the most
important merger
law
decision
ever
issued").
57.
Brown
Shoe Co.
v.
United
States, 370
U.S.
294
(1962).
58.
See
4
PHILLIP
E
AREEDA
&
HERBERT
HOVENKAMP,
ANTITRUST
LAW: AN
ANALYSIS OF
ANTITRUST
PRINCIPLES
AND
THEIR
APPLICATION
I
902c
(3d
ed.
2006)
(providing a
similar
interpretation).
59.
Brown
Shoe,
370
U.S.
at 332
n.56,
334
n.61
(citing
Stigler,
Mergers
and
Preventive
Antitrust
Policy,
supra
note
15).
60.
Id.
at
322
n.38.
416
[Vol.
42:2
What
Structural
Presumption?
that
market
share
statistics
do
not
exist
in
a
vacuum and
that
full
examination
of
a
market's
"structure,
history
and
probable
future"
is
needed
to
determine
the
likelihood
of
anticompetitive
effects
in
Section
7
analysis.61
It
is
against
this
backdrop that
the
Court's
commentary
on
the
role
of
market
concentration
evidence
in
United
States
v.
Philadelphia
National
Bank
must
be
read.
62
As
previously
noted,
one
passage
from
Philadelphia
National
Bank
is
now accepted
as
the
settled
citation
for
the
structural
presumption.
The
full
passage
is
as
follows:
[I]ntense
congressional
concern
with the
trend toward concentration warrants
dispensing,
in
certain
cases,
with
elaborate
proof
of
market
structure,
market
behavior, or
probable
anticompetitive
effects.
Specifically,
we
think
that
a
merger
which
produces
a
firm
controlling
an
undue
percentage
share
of
the
relevant
market,
and
results
in
a
significant increase
in
the
concentration
of
firms
in
that
market
is
so
inherently
likely
to
lessen
competition
substantially
that
it
must
be
enjoined
in
the
absence
of
evidence
clearly
showing
that
the
merger
is
not likely
to have
such
anticompetitive
effects.
63
Today, this
statement
is
widely assumed
to
mean that evidence
of
undue
concentration activates
a
rebuttable
presumption
of
likely
anticompetitive
effects.
64
it
may
thus
come
as
a
surprise
that
the
case
contains
no
discussion
of
rebuttable
presumptions
whatsoever.
In
fact,
read
comprehensively
and
in
full
historical context,
Philadelphia
National
Bank
is
best
understood
as
simply
articulating
a
substantive
inference
that courts
would
be
permitted
to
draw
upon
finding that
a
merger
would
lead
to
undue
concentration.
The
probative
value
that
Philadelphia
National
Bank
attributes
to
evidence
of
undue
concentration
explicitly derives
from
economic
thinking
on
the
relationship between
market concentration and
anticompetitive
harm.65
The
primary
authorities
the
case
cites
for
the
structural
presumption
are
scholarly
works
by
Stigler,
Kaysen, Turner, and
Markham.66
Each
of
these
works
expresses confidence
in
the
probative
value
of
market
concentration
as
evidence
of
likely
anticompetitive
effects,
and
each
proposes
some
threshold
level
of
market
concentration
beyond
which
a
merger could
be
deemed
likely
to
have
anticompetitive
effects
even
without further
proof.
67
The
most
natural
reading
of
this
structural
presumption
is
as
a
permissible
substantive
inference. The inference
is
substantive
in the sense
that
the
likely
61.
Id.
62.
United
States
v.
Phila.
Nat'l
Bank,
374
U.S.
321
(1963).
63. Id.
at
363.
64.
See,
e.g.,
United
States
v.
United
Tote,
Inc., 768
F.
Supp.
1064, 1068
(D.
Del.
1991)
("If
it
demonstrates
that
the
merger
further
consolidates
an
already
highly
concentrated
market
for
a
given
product,
the
Government
establishes
a
rebuttable
presumption
that
the
merger
is
illegal
under
Section
7.").
65.
Cf
Thomas
E.
Kauper,
Influence
of
Conservative
Economic
Analysis
on
the
Development
ofthe
Law
ofAntitrust,
in
HOW
THE CHICAGO
SCHOOL
OVERSHOT
THE
MARK:
THE
EFFECT
OF CONSERVATIVE
ECONOMIC
ANALYSIS
ON
U.S.
ANTITRUST
43
(Robert Pitofsky
ed.,
2008)
(commenting
that
among
important antitrust
cases
of
the
1950s
and
1960s,
only
Philadelphia
National
Bank
rested
on
economic
analysis).
66.
Phila.
Nat'l
Bank,
374
U.S.
at
363-64
&
n.38-39,
41
(citing
Stigler,
Mergers
and
Preventive
Antitrust
Policy,
supra
note
15;
KAYSEN
&
TURNER,
supra
note
15;
Markham,
supra
note
15).
67.
Id.
at
364
n.41
(describing
the
market concentration
thresholds
suggested
by
different
economists).
417
2016]
The
Journal
of
Corporation
Law
anticompetitive
effects
of
a
merger
are
inferred
circumstantially
from
economic
reasoning
on the
value
of
market concentration
evidence.
The inference
is
permissible
in
the
sense
that
the
potential anticompetitive
consequences
of
a
merger
may
be
inferred
from
market concentration
evidence
without
further inquiry
into the
other
factors
discussed
in
Brown
Shoe.
Indications that
the
Court meant
to
draw
a
substantive
inference
from
market
concentration
evidence
are
evident
on
the
face
of
the
opinion.
For
example,
in
defining
the
test
of
undue
concentration,
the
Court
refers to
concentration
thresholds
proposed
by
economists
for
predicting
whether
a
merger
would
have
anticompetitive
effects.68
In
explaining
the
limits
of
the structural
presumption,
the
opinion further
states
that
it
"lightens
the
burden
of
proving
illegality only with
respect
to mergers
whose size
makes
them
inherently
suspect."69
Similarly,
in
its
findings
on
the
actual facts
of
the case, the
Court explains that
prior
case
law
and
the
above-noted
economic authorities
support
"the
inference
we
draw
in
the
instant
case
from
the [market
concentration]
figures
disclosed
by
the
record."
70
By
contrast,
a
number
of
considerations
militate
against reading
Philadelphia
National
Bank
as
creating
a
burden-shifting rebuttable presumption.
First,
nothing
in the
opinion expressly
states
the
creation
of
a
rebuttable
presumption.71
The
Court
never
uses
the
word
"presumption"
and
never refers
to
burden-shifting
of
any
form.
The
absence
of
any
explicit
mention
of
presumptions
or
burden-shifting
would
be
only
natural
in
a
case
articulating
a
simple substantive inference
but
is
hard
to
reconcile
with an
opinion
purported
to
create
new
law
in
the
form
of
a
novel
rebuttable
presumption.
This
negative
inference
is
particularly
strong in light
of
the
Court's
articulate
discussion
of
rebuttable
presumptions,
the
rules
of
burden-shifting, and rebuttal
procedure
in
other
cases
of
the
72
time.
Second,
nothing
in
the opinion
even
implies
the
possible
creation
of
a
rebuttable
presumption.
The
closest
the
Court comes
is
an
ambiguous
statement
that "[tihere
is
nothing
in
the
record
of
this
case
to
rebut the
inherently
anticompetitive tendency
manifested
by
[the
market concentration
evidence]."
73
Of
course,
the
term
"rebut"
is
often
used
in
the
lay
sense
of
tending
to
disprove
an
adverse
inference,
74
and
this
use
of
the
term
better
fits the
statement's
explicit reference
to
the
inferred
(inherent)
tendency
of
68.
Id.
69.
Id.
at
363.
70.
Id. at 366
(emphasis
added).
71.
Cf
Paul
R.
Rice
&
Slade
S.
Cutter,
Problems
with
Presumptions:
A
Case
Study
of
the
"Structural
Presumption" of
Anticompetitiveness,
47
ANTITRUST
BULL.
557,
565-66
(2002)
(arguing
that
Philadelphia
National
Bank
articulates
a
per
se
rule
as
opposed
to a
rebuttable presumption).
72.
See,
e.g.,
Dick
v.
N.Y. Life
Ins.
Co.,
359
U.S.
437,
443
(1959)
("Proof
of
coverage
and
of
death
by
gunshot
wound shifts
the burden
to the
insurer
to
establish that the
death
of
the insured
was due
to
his suicide.
Under
[applicable state
law],
this
presumption
does
not
disappear
once
the
insurer
presents
any
evidence
of
suicide. Rather,
the
presumed
fact
(accidental
death)
continues, and
a
plaintiff
is
entitled
to
affirmative
instructions
to
the
jury
concerning
its
existence
and weight.");
id.
at 443 n.4
(noting contemporaneous
debates
about the procedural
implications
of
rebuttable
presumptions
and
the effects
of
producing
rebuttal
evidence).
73.
United States
v.
Phila.
Nat'l
Bank,
374
U.S.
321,
366
(1963).
74.
See,
e.g.,
GARNER,
supra
note
43,
at
286
(defining
"disprove;
refute;
confute;
rebut;
controvert");
cf
id.
at
754
(defining
"rebuttable
presumption").
418
[Vol.
42:2
What
Structural
Presumption?
undue concentration
to lead to
anticompetitive
effects.
Nor
does
the
idea
of
a
rebuttable
presumption
enter
by
incorporation.
Other
than
Brown Shoe,
the
only other
legal
authority
the
Court
cites
in
the
structural
presumption
passage
is a
district
court
opinion:
United
States
v.
Koppers
Company.
75
This
case
discusses
the
legislative intent
of
Section
7
at length,
but
contains
no
mention
whatsoever
of
market
concentration
or
rebuttable
presumptions.
Third,
the
prospect
that
Philadelphia
National
Bank
creates
a
novel
burden-shifting
presumption
is
inconsistent with
the
opinion's
self-description. The
Court expressly
denies that
it
does
more
than
simply
apply
the
test
of
illegality
announced
in
Brown
Shoe,
stating
that
"we
analyzed
the
test
[of
merger
legality]
in
detail
in
[Brown Shoe],
and
that
analysis
need
not
be
repeated
or extended
here,
for
the
instant
case
presents
only
a
straight-forward
problem
of
application
to
particular
facts."
76
As
noted
above,
Brown
Shoe
treats
market concentration
evidence
as
supporting
an
important
substantive
inference
of
the
likely
competitive consequences
of
a
merger but
does
not
contemplate
a
burden-shifting presumption
of
illegality.
77
Similarly, the
Court's
description
of
the
structural
presumption
as
a
way
to
"simplify
the
test
of
illegality"
is
reconcilable
with
general
deference
to the
test
of
Brown
Shoe
and
the
need for
Section
7
analysis
to
be
"based
upon
a
firm
understanding
of
the
structure
of
the
relevant
market"
only
if
understood
as
a
permissible
substantive
inference.
Fourth,
subsequent
merger
cases
in
the
1960s
do
not
bear
out the
tacit
creation
of
a
rebuttable
presumption
in
Philadelphia
National
Bank.
For
example,
in the five
merger
cases to
reference
Philadelphia
National
Bank
in
the
year
following
its
announcement,
the
opinion
is
mainly cited for
the
possibility
of
dual
regulation under
the
antitrust
laws
79
and
for language
on the
importance
of
curbing
concentration in
already
heavily
concentrated
markets.
80
The
closest
the
Court
ever
comes
to the
idea
of
a
rebuttable
presumption
is an
isolated statement
in
United States
v.
Continental
Can
to
the
effect
that
the
combined
share
of
the
mer
ing
firms
"approaches
that
held presumptively
bad
in
[Philadelphia
National
Bank]."
This and
one
other
use
of
the
word
"presumption"
in
a
footnote
in
one
of
Justice
Harlan's
many
dissents,82
provide
no
serious
basis
for
thinking
that
the
structural
presumption
was
treated
as
more
than
a
substantive inference.
The
same
is
true
of
other
important
merger
cases
of
the
1960s.
In
United
States
v.
Von's
Grocery,
for
example,
no
mention
is
made
of
presumptions,
rebuttals,
or
burden-
shifting.83
The
opinion
of
the
Court
in
United
States
v.
Pabst
Brewing
is
similarly devoid
75.
United
States
v.
Koppers
Co.,
202
F.
Supp.
437 (W.D.
Pa.
1962).
76.
Phila.
Nat'l
Bank,
374
U.S.
at
355.
77.
See
supra
notes
59-61
and
accompanying
text
(describing
how
Brown Shoe
characterized
the
role
of
market concentration
evidence
in
competitive
effects
analysis).
78.
Phila.
Nat'l
Bank,
374
U.S.
at
362.
79.
United
States
v.
El
Paso
Natural Gas
Co.,
376
U.S.
651
(1964);
United
States
v.
First
Nat'l
Bank
&
Tr. Co.
of
Lexington,
376
U.S.
665
(1964).
80.
United
States
v.
Aluminum
Co.
of
Am.,
377
U.S. 271
(1964);
United
States
v.
Penn-Olin
Chem.
Co.,
378
U.S.
158
(1964).
81.
United
States
v.
Cont'l
Can
Co.,
378
U.S.
441,
461
(1964).
82.
First Nat'l
Bank
and
Trust,
376
U.S.
at 676
n.6
(Harlan,
J.,
dissenting)
(referencing
without
elaboration
"the
'presumption'
that
the
Court
laid
down
in
Philadelphia National
Bank").
83.
United States
v.
Von's
Grocery
Co.,
384
U.S.
270
(1966).
2016]
419
The
Journal
of
Corporation
Law
of
any
mention
of
rebuttable
presumptions.
84
Both
of
these
cases
proscribed
mergers
of
firms
with
tiny market
shares
by
modem
standards and
few
would endorse
their
conclusions
today.
But
however
objectionable
their
extreme
hostility
to
increasing
market
concentration
may
now
be,
85
there
is
nothing
in
the language or
reasoning
of
even
these
now infamous
cases
to
suggest
that
the
holdings
rested
on
more
than
a
strong substantive
inference
of
the
likely
anticompetitive
effects
of
increased market concentration through
merger.
To
summarize,
the
structural
presumption
of
Philadelphia
National
Bank
is
not
a
rebuttable
burden-shifting
presumption. Taken
as
a
whole-the
argument
and
language
of
the
case,
its
purported
objective,
and
its
application
in
subsequent
cases-all
indications
are
that
the structural
presumption
is
a
permissible
substantive
inference.
To
be
specific,
Philadelphia
National
Bank
does
not
vary the
import
Brown
Shoe
assigns
to
the
probative
value
of
market
concentration
and other
factors. Rather,
it
recognizes
a
permissible
inference
of
illegality
where
market concentration
evidence
is
so
overwhelming
that
it
obviates
the
need
for
the
plaintiff
to
produce
further
proof
of
the
likely
anticompetitive
effects
of
a
merger.
B.
The
1970s
(General
Dynamics)
Supreme
Court
cases
in
the
1970s
changed
the language
of
the
structural
presumption
by
describing
proof
of
undue
concentration
as
a
prima
facie
showing. For
example,
in
United
States
v.
Citizens
and
Southern
National
Bank,
the
Court
commented
that
by
providing
evidence
of
undue
concentration,
"the
Government plainly
made
out
a
prima
facie case
of
a
violation
of
[Section]
7."
86
The
leading
case
for
this
use
of
prima
facie
language
in
discussion
of
the
structural
presumption
is
United
States
v.
General
Dynamics.
87
But
despite the change
in
language
it
precipitated,
General
Dynamics
does not
appear
to
vary
the substance
of
the
structural
presumption
as
articulated
in
Philadelphia
National
Bank.
General
Dynamics
describes
cases
of
the
1960s
as
having
found
"prima
facie
violations
of
[Section]
7
of
the
Clayton Act
from
aggregate
[market concentration]
statistics,"
the
effect
of
which
is
"to
allow
the
Government
to
rest
its
case
on
a
showing
of
[undue
concentration]."
The
meaning
of
prima
facie
language
is
ambiguous,
and
this
description
is
technically
consistent
with
both
a
substantive
inference sufficient
to
sustain
the
plaintiffs
burden
of
production
and
a
rebuttable
presumption
that causes
a
legal
shift
in
the
burden
of
production.
Taken
as
a
whole, however,
any
ambiguity
in
General
Dynamics
resolves
in
favor
of
continuing
the
substantive
inference
interpretation
of
Philadelphia
National
Bank.
Both
the
majority
opinion and
the
dissent
are
devoid
of
any
mention
of
presumptions, burden-
84.
United
States
v.
Pabst
Brewing
Co., 384
U.S.
546
(1966).
85.
Cf
Baker
&
Shapiro,
supra
note
1,
at
237
(noting
the
extremely low
concentration
figures
proscribed
in
these
cases
and suggesting
that
modem
economic
thinking, though
sensitive
to
the
importance
of
market
concentration evidence
generally,
would
not support
the
apparent
reasoning
of
these
cases).
86.
United
States
v.
Citizens
&
S.
Nat'l
Bank,
422
U.S.
86,
120
(1975).
87.
United
States
v.
Gen.
Dynamics
Corp.,
415
U.S.
486
(1974).
88.
Id.
at
496.
89.
Id.
at
497.
420
[Vol.
42:2
What
Structural