FEDERAL TRADE COMMISSION

Agency decision

Ask Donna

What actually matters in this document.

Text

I N D E X

OPENING REMARKS

PAGE

CHRISTINE VARNEY

PANELS

PAGE

139

183

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

FEDERAL TRADE COMMISSION

)

WORKSHOP ON:

)

HORIZONTAL MERGER GUIDELINES

)

Matter No.

REVIEW PROJECT

)

P092900

)

---------------------------------)

TUESDAY, JANUARY 26, 2010

Conference Center

Federal Trade Commission

601 New Jersey Avenue, N.W.

Washington, D.C. 20580

The above-entitled hearing was held, pursuant

to notice, at 9:00 a.m.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

P R O C E E D I N G S

MR. SHELANSKI:

Good morning.

Thank you very

much for coming to this fifth of our Horizontal

Guidelines Workshops.

started, one a chore and one a pleasure.

to read you a security briefing, which we're required to

read, and the pleasure is to introduce our opening

speaker.

I have two tasks before we get

The chore is

So here's the security briefing.

Anyone that

goes outside the building without an FTC badge will be

required to go through the magnetometer and x-ray

machine prior to re-entering into the conference center.

In the event of a fire or evacuation of the building,

please leave the building in an orderly fashion.

worries me with this crowd.

building, you need to orient yourself to New Jersey

Avenue.

That

Once outside of the

Across from the FTC is the Georgetown Law

Center.

rallying point.

not relevant to us.

person accounting for everyone in the conference center.

That person will make themselves known to you.

Look to the right front sidewalk.

That is our

Everyone will rally by floors.

That's

You need to check in with the

In the event that it is safer to remain inside,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

you will be told where to go inside the building, and if

you spot suspicious activity, please alert security, and

that does not refer to my colleague Phil Weiser's

remarks.

Towards the end of each panel or during each

panel, somebody will distribute these question cards.

We ask you, please, to fill them out, and they will be

brought up front.

most effective way to get questions.

the audience, however, we may opt to just take your

questions.

people here.

That has tended, we think, to be the

Given the size of

It might be simpler, given the number of

Finally, there will be a lunch break from 12:30

to 2:00.

here, but there are a couple of cafes.

a left out the front door, there are a couple of places

to eat on that block, and then there's Union Station

just a couple blocks away.

closest bets.

The offerings are not enormous right around

If you just make

Those are probably your

Finally, I would like to introduce Assistant

Attorney General Christine Varney, of course, she needs

no introduction, who will give us some opening remarks

today, and just so you all know, all remarks today will

be on a transcript that will be posted online within a

week or so.

Thank you.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

MS. VARNEY:

Thanks, Howard.

Good morning,

everybody.

Don't take any notes because my remarks are

going to be posted on the DOJ website at greater length.

I wanted this morning is talk about what we've learned

from the workshops and the comments we've received so

far and give you some preliminary views about what we've

heard during the process and where I, and only me,

believe consensus may be emerging.

As we complete the five workshops, we welcome

additional comments, including comments on the topics

that I'm going to outline this morning.

A consistent theme running through the panels is

that there are indeed gaps between the guidelines and

actual Agency practice, gaps in the sense of both

omissions of important factors that help predict the

competitive effects of mergers and statements in the

guidelines that may now be inaccurate.

something that we are all aware of.

These gaps are

The guidelines need to inform practitioners and

the business community of the Agencies' standards for

evaluating mergers.

what we actually do run counter to our goal of being

transparent.

accurate predictions about our likely enforcement

intentions adjust new behavior accordingly.

Gaps between what we say we do and

Transparency helps businesses make

Lack of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

transparency creates uncertainty, and uncertainty

results in unpleasant surprises.

We want to avoid that.

Similarly, the agencies rely heavily on the

merger guidelines in our competition advocacy efforts

both here and abroad.

pervasive, the guidelines must reflect our best thinking

about the competitive effects of mergers and appropriate

merger enforcement policy.

To be effective as well as

Courts also rely on the guidelines, in the words

of the Fifth Circuit, "to provide persuasive authority

when deciding if a particular acquisition violates

antitrust laws."

inaccurately reflect enforcement or omit crucial

considerations, we do a disservice to the law as well as

the business community.

When the guidelines either

At the same time, I do not want to overstate the

magnitude of these gaps.

whether a merger is likely to create or enhance market

power, resulting in anti-competitive effects, remains

the heart of merger analysis.

The focus in the guidelines on

The guidelines articulation of possible

unilateral and coordinated effects entry and

efficiencies accurately reflects the key concerns of

merger analysis.

that I did not envision radical review of the

I said at the outset of this project

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

guidelines.

workshops has changed my assessment.

guidelines, however, does appear to be worthwhile in a

number of areas.

Nothing so far in the comments or the

Updating the

Turning to some specifics, there are a few areas

where consensus appears to be emerging.

To begin, many

of our panelists have noted that the Agency's do not

mechanically apply the five-step process set forth in

the guidelines where markets and market shares are first

assessed, followed by a sequential consideration of

potentially adverse competitive effects, entry,

efficiencies, and then failing firm defenses.

our panelists advocated following that sequence as the

best way to either assess every merger's likely

competitive effects or to reach an enforcement decision.

None of

To be sure, the guidelines themselves offer a

note of caution regarding the potentially misleading

results that can follow from mechanical application of

the guidelines.

flexibility is both the norm of actual Agency practice

and appropriate given the diversity of considerations

that are presented in the range of transactions viewed

by the agencies.

Panelists have noted that far more

Thus, as a matter of actual practice and sound

theory, some adjustment of the description of the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

analytical process used by the Agency seems appropriate.

Implicit in deemphasizing the sequential nature of the

guidelines inquiry is a recognition that defining

markets and measuring market shares may not always be

the most effective starting point.

Remember the purpose of defining a market and

assessing shares is to assess the potential harm.

it is clear that either certain vulnerable customers are

likely to be harmed by a merger or that certain

customers have, in fact, been harmed by a consummated

merger, the need to define a market to assess likely

competitive effects is obviously diminished.

When

For instance, the consumer harm that followed

from the consummated Evanston Hospital transaction

lessened the importance of the Commission's market

definition and market share analysis.

have largely confirmed the view that market definition

should not be an end-all exercise.

something to be incorporated in a more integrated, fact

driven analysis directed at competitive effects.

Of course this is not news.

Our panelists

Rather, it is

The commentary on

the Horizontal Merger Guidelines explains that the

agencies apply the guidelines flexibly, and those

practicing before the agencies have been aware for some

time that market concentration is more important in some

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

cases than others.

new, disruptive entrant may well impact competition far

more than market shares might suggest.

For instance, a merger involving a

Similarly, the Division's merger review process

initiative has recognized the appropriateness of a

tailored second request schedule designed to enable the

division to take a quick look at potentially dispositive

issues such as failing firm and entry at the outset of

an investigation; thus precluding the need for full,

sequential review outlined in the guidelines.

acknowledging this flexibility in the guidelines

themselves seems to me to be prudent.

Expressly

The next area I would like to discuss is one

where the guidelines appear to be inaccurately

describing the Agencies' enforcement policy.

come as no surprise to you, the merger challenges data

that we collect confirms that it is rare for the

agencies to challenge mergers that will lead to HHI

concentration levels below 1,800.

indicate that such mergers potentially raise significant

competitive concerns.

It will

Yet the guidelines

Similarly, the guidelines suggest that a 100

point increase in HHI concentration level raises

anticompetitive concerns.

agencies have only infrequently, indeed I say rarely,

In actual practice, the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

challenged a merger unless they increased concentration

several times that much.

More broadly, our panelists have generally

confirmed that the guidelines overstate the importance

of HHIs in merger analysis.

you that HHIs have not been the focus of any party

presentation or any staff recommendation since I've been

the Assistant Attorney General.

the current state of merger analysis where HHI levels

are given far less prominent place as a predictive tool

for assessing competitive effects than suggested by the

guidelines.

Again, it will not surprise

That reality reflects

In that vein, I note that while many panelists

have acknowledged their usefulness as a tool for

assessing likely competitive effects, none has

maintained that HHIs should be the key driver in an

enforcement decision.

It is clear that the HHI threshold set forth in

the guidelines no longer capture Agency practice or

economic learning about the kinds of mergers that are

most likely to lead to consumer harm.

thresholds to express accurately how the agencies use

them seems not just appropriate but also necessary to

overcome what is at this point an affirmative

misstatement.

Revising the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

A third area where the guidelines may be

usefully updated is unilateral effects.

where economic thinking and Agency practice have

progressed significantly since 1992, when the concept of

adverse unilateral effects was first explicitly

introduced in the guidelines.

This is an area

That introduction was a major step forward, but

the treatment of unilateral effects was sparse, and

several of our panelists and commentators have noted

that significant advances in thinking have taken place

since 1992.

Unlike the HHI threshold where gaps are more in

the nature of misstatements, in unilateral effects, the

gaps are more in the nature of omissions.

important considerations that the agencies routinely

employ when assessing unilateral effects that are not

mentioned or even alluded to in the guidelines.

There are

Our panelists identified a number of

considerations routinely used to assess unilateral

effects.

reports, customer switching patterns, the views of

competitors, customers and industry observers, for

instance, are all tools we use to analyze mergers of

firms selling differentiate products.

Diversion ratios, price cost margins, win/loss

Yet the guidelines say little about how these

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

types of evidence are used to assess unilateral effects.

In fact, when assessing pricing effects in markets with

differentiated products, both agencies employ a variety

of techniques to evaluate whether or not the merger is

likely to lead to higher prices.

There is a growing body of evidence that

measures of upward pricing pressure, which focus on

diversion ratios and price cost margins, more accurately

evaluate the likelihood of unilateral pricing effects

where HHIs may be more productive in a coordinated

effects analysis.

Unilateral effects can arise along many

dimensions of competition, including pricing of

differentiated products, negotiations between buyers and

sellers, output and capacity for more homogeneous

products, product variety and innovation.

have accumulated a great deal of experience analyzing

such effects, and that expertise is not reflected in the

current guidelines.

The agencies

Updated guidelines can enhance transparency by

explaining how the agencies currently evaluate

unilateral effects.

areas where clarification of the guidelines appears to

be worthwhile.

I'll briefly mention five other

First, the discussion in the guidelines of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

targeted customers and price discrimination could be

clarified.

discrimination.

discrimination is quite abbreviated.

Many of our cases involve price

Yet the guidelines treatment of price

Second, the guidelines could more accurately

convey actual Agency practice by indicating that market

shares are normally assessed using recent or projected

sales in the relevant market while explaining the

conditions under which other measures, such as capacity,

may be used.

Third, different parts of the guidelines employ

closely related concepts of supply side responses by

non-merging firms:

in the relevant market, uncommitted entry, repositioning

and committed entry.

more unified approach to these concepts.

Expansion by firms already selling

A number of panelists suggested a

Fourth, the guidelines could clarify that

coordinated effects can arise through accommodating

behavior among a small number of rivals without the

necessity of reaching the terms of coordination.

Lastly, several panelists have pointed out that

the guidelines are virtually silent regarding

innovation, despite wide spread recognition innovation

generates enormous value for consumers over the long

run.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

A revision could move the guidelines into the

21st Century by explaining how the agencies account for

market dynamics, the pro-competitive role of disruptive

entrants and a merger's effect on innovation.

It's now time to turn to our expert panelists,

but I would first like to reiterate several notes of

thanks.

have submitted or will submit comments, thank you for

volunteering your time and expertise.

To all our panelists and to those who either

I would also like to offer my warm thanks to the

Federal Trade Commission staff and Antitrust Division

staff who have worked very, very hard to organize these

workshops, and I'm very appreciative for the very

wonderful and cooperative relationship the FTC and the

DOJ have enjoyed on this project, and I thank the FTC

for their hospitality today.

So I'm going to turn it over to my deputy, Phil

Weiser.

Thanks, everyone.

(Applause.)

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

PANEL 1:

PRESUMPTION.

MODERATOR:

General

PANELISTS:

MARK COOPER, Director of Research, Consumer Federation

of America

ALBERT A. FOER, President, American Antitrust Institute

ANDREW I. GAVIL, Professor of Law, Howard University

School of Law

CHARLES F. RULE, Partner, Cadwalader, Wickersham & Taft,

LLP

MARKET CONCENTRATION AND THE STRUCTURAL

PHIL WEISER, Deputy Assistant Attorney

MR. WEISER:

So thank you, Christine.

That was

a terrific way to kick off our last session, and thank

you so much for your engagement and support in this

effort.

panel to come up, hopefully sitting where your name card

is, I will start the process of introducing you all.

If I could ask the panelists for the first

The idea of this workshop in general was to get

a variety of perspectives, and there are many people who

said, Well, are there things you want us to say, and the

answer was, No, we want you to have a thoughtful,

engaging discussion, and with the folks we have here,

I'm very confident we will.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

Sitting to my left, Mark Cooper, who is one of

the premier consumer advocates, probably by all accounts

would be in the consumer advocate hall of fame.

the research director at the Consumer Federation of

America.

there, and I was at the Antitrust Division, and he is an

economist trained at Yale where he has a Ph.D. and is,

among other things, a fellow at the University of

Colorado's Silicon Flatirons Center.

He's

I first met Mark in the '90s when he was

Next to him, Bert Foer, who is what I guess you

would have to call a policy entrepreneur.

the American Antitrust Institute and has turned that

entity into a force.

not familiar with AAI, it brings together a bunch of

people on an advisory board as well as some resident

fellows to advocate on antitrust policy, and we're so

glad to have him here.

background in both private practice and also at the FTC

and in the industry.

He founded

It is a unique enterprise.

Those

Bert comes from a rich

Next to him Andy Gavil, who is one of the

leading lights in the academy, a law professor here at

Howard University.

antitrust case book with Bill Kovacic --

He is also a coauthor of an

MR. GAVIL:

And John Baker.

MR. WEISER:

-- and John Baker.

John is now

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

chief economist with the FCC formerly of the FTC, and

Bill Kovacic, of course, is a Commissioner at the FTC.

Finally, Rick Rule, who is one of the former

Assistant Attorney Generals, who has been so kind to be

supportive and engaging on this project.

of a long-standing place at the bar.

Cadwalader where he is head of their antitrust group.

He represents a number of major clients and was head of

the Antitrust Division in the '80s.

Rick has sort

He is now at

The folks we have gathered here are going to

talk about market concentration.

The way we're going to

do that is have a series of questions, have a give and

take, and I want to start with what Christine teed up

for us, which is we have these HHI figures, which many

acknowledge are, as Christine put it, a misstatement of

actual Agency practice, and also some would suggest out

of sync with economic learning.

question is:

makes it appropriate, and possibly if you would want to

supplement it, some have said maybe HHIs are not the

right framework.

I guess the broad

How do we think about this issue?

What

We could think instead of significant

competitors or market shares of the merging firms, what

have you.

think, were you there when the '82 guidelines were done

Rick, you've been around this for awhile.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

or the '84 guidelines?

MR. RULE:

I came shortly after the '82

guidelines were done and was there and in theory, sort

of the deputy, who was managing the process for

rewriting the guidelines in '84.

MR. WEISER:

What I heard is back then they were

using four firm concentration ratios, and they tried to

reverse engineer from that an HHI figure, and they came

up with 1,800, which has been -- was 1,800 taking it

back in the '80s and they kept the '92?

MR. RULE:

Well, what happened was in the '68

guidelines, the original guidelines, they used four firm

concentration ratios.

Werden, who is in the audience I think, is the source of

my information on this.

conversion in '82 was the introduction of what was

viewed at that time as a relatively revolutionary tool,

the HHI.

I wasn't there in '82, but Mr.

As I understand it, the big

As I understand it, Bill Baxter decided that it

made sense in making that change not to change the

thresholds, to largely have something that was similar.

Also the numbers, a thousand, 1,800, are nice round

numbers and sort of equate to certain things that make

sense, and so that's how that came about.

I should say it's also important to keep the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

historical perspective in mind.

performance paradigm, so to speak, was alive and well,

reflected in the courts, and in fact even, one might

say, more extremely so, and that's where the four firm

concentration ratios came from.

In '68, the structure

I can remember as a summer associate, in fact,

in the late '70s writing a paper that sort of summarized

the things other than structure that were relevant in

antitrust analysis and doing that for Ed Zimmerman, who

had also been an AAG, and he found that quite amazing,

that there were things that were relevant other than

structure.

In '82, I think it is fair to say that there was

the new learning that had occurred.

recognition that structure might be less important, but

there was still a strong sense within the Division, and

I think in Bill Baxter's case, that structure still was

the significant factor, and a lot of the analysis really

took the form of market definition issues, and of course

the focus in '82, in addition to the HHI, was the SSNIP

test, which of course was the principal lasting

innovation of the '82 guidelines.

There was a

In '84, there was a sense that -- and if you go

back to the '84 guidelines, you will see that the HHI

thresholds, even there, there's a line that sort of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

indicates they are at the beginning of the analysis.

There were some of us in the Division who thought we

should be even more explicit, that they really were safe

harbors as opposed to determinative guidelines, but

there was the sense that politically, that would not be

a fruitful exercise.

So instead of changing the numbers to reflect

what was emerging as the reality in '84, there was that

line.

analyses that have been done of mergers really are from

1990 forward, and if you go back into the 1980s and

around that time, there actually were mergers that were

getting scrutinized, and at times challenged, as I

recall it where the HHI post merger was below 1,800.

Now, I will say, if you go back, most of the

What I think was happening over that period of

time, and you can see it in some of the speeches, you

can actually see it in the international guidelines that

came out at that time, there was a recognition in the

Division that the notion that structure or market shares

were really a beginning and that what you had to do, and

the way I like to say it, is you have to tell a story.

Inevitably you have to focus on structure

because what a merger does is changes the structure of

the market, and what you have to decide is, as the

agencies do, whether or not that change in structure

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

makes it likely, in light of all of the relevant

circumstances, that prices may go up, and I think

through the '80s there was a recognition that it was

more complex than just structure.

than could just be handled in a SSNIP test.

It was more complex

Entry was the factor that everybody focused on

in the '80s, but there were other factors that were

developing.

that the unilateral effects analysis was specifically

stated.

being looked at by the Division in the '80s.

I think notwithstanding it wasn't until '92

It was something that, in various forms, was

So by the late '80s, by early '92, I think it

was very clear to anybody inside the Division that those

numbers were really only safe harbors and that they were

just the starting point, and at that point, structure,

market share was really only one part of the holistic

effort to tell the story, to show what the linkage was

and decide whether or not that linkage was a concern in

light of all of the relevant factors.

So I think that the history is important.

me, I think that if you look at those numbers that were

generated by the FTC and the DOJ in the early part of

the last decade now, you will see that I think probably

2,000 is the cut off or 2,500, as it now turns out, and

the few outliers where there's still cases that had been

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

brought below 1,800, I think there are kind political

explanations for them, and if anything, those political

explanations led the agencies to bring cases in those

areas because the guidelines were there.

So as far as I'm concerned, they are, as

Christine said, safe harbors.

because I think today they're inaccurate, but I think

that the sort of trend away from just focusing on

structure is again a 30 year or longer occurrence.

ought to continue and the guidelines ought to be very

clear that structure really in and of itself can only be

a starting point and really can only be part of a much

larger effort of looking at a variety of factors.

MR. WEISER:

They ought to be raised

Andy, how do you conceive of the

HHIs and their proper role in the guidelines?

MR. GAVIL:

I think what I would add to what

Rick said is that this 30 year evolution is larger than

just merger law.

years.

Section 1, Section 2, our thinking and our reliance on

structure has changed generally in many ways in

antitrust law.

Antitrust law has changed in those 30

In you look back at cases in other carries,

So I think it's clear that we don't rely on

structure to the same degree that we once did, but

having said that, I worry about two things in throwing

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

the baby with the bath water out.

structure is of some utility?

utility at certain very high levels?

Do we still think

Do we think it's of some

One of the big challenges in antitrust generally

is that there is a trade-off between reducing error

costs, false positives, false negatives, and increasing

the direct costs of deciding and litigating cases, so

does the structural assumption have some utility?

think it still does.

The other thing I worry about is every sentence

the agencies now add to the guidelines will be cited and

held against them in court when they litigate as a

constraint on their discretion, so to the degree you

move away from the structural presumptions and you

started adding, Well, we ought to look at this factor,

look at that factor, when you get into court and

litigate, people will say, Well, you didn't look at that

factor in this case, and I think there's a long history

under the guidelines since 1982 of courts holding the

agencies to their own guidelines and it not always

working out well for the agencies.

I completely agree, however, that moving away

from the 1,800 makes sense.

The assumption there was

sort of a six to five was the threshold where we would

start getting concerned, it looks more like where we are

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

today is five to four or four to three, but I get very

concerned about the agencies saying that, creating the

impression of safe harbor and actually constraining

their own discretion, if a case based on other factors

that happens to be six to five or happens to be five to

four or gives them some concern, how you draft the

guidelines could wind up making it more difficult to

litigate and win that case if you had to.

So it makes sense to me to change it.

the case for changing the increased thresholds, the

1,500, that stuff is clearly -- that's too small to be

of some use, but I would just caution a little bit about

balancing the value of increased guidance against

constraints you can place on the Agency by adding

additional factors that you want to look at, which will

become de facto requirements when you litigate.

MR. WEISER:

I think

So you don't think the old lawyerly

construct of including, but not limited to, or

illustrative, but not necessarily required, is going to

do the job because there's a tension between providing

guidance and giving people transparency into what we do,

and the other side is you worry about pinning yourself

down?

MR. GAVIL:

The guidelines currently say and

even the announcement of this process said, this is just

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

how we make decision, not how we litigate, and then it

proceeded to list all of the cases in which the courts

had used the guidelines as a framework with shifting

burdens of production and proof, and I think that's just

a reality that you have to be aware of.

You can put the conditionality in there, but

thinking back to Baker Hughes and the language of the

guidelines on entry at the time, it didn't stop the

court from saying, Well, the language you've used is not

persuading us.

MR. WEISER:

Bert, you at a conference last year

in Colorado said something to the effect of there is I

guess an indisputable gap between practice and the

guidelines, let's say it's 1,800 and 2,500 as Rick

suggested, and many people have said, as Rick noted, you

can just raise it, and I think you said at the time,

Well, you can start bringing more cases that are in the

1,800, 2,000 range.

How do you approach that issue?

Is that still kind of your view?

MR. FOER:

My view is that we're moving in the

wrong direction.

direction would be to conform practice to the

guidelines.

metric is that we should look for, and it seems to me

that the proper metric is Congressional intent.

We have a gap, but that the proper

You've asked us to talk about what the

It's

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

not economic theory.

And, the Congressional intent, as

defined unfortunately perhaps long ago, by the Supreme

Court is that this is an incipiency statute.

The whole purpose is to avoid high levels of

concentration, and if you step back the way, for

instance, the Antitrust Modernization Commission did not

and you ask, Where are we, where have we come, how

concentrated has industry become?

to fail issue that everybody is worried about today?

What is this too big

I think you have to say that we have not

succeeded in fulfilling the Congressional mandate.

Now,

why else do we hear our marching orders if not from

Congress and the Supreme Court?

is whether the Incipiency Doctrine can be utilized more

than it is in the guidelines.

in one sentence and virtually ignore it.

So the question I ask

The guidelines mention it

It does not seem to me that this brings us back

to Von's, which nobody wants, but I thought that the

original guidelines, looking at basically a six to five,

and basically saying, okay, five companies competing

should be kind of a model, not inevitable, not

irrebuttable, but when you get to five, you should be

worried.

I think that the reality today is much more when

you get to three, you're worried.

Well, by then it's

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

awfully late to be worried.

good at is creating competition.

in theory at least, preserving competition, and we know

that by the time you get to three or four or five,

collusion is much easier to accomplish.

accomplished in a more fragmented industry, but common

sense tells us it's easier to accomplish.

One thing antitrust is not

It's much better at,

It can be

Also, we don't really have a way of getting at

parallel behavior, so since we can't get at parallel

behavior very well, we should try to maintain a

structure in which it's less likely to occur.

Therefore, I would shift the burden, when we get

to high levels of concentration, and I would say instead

of starting with the proposition that underlies our

current policy:

good.

to the overall welfare.

Mainly, that mergers by and large are

They're efficient.

They're likely to be useful

That's the golden proposition, and it works

pretty well up until high levels of concentration, but

then it no longer works, and when we look at the results

of mergers, most of them don't work out very well.

There's some that work out very well, and we've got to

not preclude those, but generally speaking, they're not

terribly successful, because the externalities of a

merger are not calculated into the analysis.

There are

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

real externalities, and I think we're coming out the

wrong way.

level of concentration, there should be a strong

presumption against it, and the burden of demonstrating

that it's in the public interest should be on the

parties that want to go forward.

In other words, when we get to a very high

MR. WEISER:

So there are two ideas here.

I'll

take them both.

One is the virtues and vices of a safe

harbor and what should that be, and the second is the

virtue and vices of a I think it's called a structural

presumption, which is where Bert is going, and, Mark, I

want to ask you to address the second one.

At what point, and Bert suggested six to five or

five to four, others I think would suggest four to three

or even three to two, should a structural presumption

give some weight?

as a predictor of actual competitive effects has become

more questioned, although the guidelines still today

have a commitment to a structural presumption.

As Rick noted, the focus on structure

Is that something that should be retained, and

how should the agencies look at it?

MR. COOPER:

Well, I think the critical point is

if you're going to set a threshold, it's important to

know what the threshold means.

what it's going to mean, where you set it is a shot in

Unless you really know

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

the dark.

I like to analogize and a couple people have

heard this before, for the last 25 years, the merger

guidelines have been sort of like the pirate's code in

the Pirates of the Caribbean.

the movie, so now I get to tell the story.

Not too many people saw

It's a comedic device throughout the movie that

actually really gives you some insight into life in

pirate society, and essentially what happens is at each

key point -- it's an older crowd, you don't have young

kids here.

someone is about to do the morally incorrect thing,

another character says, But wait a minute, what about

the pirate's code, which of course tells you to do the

opposite thing, don't abandon your friend, right, when

you're about to jump ship?

At each key moment in the movie, when

The pirate's code will always tell you to do the

opposite thing, and everyone violates it, except of

course for the heroine, who is not a member of pirate

society, and they play this routine throughout the

movie.

committed, someone says, What about the pirate's code,

and they go off and do the wrong thing anyway.

Every time some dastardly act is about to be

At the end of the movie, the chief villain is

challenged, and they say, Well, what about the pirate's

code, and he says, The pirate's code, they's only

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

guidelines, and he does the wrong thing.

The fascinating thing is in the beginning of the

second movie, they introduce this very early, and

someone is about to challenge him, and he just waves his

hand and says, Don't give me that parlay stuff, he's not

going to hear the pirate's code, and the pirate's code

disappears from the last two movies.

The key here is that if the thresholds are going

to be meaningful, they will be useful, but lax law

enforcement is bad in antitrust, just like every place

else in law enforcement, so if you're going to give me

thresholds, they have to be meaningful, and I would say

the following:

many, I'm a ten guy, but that's okay, times have

changed.

I can live with four is few and six is

I can live with four is few and six is many if,

when you get above 2,500, you pretty well know that

you're going to end up in court.

be meaningful, and you need to know that between six and

four, there's going to be a parlay going on.

the central theme in the pirate's code is whenever

you're about to get off, you would say, wait, parlay,

and in theory the pirates were supposed to negotiate.

If you come in with a six to five or a five to

four, you should know that there's going to be a really

The threshold has to

That was

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

tough conversation about the harm to competition, so for

me, I think the reality has moved there, and I agree

with Bert.

there's a gap between practice and the guidelines, we

assume that the guidelines are wrong and the practice is

right.

right, and that practice is wrong, but four to six is a

number that I think we can begin to live with.

When we have this conversation, we say

Some of us actually think the guidelines were

Let me say, the other question, non structural

issues.

structures still have utility one step further.

the current state of the economic discipline, the

question is:

some utility?

I would take Andy's statement about that

Given

Does neoclassical economics still have

Let's be clear.

The fundamental assumptions

that we've used to analyze the performance of markets

has been shaken, sometimes I like to say buried, if not

dead, beneath the financial rubble of Wall Street, and

so we need to ask ourselves the question that

transaction costs economics and behavioral economics

teaches us things about economic performance that are

directly contradictory to neoclassical assumptions and

predictions, and the question we should ask ourselves:

Does the teaching of these two disciplines make it more

or less likely that market power will be abused?

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

I believe structure still has some importance,

and I frankly believe that behavioral economics teaches

us that once you have market power, given the role of

inertia and social influence and things in human

society, market power is liable to be more durable than

you thought, not less, because the assumptions you make

about human behavior are incorrect.

reality.

They don't reflect

Clearly a good clear statement of four is few,

six is many, with a precise understanding that this

stuff is going to meaningfully dictate future Agency

behavior would, in fact, be a better place to live than

where we've been for the last 25 years.

MR. WEISER:

So we're going to come back to the

following formulation as you put it.

You said in some

context, you need to be able to put on the spot, so we

can tell the story.

safe harbor, and in other contexts, some argue there

should be a structural presumption and maybe likely to

challenge.

In some contexts you need maybe a

Mark has put on the table if you have a six to

five or five to four merger, you have to have a good

story.

expect to be challenged, and the Agency should get a

presumption.

If you have a four to three merger, you should

What's your take on that proposal?

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

Is it

something you can live with?

MR. RULE:

Well, let me start by saying, as I

get older and mellower, I find it hard to resist

propositions from Mark and Bert and others.

probably has something to do with the fact that I tend

to represent more plaintiffs these days.

It also

MR. WEISER:

There is a transcript of this.

MR. RULE:

plaintiff clients.

the structural numbers are relevant frankly to the world

at large, and I think all of us know, we run into

clients who have heard this thing called HHIs, and so it

becomes a big issue for them.

That's okay.

I will give it to my

No, what I would say is, look, to me

It is relevant to them.

It's relevant to people who are planning, who

don't want to necessarily go out and hire an antitrust

lawyer when they're putting together two Kansas wheat

farms to basically say, Look, if you're under this

level, there is not a problem.

people maybe don't like this, that it ought to be a safe

harbor, and that may tell you that you want to set these

numbers a little lower, so maybe not 2,500.

is the right number, but again, you also have a thousand

in the guidelines.

That's why I say, and

Maybe 2,000

I think that's what gets communicated.

The

problem I have with what Mark says, as I say, even

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

though I find some attraction to it is I think the

experience of the last 25 to 30 years has taught people

who do this, both inside the government and outside the

government, that there really are potentially a lot of

relevant considerations so that in some cases, I will

grant you that six might be too few.

I mean, I'm old enough I guess that I don't find

1,800 to be completely appalling, although I think if

you go from six to five, maybe that's a better area or

even five to four.

and under some circumstances that might be a problem.

I think though what the agencies have found is

I mean, I can understand the theory,

there are a lot of other pieces of evidence.

a lot of other facts that can inform one as to whether

or not a merger that goes from six to five, five to

four, four to three, three to two, in fact is a problem.

If we can know the answer better than relying on

something like a market definition, which is not

perfect, and then some heuristics that don't necessarily

have support empirically, that's what we pay the

government to do, to try to get the answer right.

There are

Unless they can tell a story that the merger may

substantially lessen competition in some line of

commerce, in some section of the country, then frankly

the law doesn't allow them to challenge it, but to me, I

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

guess we do know more today than we did 25 years ago.

I think analysis has to take that into account,

and one of the things we do know is that structure

doesn't hold the same significance for determining the

outcome of an effect of a merger as we thought it did 30

years ago, and I think given that reality, the

guidelines ought to be changed to reflect it.

much as on some days Mark's proposal might make sense, I

think it ignores the fact that we may actually be able

to get it right in a particular case more often than

using what is a rather crude rule that Mark plays out.

MR. WEISER:

Again, as

So let me take that line of

discussion, and I have several different ones that I

want to follow but we'll follow this one:

industries have what you might call different minimum

efficient scale, meaning it's hard to sustain, let's

say, five competitors in certain types of industries.

Different

The DOJ filed comments recently in the Broadband

Plan, noting in that broadband markets you're not going

to see textbook competition.

broadband providers for that matter.

You may well not see six

What do you say if there are claims in an

industry where it's moving from, let's say, four to

three, and they're saying it's four and three merging,

and we need to be stronger.

On a pure structural case

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

you might not want to allow that, but as Rick says,

there might be other reasons to believe that merger is

benign and may be pro-competitive.

How does that square with a concern that Mark

articulated about departing from what you're pre

committing to as a particular code?

start with that one, Mark?

MR. COOPER:

Do you want to

Well, yes, because I have the

experience of working lots of industries, and so in the

first year of this administration, I've had

conversations with the antitrust authorities over

airlines, railroads, newspapers, wireless companies,

broadband service providers, all of which are industries

where four would be heaven.

small number of competitors, and the antitrust

authorities lose their primary weapon, which is lots of

competitors, to ensuring an efficient economy.

We have this problem of a

So I have a series of principles, five quick

principles, and I will file them.

Basically, when I'm

looking at a situation, first of all, you really do have

to test the limits of minimum efficient scale.

is going to come in and say:

support more than two or three or four.

challenge that, but if it's true, you really have to

make sure that you get the maximum number of competitors

Everyone

Hey, this market won't

You need to

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

you can, and that's going to be a fight about whether or

not the weaker of the two merged parties is viable.

In the case when you conclude that there is

going to be less than four, then you have to be really

worried about market power because economics teaches us

that the ability of the small number of players to

extract rents and otherwise avoid the inconveniences of

competition is great when there's that small number of

competitors.

So we need to really worry about things like

artificial barriers to entry, refusals to deal, efforts

to monopolize neighboring markets.

a tremendous need to analyze small number of competitors

from the Agencies' point of view, both eventually

prophylactically setting out a policy by which you might

bring other cases under other sections of the Act, but

also as a framework for analyzing what we understand

about the conditions we have to put on these mergers.

So, for me, there's

So if I'm confronted with a four to three and I

conclude that it's a necessary outcome in terms of

minimum efficient scale, then I have to really worry

about the ways that the resulting market power would be

abused.

You will notice I stopped at four to three.

concepts of a dynamic duopoly or a benign monopoly

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

The

simply don't exist in my vocabulary.

I just don't think

this Agency or the antitrust authorities in this country

can, in fact, be comfortable with the theories that led

us get down to those extremely low numbers.

MR. WEISER:

Bert?

MR. FOER:

I don't have five points.

encapsulating point, and that is the principle that as

the level of concentration increases, the size and the

certainty of the offsetting benefits have to increase.

In other words, the higher the level of concentration,

the more skeptical, the more intensive the

investigation, the greater certainty that these

efficiencies are going to be there, and that they will

be passed on, in substantial part, to consumers.

I don't think I disagree with Mark.

I have kind of one

It's a sliding scale.

It's the Heinz Baby Food

test where you had apparently very high level of

efficiency demonstrated, but it wasn't high enough

because the level of concentration was going to be so

high.

very difficult to become more scientific about it

because in part, we've created a pseudoscience.

I think that's the right approach, and it gets

Sorry all my economist friends, but I think that

we've made it into more of a science than it really is

or it can be, and that one of the prices we pay for that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

is a lack of intelligibility to the public in our merger

policies.

three than it is about HHIs.

internally inside the Beltway, but we've also got an

audience outside the Beltway that we've largely ignored,

and I don't think they are very supportive of what we

do.

It's so much easier to talk about a four to

MR. WEISER:

Maybe we can do it

Andy, you can jump in here.

not, I have another question for you.

MR. GAVIL:

No.

MR. WEISER:

The other question goes like this:

We've talked now for the last half hour or more about

concentration broadly speaking, not differentiated

between coordinated effects and unilateral effects, and

part of what happened I think is that the '82 guidelines

and '84 guidelines largely were thinking about and

governing the concern about coordinated effects, and

since '92, most of the Agency's cases have been on the

unilateral effects side, still also invoking the HHI

structural presumption.

So let me start with coordinated effects.

coordinated effects, the question would be as follows,

and this gets to something Rick said earlier:

have a structural case, say a four to three merger or a

three to two merger, is that enough based upon what we

If you

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

know that we should be worried about, coordination which

I believe Bert said or Mark said is something that if

it's tacit coordination is not a problem under Sherman

Section 1?

that's not something Sherman 1 does anything about.

You can have conscious parallelism, and

So is that enough of a reason to worry about it

or might we want to say, as the guidelines do, that

there are certain pre conditions we need to look at to

understand whether or not collusion or coordination or

even, as Christine said, accommodating behavior is

likely, and the Agencies need to have some evidence of

that in addition to the structural conditions?

you think about that coordinated effects question?

MR. GAVIL:

Two things.

How do

First, this is sort of

a broader comment, and I'm glad we made that transition

because in the revisions that are being talked about for

the guidelines, we're trying to separate out what is a

dilemma of the guidelines.

multiple strands of intellectual history in merger and

economic thinking.

strong because of the '68 guidelines, because of the

influence of the structure conduct paradigm.

The guidelines reflect

The structural paradigm was very

We then introduced oligopoly theory, game

theory, and we've sort of layered different strands on,

and I think the tension that's now being addressed

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

between the structural concepts and unilateral effects

is a good example of it.

away from, in all of our discussion to this moment this

morning, is that all that structural stuff may not be as

relevant in unilateral effects cases.

What you really want to get

That's a reflection of there being these

different strands, but the structural paradigm was very

well established in the case law.

established in the literature, so it got written into

the '68 guidelines, carried over in '82 as not to appear

to be a too radical departure.

It was very well

So we have these competing strands of

intellectual history, and I think part of the challenge

in the rewrite is to explain that and separate that out

and explain which models work under what circumstances.

Now, to get more directly to the question.

The

guidelines in essence already answered your question,

made that decision, that structure alone was not enough,

that there's a separate inquiry about anticompetitive

effects.

guidelines, but again, as I said earlier, the agencies

found they were being held to that when they went and

litigated.

It's all in the same section of the

When you went out and said, Well, here's

structure, well, the statistical case is pretty much no

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

longer going to be enough, except at maybe a very high

level.

are the conditions for coordination in this market?

will this merger alter those conditions and facilitate

better coordination?

You have to tell the coordination theory.

That's become part of the analysis.

What

How

It is

clearly added to the burden of the agencies in

challenging a coordinated effects case.

should be?

shouldn't have to do that?

Is that as it

Are there some levels beyond which we

Again, I think if you're going to write that

into the guidelines, two to one, do you need to show it?

Bert mentioned Heinz.

delineating what theory of anticompetitive effects is

there.

is just fine with us, and the sliding scale approaches,

and we're not going to really demand that, but you look

at cases like Arch Coal, and the Court wants to know

where your evidence is of coordinated effects.

Heinz is not very careful about

It's just saying at some point the presumption

The last point I would make is, yes, it is very

important that we use the merger laws to stop structures

from forming which could lead to coordination that we

could not reach under Section 1.

traditional purpose of Section 7, because we recognize

that oligopolistic coordination, which can't be reached

That has always been a

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

under Section 1, is still bad.

higher price, and the only tool we really have in

antitrust, because we've essentially walked away from

the idea that Section 1 can reach interdependent

pricing, is stopping the structure from forming that

will make it easier.

It still results in

So I think on that sense, Section 7 really does

provide a very important -- it goes back to the

incipiency idea to some degree.

barrier that keeps us from getting to structures and

problems that we can't reach under other parts of the

antitrust laws.

It is an important

So I think that role is still important.

does, I think, require us to tell a coordination story.

How will this merger incrementally increase the ability

of firms post merger to coordinate is an important

question to answer, especially if we're talking about

six to five, five to four.

three to two and two to one, maybe the story doesn't

matter as much.

Like I said when you get to

We're just too scared to go there.

MR. WEISER:

Rick, to kind of capture Andy's

point, if you have let's say a four to three or three to

two merger where you have conditions that you would seem

to facilitate coordination, let's say very difficult to

enter, homogenous product and maybe some story you can

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

tell about how coordination happens, should that be

enough for a Court to, under incipiency theory and under

the structural presumption, be able to stop a merger?

Others have argued in comments that the whole

idea of a structural presumption and this concern is not

one the Agency should focus on.

that question?

MR. RULE:

How do you come down on

Again maybe it's a reflection of my

age, but I tend to agree with Andy on this one.

To me,

one of the issues, and I think a number of commentators

have raised this, I think that currently the guidelines

are a little confusing in the use of the term

coordinated effects versus unilateral effects, and I

think you should probably get away from that.

I think I heard Christine say that there should

be a more detailed description of what an adverse price

effect means, and I agree with that.

that, and I think the evidence is consistent with the

fact that in some industries, for example, the

characteristics that you laid out, a reduction in the

number of competitors can raise a threat of a price

increase.

But I also believe

Now, I would say that even in that circumstance,

one ought to be willing to look at efficiencies and that

sort of thing, and so I would say in appropriate

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

circumstances, that could be a basis for concluding that

a merger violates the antitrust laws.

On the other hand, and this is I think part of

experience, but it's also part of the change in the

economy:

modern industries just doesn't work very well, and so

the notion that there's going to be some sort of

coordinated interaction in some industries, for example,

the information industry, is just, to me, not very

credible.

Trying to apply that paradigm to a lot of

I don't think there's a lot of empirical basis

for that, so I think you've got to reach the conclusion

first that this is an industry that is likely to witness

tacit collusion, in the old term, before you reach that

conclusion.

The other point I would make, and I think this

is just an interesting observation, while I agree with

everything Andy said and what I just said, it's also

kind of interesting that the law actually, under Section

1, has moved in the direction of capturing more of what

might be called tacit collusion, leaving Twombly aside

and the difficulty of pursuing those cases, if you look

at Posner's opinion in high fructose corn syrup, there

are ways I think today that I would have been much more

skeptical about 25 years ago of actually creating an

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

inference of a conspiracy using some of the analysis of

Posner.

So that there's an argument today that maybe we

can reach some of that conduct under Section 1, that 25

years ago when all this was developed, there was kind of

the sense that you just couldn't find an agreement under

those circumstances, even though there was tacit

collusion going on, and so approaching it and trying to

stop it structurally was more important.

least that argument.

MR. COOPER:

There's at

I really agree with that, except I

don't want to call it tacit collusion.

it noncooperative games because I think we're talking

about the same thing, and I think that analysis of

noncooperative games is the bridge between coordination

and unilateral action, and he did win a Nobel Prize for

it, and we have spent 25 years, -- and it's almost

exactly 25 years that the theory has received an immense

amount of attention.

I want to call

While I'm not a lawyer and haven't reviewed the

cases very closely, I don't think the influence of

noncooperative game theory has been fully felt, nor has

the influence, as I said, of behavioral economics, and I

think that that needs to get reflected, so I'm agreeing

with that.

I just don't want to call it tacit collusion

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

because that has that old style ring to it of, there's a

collusion here; no, these are just people who, as the

lion in the movie says at the second bar room scene,

Adam Smith was wrong.

It's a wonderful line because --

MR. WEISER:

Which movie are you talking about

MR. COOPER:

This is in A Beautiful Mind.

now?

I'm

sorry, I'm a veracious consumer of popular culture, so

in A Beautiful Mind, Nash is struggling with his theory,

and in the second bar room scene, there are nine guys

and nine gals, and one very pretty gal and other very

intelligent women, and he looks at it.

going to happen here, right?

all compete for the one good looking woman, eight of

them will be disappointed.

He says, what's

He realizes that if they

He then goes back and writes his theory of how

the nine guys will learn very quickly to allocate who

ought to chase whom, and the ability of a small number

of people to capture the monopoly rents available

without colluding is a really important observation to

which the economics discipline has devoted a great deal

of attention for exactly a quarter of a century since

the guidelines were adopted.

MR. WEISER:

I just want to point out for those

who missed it, we have Mark Cooper and Rick Rule in

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

agreement, so the idea of an emerging consensus, we can

stop right here.

We can stop right here and see if we have

questions from the audience.

I think, as Howard said,

we have a small enough group of folks that rather than

asking you to submit written ones, if there are people

who have questions that they want to ask, we have time

for a question or two, and as sort of a professor on

leave, I'm not afraid to call on people either.

Any questions folks want to ask?

too, but if there were any questions?

the back, do you have a question?

MR. ABBOTT:

I have more

Is that Alden in

Yes, thank you very much.

The

question would be directed at Rick.

Posner's opinion in high fructose corn syrup, but given

recent case law, some might argue that it's becoming

very, very hard to win a Section 1 case.

is viewed by many as a minority view.

He pointed out

Posner's view

I would say there are lots of other commentators

who have challenged that, so if that is the case, how

likely are you going to be able to pursue a Section 1

case, and does this get back to the notion that Section

7 is an incipiency statute, and because of the very

difficulties in approving a quote, unquote agreement,

despite Posner's views, Section 1 may be a less than

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

ideal vehicle?

MR. RULE:

I mean, I don't disagree with that,

Alden, and as I said, I agree that under the right

circumstances where you believe that whether you want to

call it tacit coordination or some sort of game theory

tells you that there's going to be a likelihood that

prices will increase, I think that's a basis for

stopping a merger.

The only point I would make on Section 1, the

fact that Posner's decision is out there, I think it

lends credibility to an argument that frankly ten years

ago would not have gotten you very far.

principal issue on Section 1, for what it's worth, in

terms of being difficult to win is Twombly actually.

Twombly is the one that creates the biggest obstacles,

but that's a different panel.

I think the

I do think that that is one theory that could

motivate a merger challenge.

that structure is not by itself determinative of whether

or not an industry is going to exhibit that sort of

conduct.

experience of the last 30 years, and I think that's

really what needs to be captured by the revision of the

guidelines.

Again I think the point is

You have to look at other factors.

MR. WEISER:

That's the

Do we have another question?

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

MR. CARY:

George Cary.

I guess I am finding

this discussion really fascinating, especially some of

Rick Rule's comments.

have is:

explication of the role of non price coordination in

merger analysis?

I guess the question that I would

Is there room in the guidelines for a greater

The guidelines seem to focus on pricing.

They

don't seem to elaborate very much about how non price

competition might be the subject of coordination.

tends to be relegated into the unilateral effects part

of the guidelines, and I wonder whether there isn't room

for some discussion about competitors channeling their

competitive efforts into elements of competition where

consumers could be harmed, where they still compete, for

example, on marketing rather than on price or on some

forms of innovation rather than other forms of

innovation, or is that too big a project?

That

I guess the sub theme here is whether the

unilateral effects analysis has ignored the role of

coordination among firms producing differentiated

products and whether that ought to be spelled out

somewhere?

MR. WEISER:

So, George, that's a great

question.

I was going to add, let me put my related

point on the question, and then I'll let the panelists

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

answer.

More broadly:

Should the structural presumption

not only be tied to and motivated by a story about

price, but other elements of competition, be it quality,

product variety what have you?

MR. COOPER:

Mark?

Well, as the consumer advocate who

is always accused about only caring about price, let me

say we care about a lot more, and we always have.

are other two areas that are really important, one, is

termination fees in cell phones, wireless, for instance

is a really onerous condition on consumers, and there

will be people who will disagree with that, but we look

out at bundling in the cable industry as a term of

service.

There

We have got complaints about --

We look out at the competition of big fat

bundles in the triple play, and these are key questions

about everybody's offering me the same package, and it

only serves a quarter of the market, so, yes, I think

the terms of service is a second area that's really

important in addition to the price, and then the big

enchilada is innovation and long-term competition.

We have tried very hard not to go for the

short-term, near term buck, so frequently people will

come forward and argue that, hey, the prices will be

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

lower next week, and we say, yeah, but what about next

year or ten years from now, so the second area is

innovation and long-term competitive structure.

The guidelines have a footnote here and there,

they need to be much more prominent because they are at

least as important, and, in the case of the second one,

innovation and long-term competition, probably more

important than price and terms of service.

guidelines should be oriented around that, they should

be forward looking.

I think the

It's interesting, Bert talks about the

Congressional mandate, and I ask myself:

What would the

Congressional mandate look like if this Congress were

working on it?

I think the most important thing they would talk about

is long-term innovation and production, and that

wouldn't be a bad thing.

about price and a lot more about building an economy for

the 21st Century, and I think that would be a useful

thing for the guidelines to say.

Obviously they can't agree on much, but

MR. WEISER:

MR. FOER:

They would talk a lot less

Other comments?

Bert?

If I can challenge Joe Farrell for a

speech he once gave:

Price is usually a pretty good

surrogate for the things that we want from competition,

from the market.

We want fair price.

We want

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

innovation.

We want choice for the consumer, but in

some industries and in some circumstances, price is not

a very good surrogate.

For instance, right now there's an investigation

of a voting machine merger where the bottom line is not

so much what the price for a voting machine is going to

be as different aspects of the effects that can come,

including some possibly very important innovation

effects.

In information industries, we may not care as

much about price as we do about choice, so there's got

to be a loosening up that permits these other objectives

to become part of the analysis.

Exactly how you do that, George, I'm not sure,

but I am sure that your question is the right question.

How do we make certain that what we're getting out of

our policy are the outputs that we really want, and

price alone is insufficient.

MR. RULE:

I mean, here's another one where I

will agree, this time with Bert.

I've always kind of

viewed price as an easier, sort of more quantitative

variable to do things like understand how you define

markets.

competitive parameters can be understood or reduced in

some ways to price.

And actually I think a number of different

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

However, I agree that -- and it's contrary.

would frankly be inconsistent with my experience

recently to say that the Agency should ignore other

effects, other than price, because my sense is that if

there is a reason to be concerned about sort of non

price elements of competition, the agencies will look at

it.

Conversely, again my experience has been that

where one has an explanation, that even though there

might appear to be some minimal price effect, if there

is a countervailing non price benefit like a quality

improvement or a technological innovation improvement,

the agencies will consider that.

I think the only thing I would counsel the

Department and the FTC as they go through the process of

doing guidelines, I think it is incredibly difficult to

generalize.

again if you explain the process, then that will help

counsel.

think there's anybody else in this room who was involved

in it -- who came up with the sort of structural

presumption for R&D joint ventures.

and no particular empirical reason for doing it, but we

were trying to come up with something to put in

legislative history in the old NCRA, and that's where we

I think that's one of those areas where,

But, I can tell you as the person -- I don't

We just made it up,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

came up with this notion that so long as it was possible

to create I think we said three other joint ventures,

equal capability, there shouldn't be a problem.

It sounded good.

There was a logic to it, but

I'm not sure that -- and again it had the benefit of

sort of weeding out the things that probably aren't

going to be very interesting, but I do think that when

you start getting into non price areas, it's much harder

to make generalizations, and I personally think it would

be unwise to try.

MR. WEISER:

Andy, you get the last word.

MR. GAVIL:

I think it's hard to make

generalizations, but there are industries where it's

obvious that innovation, quality and service, those

three things, can be very important and are vulnerable

to being lost.

I realize we're out of time.

Healthcare I think

is an industry where you can see lots of examples where

you have pressure from payers to reduce payments.

allow mergers.

is service and quality and innovation as well.

One thing that could get lost in the mix

I think there have been some examples of that,

so you do, I think, have to go beyond price.

Whether or

not -- and I think George's narrow question is whether

concentration has really been linked to losses of non

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

price competition.

to support that, but I think it is important for the

agencies to not just function on price, and I think it's

easy to identify industries where these other components

of competition are especially important and are

vulnerable to being lost.

MR. WEISER:

I don't know that there are studies

I want to thank our panelists for a

great discussion.

(Applause.)

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

PANEL 2:

PRICE DISCRIMINATION/POWER BUYERS.

MODERATOR:

Economics

PANELISTS:

SUSAN CREIGHTON, Partner, Wilson, Sonsini, Goodrich &

Rosati

MARC SCHILDKRAUT, Partner, Howrey, LLP

JOE SIMS, Partner, Jones Day

JOHN THORNE, Senior Vice President and Deputy General

Counsel, Verizon Communications, Inc.

HOWARD SHELANSKI, Deputy Director, Bureau of

MR. SHELANSKI:

Okay.

Well, I would like to

welcome you to our panel on price discrimination and

powerful buyers, and we have a wonderful panel, as we do

really throughout the day.

exception of one member of our panel, has both serious

private antitrust as well as government enforcement

experience, and the one, John Thorne, who does not, has

vast experience being pursued by public enforcement

agencies, so this is really a very fit panel for this

topic.

Everybody here with the,

I would just like to briefly introduce the panel

and then open up with a couple of questions.

immediately to my left is Susan Creighton, who is a

partner at Wilson Sonsini, and a former Bureau of

Seated

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

Competition Director here at the Federal Trade

Commission.

To the left of her is Marc Schildkraut, a

partner at Howrey, who has had a very distinguished

career in both private practice and public enforcement

and is another FTC alum.

Joe Sims is well known to everyone as a leading

antitrust partner at Jones Day, who spent a long part

his career at the Department of Justice, Antitrust

Division as a Deputy Assistant Attorney General, and

then John Thorne, who is senior vice president and

deputy general counsel of Verizon, who has been a

contributor on many panels through his writing, and also

as a litigant in many regulatory and antitrust matters.

I would like to start with a very broad question

for our panel, which is how the Agency should judge a

merger's effects on price discrimination?

and criteria are relevant to judging a merger's effect

on price discrimination?

with you?

MS. CREIGHTON:

And, Susan, why don't we start

Sure.

It seemed to me that

really was sort of two questions.

and the other is:

relevant?

What evidence

One was the criteria,

What kinds of evidence should be

In terms of the criteria, I should say by way of

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

preface that it's interesting how often, particularly in

say high technology markets, price discrimination is

almost always the first thing staff is looking for, so

it comes up all the time, and because it comes up all

the time, I think that the criteria that really should

be used is whether you are able to identify the infra

marginal customers, and can you engage in price

discrimination?

price discrimination, and third is:

Is it profitable?

So those are the criteria.

I think that's the

What's the mechanism for engaging in

easy part.

The question is just how heavy should the

burden of evidence be to go from sort of just presuming

that, gee, you should be able to discriminate between

the buyers and actually having to prove it?

It seems to me that the evidence should have to

be relatively compelling that you actually would be able

both to identify the customers, and that they would have

no means of avoiding having sort of some recourse,

whether arbitrage or something else.

kind of econometric evidence that it would be

profitable.

And also, some

So just to use a high tech example, I would say

that the ability to price discriminate is -- I don't

know if John Baker is here, but I think he had used the

great example in an article a long time ago, something

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

like a unicorn or a white tiger.

I guess I would say it's neither.

It's probably

just a regular tiger, that it can be found in some

places in nature relatively frequently, but the things

that you need to be looking for are, for example, how

does the seller relate to the buyer?

reseller channel, or is it direct contact with the

particular customers?

be:

customer, or is it extensive hands on, deep knowledge of

the individual customer?

Is it through a

Then, further, an example would

Is it just sort of an infrequent dealing with the

In technology, for example, with heavy supply of

services, you have people on the premises all the time,

then all of a sudden it starts to become plausible that

maybe you actually do have some ability to know the

ability of the customer to have some kind of ability to

avoid price discrimination or not.

So that would be the kind of evidence that I

would be looking for.

I think that kind of thing can

bedevil agencies trying to figure out why one customer

likes one thing and not the other, so I think, for

example, the SunGuard case was probably a great example

of that.

It wasn't possible to draw a circle around

saying, well, it's the big customers that can self

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

supply, or it's sort of this type of customer versus

that type of customer, but because of the nature of the

customer relationships, I think if price discrimination

had been better understood and better supported in the

merger guidelines, that might have been an easier case

to say just because we have 60 declarations and they

have 80 declarations, that doesn't mean you just throw

up your hands.

It may mean, in fact, that there are 60

customers that, in fact, the Agency knows don't have

alternatives that self supply.

So bottom line, I guess what I would say is I

think there needs to be not only plausible but

demonstrable evidence that would tell a story about how

it is that you actually would be able to engage in that

kind of price discrimination, so it isn't just a story.

Of course it would be great if there's evidence

that supports that where you can show through

econometric evidence or otherwise that in fact that kind

of price discrimination already has been going on.

can to it, but I thought, for example, that was the

thrust of the econometric evidence in Oracle.

was to show that there had already been that kind of

price discrimination evidence.

MR. SHELANSKI:

Thanks.

I think

Marc?

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

MR. SCHILDKRAUT:

I think you need to go back

one step and ask yourself something more about price

discrimination, because price discrimination is

basically pervasive.

time someone uses a coupon, that's price discrimination.

Any time someone turns around and goes to a movie

theater and has a child or a senior citizen with them,

they'll probably all be at different prices.

Just to give you examples, any

Almost every airline discriminates left and

right no matter whether they have market power or not,

so this is a quite pervasive thing, price

discrimination, and the problem in the guidelines with

using price discrimination is you can make millions of

markets.

arbitrary, particularly to a Court more used to general

criteria, to all of a sudden have a case where you say,

we're going to identify this group of customers that can

be targeted.

It all seems very arbitrary.

It can seem very

That becomes very, very difficult because you

can slice and dice 500 different ways, and being able to

do that suggests to me that you actually need to be more

rigorous when you have a theory of price discrimination

to define a market than when you have a general theory

of a market definition, and that further means to me

that Susan's last remark was very important, which is

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

ongoing evidence of price discrimination is important.

It's important to show that that ongoing

evidence is not the kind of evidence that relates to

simply the every day kind of price discriminations which

I was talking about, which is pervasive, but there are

industries where we're talking about something else, and

those are usually multiple players.

There's systematic price discrimination ongoing,

particularly if you're dealing with a fungible

commodity, and then you need to ask yourself the

question:

possible that an industry like that can really price

discriminate?

What the heck is going on here?

How is it

Typically, when I was back at the FTC and I

would ask questions like that in depositions, I would

say, Well, why are you doing this, I mean, wouldn't you

be better off shaving the price to the people who are

disfavored, and you can make more money?

answer I got was, Well, if I did that, everybody would

do that, and how would I ever be better off?

when I knew that I had something I had to think about

much harder because that was indicating to me that the

propensity to price discriminate was actually

meaningful.

Usually the

That's

If you have something that's that meaningful and

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

you can then turn around and prove something like that

to a court, then you have a theory that's workable and

something that you can do something with and something

that doesn't seem arbitrary.

thing is to avoid this potential for arbitrariness, and

that requires not only having something special, it

requires being able to target.

possible.

I think the most important

It means no arbitrage is

All those different things need to come into

play, and one more thing that needs to come into play:

I think you have to think about your underlying theory

when you're doing this.

you're dealing in a case that is a coordinated

interaction case, it is very possible that, unlike most

cases, a small fringe firm that couldn't really increase

its output is going to be able to undermine that

collusion very, very easily because all it needs to do

is to shift to the disfavored customers.

What I mean by that, is that if

It doesn't have to produce another unit, so I

think all these things need to come into play.

probably answered all of your questions at once in doing

this, but I think all of those things need to come into

play when you're thinking about defining markets in

terms of price discrimination.

MR. SHELANSKI:

We have a couple follow ups, but

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

I want to hold off on them until we hear from Joe on

this.

MR. SIMS:

Well, first I think I feel compelled

to correct the historical record as laid out by Mr.

Cooper.

movie called Ghost Busters, and there was a great scene

in Ghost Busters where Sigourney Weaver, possessed by

the demon, is pursuing Bill Murray throughout her

apartment, and he's trying to resist her, and finally

she tackles him on the bed in the bedroom, and he says,

Wait a minute, wait a minute.

absolute hard and fast result, no fraternizing with the

customers, and then he looks directly at the camera and

says, Well, actually it's more of a guideline.

think that's the origin of the Pirates of the Caribbean

remark.

For those of you who are my age, you remember a

He says, We have an

So I

A theme that will run through my comments today

is practical versus theoretical.

criteria and evidence?

I would start with:

parties doing it today?

If it's not happening pre

merger, then there needs to be a really compelling story

about why it's going to happen post merger.

You asked:

What's the

Are the

Even if it is happening pre merger, there should

be some explanation of why the merger is going to make

it worse or why the merger is going to make the effect

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

of that price discrimination worse.

This seems credible only in some pretty limited

circumstances.

If it doesn't happen pre merger, how is

the merger going to change that fact?

is a potential for this, will the merger cause some

dynamic changes by customers to minimize that risk?

Indeed if there

Buyer statements, declarations, having obtained

declarations on both sides of this question from buyers

in multiple matters, I'm not very enthusiastic about

their probative value.

many customers don't really know what their options are

until they're incentivised to think about them.

It is I think pretty common that

Inertia plays a very strong role in business

behavior, and until they've actually been forced to

examine the possibilities, a lot of people will

automatically revert to the:

There isn't really a

realistic option available.

There's a lot of laziness

in preferring the status quo, so you ought to have some

evidence that these concerns are real as opposed to just

the statement of the concerns.

the same test to statements contained in documents or

otherwise from the merging parties.

Indeed, I would apply

Anybody who is an experienced practitioner in

this field knows how often they get deeply involved in

looking at a merger and come to the conclusion that one

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

or both of the parties don't understand their business

as well as you would expect them to.

The facts, as evaluated by somebody who knows

how to look at it from an antitrust perspective, as

opposed to a business perspective frequently lead you to

different conclusions than the parties reached on their

own without the advice and input of that training.

So sometimes companies don't know they have

market power.

the FTC would accept that as a defense, and the other

way ought to work too.

that they have market power doesn't mean they have

actually.

for hard evidence as opposed to conclusionary assertions

by either side.

I don't think the Antitrust Division or

The fact that they say or think

So my general point here is you need to look

MR. SHELANSKI:

John, as one of these

representatives of one of these high tech companies that

Susan alluded to, have you now or have you ever engaged

in price discrimination?

MR. THORNE:

I withdraw the question.

That's a great question.

If you've

seen any of the recent Verizon television commercials

like during the NFL playoffs, you see the guy come out,

and there's a big white sign, and it says $99 for this

package of all the voice calls you want to make in a

month, and he flips around the 9 to become a 6.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

national television in front of all the NFL viewers, our

pricing went from $99 to $69.

When you have large economies of using national

advertising, it's hard to target Albuquerque for a price

increase that's different than that.

the transaction costs often overwhelm any desire on a

tiny market basis to price discriminate, so in echo of

what Joe said, the practical constraints make it more

difficult to price discriminate than some of the

theoreticians would anticipate.

MR. SHELANSKI:

In the real world,

Let me ask a follow-up to that.

Would a hypothetical telephone company that had merged

seriatim with a number of other hypothetical telephone

companies have found its ability to engage in this kind

of price discrimination be affected by those

transactions, and if so, in which direction?

MR. THORNE:

Telecom is a hard industry to talk

about as an example because some price discrimination

increases output.

don't think if you can fly an airplane today that didn't

have differently priced seats and still fill up all the

seats, so there's some industries where price

discrimination may be output increasing, and in telecomm

in some aspects may be that.

It allows you to build a system.

I think the general trend, if you look at

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

telecom mergers, for example, in wireless or wire line

has been to unify, not to fragment the pricing

structure.

MR. SHELANSKI:

Right.

That actually leads very

nicely to a follow up question I would like to ask the

whole panel.

different kinds of effects that price discrimination can

have for consumers.

It gets exactly to this question of the

Certainly the fact that some people are willing

to pay an enormous amount of money for business class

may enable the airline to offer some very cheap fares in

the back of the plane and fill seats that otherwise

would not have been filled, so I guess the question I

would like to follow-up with, we'll start with Susan

again and just work down the line is:

merger investigation does produce the kind of compelling

evidence that you and Marc and Joe have talked about of

price discrimination or of an increased ability to

engage in price discrimination, how should the agencies

balance harms to vulnerable groups of consumers against

possible benefits to other consumers?

MS. CREIGHTON:

If a

I think that's a great question

because I think that, maybe just to step back on the

question of whether you call it sort of a localized

effect within a larger market or a price discrimination

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

market or a sub market, you really are asking or

potentially getting to the point where you're saying:

For this group of six customers, there's the ability to

price discriminate, but it could very well be that

that's in the context of a merger where the other 94

customers really want to be able to have the integration

that you're now going to be able to supply, these six

don't need it.

So I guess it seems to me that as much as the

agencies have resisted historically the notion of very

narrowly defining efficiencies and sort of offsetting

pro-competitive effects, that you can't say benefits in

this market can't be offset by benefits or sort of

detriments in that market, it has to be merger specific,

sort of all this very narrow defining down of what kind

of benefits will credit to the merger.

I think the concomitant of saying, Yes, we will

look at -- and it may in fact be sufficient for us to

challenge a merger if there are these localized

competitive effects, that it's incumbent on the agencies

simultaneously to step back and broaden their view with

respect to the offsetting competitive benefits.

MR. SCHILDKRAUT:

Yeah, I would agree with that,

and I might want to go a step further than that.

have, in the efficiencies section of the guidelines now,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

something that says we're not going to trade-off

different markets, but if you have price discrimination

markets, you can have millions of markets, and any

individual consumer could be an individual market on

that basis.

If a million consumers are going to benefit and

one is going to be harmed, in theory under the

guidelines, you have an anticompetitive effect, and the

guidelines are telling us we must prevent that from

happening.

agencies actually practice.

I think that that is not the way the

The agencies do make trade-offs under those

circumstances.

They don't announce them that way, but

if the guidelines are going to be honest about this, we

ought to look at these trade-offs and think about

whether you want to bring a case where the

pro-competitive effects to most consumers outweigh the

anticompetitive effects to some.

MR. SHELANSKI:

Would you envision doing this

within the effects analysis, or would you envision the

pro-competitive aspects as something that would come in

under the efficiency analysis?

MR. SCHILDKRAUT:

Well, it could come under the

efficiencies analysis, but it doesn't have to be a

traditional efficiency.

It is standard analysis that it

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

is ambiguous whether price discrimination is going to

lead to adverse welfare effects, so it could very well

be, I suppose, that in a merger, you're going to get the

better ability to price discriminate, but the better

ability to price discriminate could lead, on average, to

lower prices.

That's not what one would normally think of as a

standard efficiency analysis, so my answer to that is it

depends on what kind of effect you would have as to

where you would balance it, but in either case, I don't

think you should let the guidelines where tradeoffs are

verboten unduly effect the analysis where we think there

is going to be positive welfare effects from the

acquisition.

MS. CREIGHTON:

If I could just maybe interject

something Marc says triggers, and this is a bit of a

detour, but one of the issues that you have is the

question of who your audience is for the guidelines, and

I guess I would encourage you to be thinking about

District Courts as your audience, over and above

everybody else, and one of the benefits that the

guidelines have had is a tremendous amount of buy in

from the court system.

To preserve that, I think if you're going to

have buy in on the notion of sort of more localized

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

competitive effects, you are going to have to sell this

as reasonable.

to sort of the common sense kinds of reactions you're

going to get from judges along the lines of what Marc

was describing.

You're going to have to sort of respond

If you're trying to say:

Under our guidelines

we don't have to look at the fact that overall our

prices are going to go down, you're just not going to

get the kind of buy in that the '92 guidelines had.

So just at a very practical level, I think it's

important to be taking cognizance of that, that this has

to be a realistic and accord at some level with the

intuitions of general stretches.

Sorry.

MR. SHELANSKI:

MR. SIMS:

Joe?

Let me just first follow-up on that

point.

seems to me are two:

world who are trying to figure out how to look at

mergers based on the way the government will look at

mergers, number 1, counselors, internal and external

counselors, and number 2, the courts.

I mean, your basic audiences for guidelines it

All those folks out there in the

I agree with Susan, the courts are a lot more

important than the counselors.

figure it out over time.

The counselors can

The courts will hold you to

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

what you say and what you layout as the standards.

only will they but they should.

boundaries are there?

boundaries, so there is a tension.

Not

Otherwise, what

The statute provides very little

There's a tension between trying to write

guidelines that are descriptive, in fact, of what the

agencies do and descriptive enough so that people can

actually figure out what the agencies do in some detail

on the one hand and writing guidelines that will

actually be followed and useful to courts when you're

dealing with that.

Now, most mergers don't go to courts, and so a

reasonable person could say:

over counselors?

It's not the guidelines, and it's not the agencies.

It's the courts, and those are, to me, the most

important audience.

Well, why emphasize courts

The courts set the rules in the end.

The other sort of side point to this, and I'll

come back to your basic question, is:

Economics is

critically important in intelligent analysis of mergers,

and for that matter almost anything else in antitrust,

but courts deal in English.

and so you can't really in my view assert economics, an

economic analysis only or primarily as the basis for

challenging a transaction.

They don't deal in math,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

Economic analysis has to be supportive of an

intelligible and pervasive competitive effects story.

You have to tell a story.

sitting on the bench.

are like me, who have to have economics interpreted to

them by intelligent economists, and what happens in

trials is we have our intelligent economists, and you

have your intelligent economists, and the court sits up

there and says, I don't have a clue which one of these

is right and they wash, and you end up with a decision

based on something else, so I apologize for the

divergence.

There are damn few Posners

More of the people on the bench

On the question of how do you balance, I think

this is actually the single most important question

that's connected to price discrimination, and I think I

agree with both Susan and Marc, if I understood them.

You really can't, as a practical matter, expect

to be successful in challenging transactions which have

apparent anticompetitive effects only on very small

audiences and positive or neutral effects on much larger

audiences.

A court is going to look at this not as an

exercise in trying to find the group of consumers who

might be injured, but they're likely to look at it as an

exercise in figuring out whether this transaction is in

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

the public interest from an antitrust perspective?

net good or a net bad?

94 does not come out to net bad.

Is a

To use Susan's example, 96 plus

Now, the commentary recognizes that the agencies

will deal with this thing.

That Guy Bakery case that's

noted in the commentaries talks about maybe there was

some anticompetitive effects on some institutional

customers, but there were procompetitive effects on

everybody else, and the institutional customers are only

20 percent of the customers, and the efficiencies were

uniform across the board, and so we didn't challenge the

transaction, even though there arguably was a basis for

challenging it.

That is the kind of analysis that I think the

agencies have to do, and more importantly, it's the kind

of analysis that they have to be prepared to defeat, if

they don't do that analysis and try to go to court to

protect that 20 percent or in many cases, that 1 or 2

percent of the potential audience.

MR. SHELANSKI:

MR. THORNE:

I thought you were about to switch

to a new topic.

MR. SHELANSKI:

John, do you want to

comment?

Thanks.

topic.

I'm about to switch to a new

We may look back and follow-up on some of this,

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

but I would like to switch to our distinct but related

topic of large buyers, and I would like to start with

you on this, and I have a couple of different questions.

Why don't we start with this one, which is

basically:

How powerful buyers , and I actually want to

use the term powerful buyers instead of large buyers.

MR. THORNE:

term, but go ahead.

I can tell you that's the wrong

MR. SHELANSKI:

That can be part of your answer.

I would like to hear your thoughts on that, but how

should powerful buyers factor into the analysis of

competitive effects, and specifically how should

agencies determine whether powerful buyers will protect

all buyers or just themselves?

dictate the market price or just their own price?

MR. THORNE:

To what extent do they

That's a good question for me

because most of my experience, most of Verizon's

experience with the agencies is as a buyer.

frequently called by Agency staff about other people's

mergers, and we do a little bit of merging ourselves,

but most of our interactions is in the context of:

You've been named as one of the 20 largest buyers of so

and so's product, they're merging, can you put somebody

on the phone that can explain the jargon of the industry

because nobody knows what these products are or even

We're very

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

know who these various participants are.

Then usually the interview goes a bit deeper.

They ask for the views of a customer, and I think the

views sometimes matter.

and some papers by Joe Farrell on how often customer

viewpoint is important, but the important question comes

after that, and that's:

through?

What are you going to do to protect yourself?

do anything?

There's a paper by Ken Heyer

Well, what if the merger goes

What if we, the Agency, allow it to happen?

Can you

Who would you turn to?

There we usually -- I don't think Verizon is

unique in this, but usually we have something to say.

Occasionally I get a guy on the phone that's being

interviewed, and I don't know, I'm expecting you to

block it, but often as a buyer, there's a strategy for

dealing with a merger or with anything that might

threaten the price increase or a change in terms of

dealing.

I'm tempted to tell a story.

Maybe I'll tell it

quickly and then get away with it.

Just on the

visibility into this thing over the whole period, when

the Bell System was broken up, you had the Baby Bell

Companies freed from buying Western Electric Gear.

had always for -- not a whole hundred years, but for

most of the century been buying their house product, and

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

They

all of a sudden, like being subject to a merger to

monopoly, you're freed to try to do something else.

You've been buying Western Electric but now

you're allowed to do something else.

immediately went to Canada and bought Nortel, the

supplier there to the American market, so there were

two.

The Bell Companies

Bell Atlantic, the precursor to Verizon, wasn't

happy with just two suppliers, and went to Germany.

Siemens, which was on a totally different standard than

the North American telecom standards, agreed with some

nudging and promises, to bring its gear to the North

American market.

gear, get it tested, guaranteed enough purchasing to

make it worth it as well.

Bell Atlantic helped to qualify the

Now, we had three suppliers.

Lucent, the

renaming of Western Electric, felt threatened and

retreated to a strategy of:

locked in supply, let's milk it for all its worth, let's

make it hard to or expensive to get increased capacity

and new features on the locked in devices.

Atlantic again, with the help of Bellcore, the standards

groups of the Bell Companies, set standards to break the

points at which we were locked so we could add capacity

in other people's gear that connected through a standard

Well, I guess we have some

Bell

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

interface, add features to get another box that was

connected through another standard interface.

Lucent, again I'm just going to tell you what

the allegations were because that case -- although the

trial started but we didn't finish, Lucent tried to

thwart the standards.

East Texas.

I brought an antitrust case in

We settled.

Moral of the story was that the strategy of

opening up what was at one point a total hundred percent

self supply to multiple competitors and evading even the

lock in on the residue, the strategy succeeded and Bell

Atlantic, Verizon was not the only beneficiary, but tiny

Seelex, the competitive local telephone companies under

the '96 Act, entered the market getting all sorts of

cheap product as a spillover from probably a group of

sophisticated buyers doing the work of opening up that

market, attracting supply, supporting and qualifying new

entrants; then with standard setting and redefining the

product, making it possible to have mix and match

capability for the things you had to add to the locked

in piece of it.

That's the long story I meant to make short, but

let me just outline how I think buyers are important.

First, as a matter of fact, buyers often can

self protect, and if you think in symmetry terms, this

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

is a problem with being a math major, the guidelines

spend so much time on what other suppliers do.

enter to discipline a merger?

The other half of the transaction, what can they do to

self protect?

Can they

Well, what can buyers do?

We submitted some comments that cite some of the

relevant articles on the subject, but there's one thing

we missed and I want to point out.

the year after the '92 guidelines, in the '93 Antitrust

Law Journal Winter Edition, has a wonderful little

article canvasing the ways that buyers can defeat

oligopoly pricing.

Mary Lou Steptoe,

So the first point is the buyer's side of the

market is important.

my own experience, the agencies seem to be looking at

whether buyers can self protect.

the interview phone call and say, yeah, we think we can

take care of this, that helps an Agency decide not to

challenge.

The second point, and this is just

If the buyers answer

The third thing, recently courts have begun

taking seriously the buyer's self protection.

For

example, Verizon supplied a witness for the DOJ case

against Oracle PeopleSoft.

these witnesses seem like they can take care of

themselves and disregarded their concern about the

The good Judge wrote that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

merger.

So I think the merger guidelines, as presently

written, insufficiently reflect what the courts are

doing, what the Agency staff is doing and the importance

of this.

sort of come back to price discrimination.

Now, that's not the end of the story, and we

The fact that some buyers can protect themselves

doesn't mean that all buyers can protect themselves or

the spillover is perfect.

protect themselves.

The same skepticism that applies on

the supplier side:

Will entry be timely, likely,

sufficient, you can apply a similar skepticism to a

story about buyer self protection, but it's still an

important element of how the markets or some markets

tend to work, and I think it deserves some attention in

the guidelines, the way the supplier side entry stories

get attention.

MR. SHELANSKI:

Not all buyers can always

Okay.

I've got some follow ups,

but I think before we go to those, I would like to hear

from the rest of the our panelists.

some thoughts on this?

MR. SIMS:

Yes.

Joe, do you have

I guess the only thing I would

add to what John said, all of which I agree with, is

that the key question to me is not so much can some

buyers protect themselves, but can most buyers protect

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

themselves?

There are lots of different ways to protect

yourself.

consumer products, and when we're talking about the

reaction of distributors like supermarkets or the

Wal-Marts of the world, and they sell those distributors

30 or 40 different products, misbehaving with respect to

one product creates serious dangers with respect to

other products.

I've done a lot of work for people who make

I know a lot of economists find that nonsense,

but as a practical matter, it is real and business

people believe it, so that constrains their behavior

because these people are important buyers.

the term large buyers because large really isn't the

issue.

the competitive environment in which they operate?

I don't like

It's how much bargaining power do they have in

Same thing can happen on geographical

differences.

You're selling to people in multiple

geographies.

You have the ability to exercise market

power in one but not in the others.

in one, does that cause them to change their behavior in

another, or the fact that the buyer has some strategic

importance to the seller?

some ways.

If you exercise it

It's a validating buyer in

So there's lots of different ways that this can

happen, but the critical question, as John says, is:

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

How far does that reach?

We get back to price

discrimination in the end.

differently with respect to some meaningful group of

consumers that don't have the same leverage, bargaining

leverage that this one group has?

MR. SCHILDKRAUT:

Can you, in fact, behave

Often this issue is self

correcting.

is very difficult, and in fact you have a group of firms

that can attempt to coordinate, to target some

unfortunate, small buyers while they can't target the

large buyers.

those circumstances is some of our colluding firms are

going to end up much better off than others, because

unless they can perfectly allocate the customers,

everybody is sharing equally, the seller who ends up

selling more to the big buyers and less to the small

guys who are at higher prices is going to turn around

and say everybody else is doing better than I am, and

that seller is going to start cheating.

What I mean by that is, let's say arbitrage

Typically what's going to happen under

And because he's just not doing as well, so

coordination becomes very hard when you're dealing with

big buyers versus small buyers, and so you have to watch

out and make sure you're actually dealing with a stable

situation, even if it looks like small firms can't

protect themselves as well, because they may not be able

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

to, but the market may end up protecting them.

Go ahead.

MS. CREIGHTON:

MR. SCHILDKRAUT:

No.

I thought you were done.

On the other hand, one other

point I want to stay in the opposite direction, I'm not

saying it's impossible that you could have a stable

situation like that and you have to, under those

circumstances if you're the Agency, watch out for what

large buyers are saying to you because if the large

buyers think they can do better off than the small

buyers can, they may not want to say that because there

may be benefits to them because the price will stay high

downstream, and they're getting the benefit and they may

think the merger is good simply because it's

anticompetitive and they're going to be able to share in

the anticompetitive effects.

Go ahead.

MS. CREIGHTON:

I was going to mention on a more

pedestrian level, I think it may be that power buyers

are the issue most often raised by the parties about

which the guidelines are completely silent, and I would

certainly say that after entry, my guess is power

buyers, you would have to poll the staff, is the defense

invoked most often as a defense.

So although I had previously said I think the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

principal audience for the guidelines is the courts, I

said that because I think there are other ways of

educating practitioners, as Joe was suggesting.

commentary was an effort at that.

speeches.

educate practitioners, but it is my impression that

powerful buyers get invoked way too often by the parties

and rejected out of hand way too often by the staff,

whereas entry I think staff really agrees is an issue.

I think their skepticism about power buyers is

reflected in the commentary, and that that was an undue

skepticism.

infrequent solution if you otherwise have a competitive

problem, but it happens with some frequency, and so sort

of things like I think you were just mentioning, Howard.

I would be curious why John thought power buyers was the

wrong term, but certainly I think many practitioners

tend to equate power buyers with large buyers.

The

You can give

There's a lot of other things you can do to

The power buyers are going to be relatively

You always see it when it's like state and local

government, oh, well they're large, but they may be the

perfect example of a customer who can't in fact on a

particular kind of purchase defend themselves.

it would behoove the agencies to follow the example in

this respect of the European Commission, and

specifically address power buyers.

I think

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

Going back to my final argument about the courts

being your ultimate audience here.

this often before the agencies, eventually it's going to

start showing up in litigation as well, and you will

want to have staked out your ground for when it is that

the power buyers are sufficient and hence, why in a

particular case that condition hasn't been met.

MR. SHELANSKI:

If it's coming up

Before I get with John on what

the right term is, let me just ask a question that

follows very quickly from that.

John and Sue and Marc is a suggestion that there are

certainly circumstances where buyers are, we call them

powerful buyers, large buyers, who obviously can't

protect themselves.

that too large of a presumption.

all of merger analysis, and then you brought in these

buyers with the suggestion that the existence of a

powerful buyers could be a defense.

What I was hearing from

We obviously don't want to make

Otherwise you swallow

Is your thought that powerful buyers should be

elevated in the guidelines to the level of efficiency as

sort of a defense, or more along the lines that I

thought I was hearing from the others?

take into account in the competitive effects analysis?

MS. CREIGHTON:

It's a factor to

Oh, I guess I'm not that sure

that I have a view on that.

I'm more getting at the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

substantive point, that if you have buyers who can't

have the same set of purchasing needs as other buyers in

the market, so you can't say, Well, sort of segregate

the products, they have an ability to find alternative

sources of supply.

in the market that they would render a price increase

unprofitable.

guess is really what I was going to answer.

How you want to put those factors in I

MR. SIMS:

They're substantially large enough

Can I make one quick point?

recognize before I make it that this is a lot easier to

suggest than to execute, but one of the problems in the

current guidelines is that they're too mechanical on

their face.

analysis is, with all due respect to the economic input,

art, not science.

involved because the facts are never really crystal

clear.

Merger analysis is not mechanical.

Merger

There is a certain amount of judgment

There are always ambiguities.

The guidelines,

to the extent possible, should be written to recognize

that this is a dynamic process, not a mechanical

process, so asking the question:

efficiency or should it be an anticompetitive effect?

don't like the question.

MR. SIMS:

Should it be an

I do like the question, and I really

think it should remain in competitive effects.

That's

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

what's really going on is we're looking at an effect of

the merger.

you're shifting burdens of proof I think at the wrong

time.

You make it a defense.

All of a sudden

I don't think that's the way to go about that.

I do think they need to be mentioned more in the

guidelines.

buyer is a large buyer can just go and negotiate with

someone, and basically say:

lower price, I'm going to go to somebody else who I

think will, and you really don't know in advance whether

somebody's going to stick to some coordinated agreement

or not under those circumstances, and you're nervous,

and you end up giving someone a lower price, and that

can then ripple throughout the industry.

Just the first order effect of a large

MR. THORNE:

If you don't give me a

Let me answer the question that I

think you asked:

Assuming you like this idea how do we

draft it?

places where the American guidelines are out of step

with the European Commission and actually a little more

severe on mergers.

As Susan mentioned, this is one of the rare

The European Commission has a fairly nice

formulation of how to consider offsetting buyer power or

buyer self protection, but if you want to Americanize

it, the Mary Lou Steptoe article has five scenarios for

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

buyers self-protecting that are pretty good, and one of

those scenarios isn't about large buyers.

well informed customers, and so I've got one prop I want

to share, and I didn't bring -- this is the January

issue of Consumer Report.

it shows that in every single American market Verizon

has the best cell phone service, that's not why I

brought it.

It's about

I didn't bring this because

If you're a Consumer Report subscriber and you

get their online version, you can find out what the car

dealer paid the manufacturer including all the discounts

and rebates -- paid for the car, and so as a well

informed Consumer Report's subscriber, you can go to a

car dealer and you start from his cost, and I know what

you paid for the car, you want a margin of $250.

my offer.

That's

Just by being a stubborn and smart buyer, you

can negotiate, well, thanks to being well informed, so

my term wouldn't be large buyer.

sophisticated buyer, well informed buyers or buyers who

enjoy the spillover effect of those who are

sophisticated or stubborn.

MR. SHELANSKI:

It would include

Your mission, John, if you

choose to accept it, is to figure out how do we

encapsulate that into one adjective that we can stick in

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

front of buyer in the guidelines and let me know when

you come up with that.

MR. THORNE:

Leave the adjective and talk about

it as buyer's self protection, what can buyers do to

protect themselves it.

MR. SHELANSKI:

That's very helpful.

With that

last word, I would like to go to the floor with any

questions or our panel?

MR. WEISER:

Phil Weiser?

So the discussion on self protected

buyers has been very illuminating.

One thing I will

want to peel back a little bit is putting the two

concepts of the panel together.

How often do you think that the dynamic John

described as for his Seelex cases, Seelex benefitted

from the sophistication of Verizon will be the rule as

opposed to where you have certain buyers, say Walmart,

who are able to be very savvy and protective but others

won't necessarily have those capabilities, and thus you

have the two concepts together?

Large buyers wouldn't benefit from that

effect but for the fact there can be price

discrimination as to the large buyers?

how to evaluate that, and is that just all part of that

effects analysis or is it just not as big of a concern?

I don't know if anyone has thoughts on that.

Any thoughts on

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

MR. SCHILDKRAUT:

I think again it's part of the

effects analysis, and it's all the different things we

have been talking about here.

Is coordinated effects going to be undermined under

these circumstances?

systematic is it?

Is arbitrage possible?

How persuasive is it?

How

All these different kinds of things are going to

give you hints as to whether a merger is going to have

an anti-competitive effect against a small insular group

of buyers who just doesn't have the wherewithal of

Walmart.

MR. SIMS:

Ever since Derrick Bok's article 30

years ago, there's been this angst about the demise of

presumptions and the rise of unique fact- situations.

am going to borrow a Marion Barry quote here, which I

don't use too often but this one seems appropriate,

which is:

Get over it.

I mean, the world has left presumptions.

It's

not going back to presumptions.

Every case depends upon its facts.

impossible I think to write guidelines that are going to

be credible and accepted in the courts that tend to try

to create or rest on presumptions as opposed to

explanations of the analysis that you use.

MR. NAIL:

Hi.

Every case is unique.

It's going to be

John Nail from FERC although

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

this is about healthcare.

United States healthcare and a couple hospitals in New

York City, and in those markets you have localized

healthcare markets, localized healthcare networks being

assembled by private insurers and public insurers and

various other entities.

I was just reading about

So how would these revised guidelines deal with

the situation where two or three local insurers were

merging in order to sort of where -- maybe these key

hospitals where there aren't good substitutes?

happening in this case are the hospitals are denying the

increases in United Healthcare and threatening them not

to be in the network.

What's

So how would you deal with those situations

where a merger between the private insurers may be seen

as beneficial in terms of thwarting the power of certain

key healthcare providers that aren't well substitutable?

You can't necessarily go out of New York City to Phoenix

to get your open heart surgery if there's certain kinds

of very key elements and how would the offices on the

ground deal with those kind of product market

definitions?

That's broad.

MR. SHELANSKI:

If I can just generalize the

question.

To what extent should the existence of

powerful buyers or monopsony power on one side effect

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

our view of the benefits of a merger?

MR. SCHILDKRAUT:

Well, I think you first have

to figure out whether we're really dealing with

monopsony power or whether they're taking away the rents

from the hospital, and the second thing we have to deal

with is the fact that United in this case is pushing

down the price, some of which may be passed along to the

consumer.

After you take all that into account, I don't --

gee, I don't think the agencies are going to turn around

and going to want to okay a deal which is going to

prevent that from happening, but I think that is too

particularized to end up in the guidelines.

hope it is.

MR. SIMS:

At least I

I don't have anything to offer on

that point, but I do have one hard and fast rule that I

would recommend to the agencies in dealing with price

discrimination, market definition based on price

discrimination, and that is you can't have a market

definition that requires more than six words.

MR. SCHILDKRAUT:

syllables.

Now it's got up to six words?

MR. SHELANSKI:

I thought it was six

We'll work on that.

We have

time for a final question from the floor for our panel?

Okay.

Then I would like to --

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

MS. CREIGHTON:

Just on that though, we had

given an example of SunGard and Oracle earlier, one of

the things that had precipitated the move to do the

commentary at the time that I was still at the

Commission was this tension between sort of having

direct evidence of competitive effects and market

definition.

definition first, instead of using competitive effects,

as I think George did well to prove what the market

definition was in Office Depot Staples, that you end up

with the competitive effects, actually even undermining

your market definition argument.

When you end up with doing market

So you end up in this weird world where you're

litigating a case completely different from the one you

investigate or you're trying it in a way that's

completely different from the one you investigated.

So I actually, maybe to beg to differ a little

bit with Joe, price discrimination is the hardest case

of that where you do end up with these multi word,

seemingly market definitions or localized competitive

effects within a larger market.

I would commend to everybody Mark Schildkraut's

article from 2005 on the Oracle case and how you might

have tried that case differently if you had started with

competitive effects and used that to prove market

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

definition, but I do think it's important for judges to

understand that you're not just making it up out of

whole cloth because otherwise the multi adjectives

really consume you.

MR. SHELANSKI:

No.

I mean that's an extremely

helpful point, and it relates to some of what we were

talking about in the earlier panel about the extent to

which you free yourself from the sort of wooden

algorithmic formula that is contained in the guidelines

now and allows for this more flexible analysis, but

taking Joe's advice seriously, that you have to really

make it credible.

You can't sound jerry rig it.

MR. SCHILDKRAUT:

I just want to make a final

point on that wooden analyses.

you're going to have guidelines, there has to be some

wood in them.

have to tell people what you want to do and to have

guidelines and commentaries, and then on top of that,

well, that's not really what we do.

I actually think if

Everything can't be vague.

MR. SHELANSKI:

MR. SCHILDKRAUT:

You really

Right.

Then having to take that to

court where everything is very vague, and everything is

very vague, that sort of gives the prosecutor the

ability to say:

intended to mean, so I'm not as opposed to wood as I

Things mean things they were never

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

think Joe may be.

MR. SIMS:

Just to put a fine point on that, I

would strongly advise against guidelines that have rules

in them.

analytical process and the relevant factors that the

agencies take into account.

in the guidelines what the result of that analysis is

going to be because that result is going to be unique to

the facts of a particular case.

The guidelines ought to describe the

You shouldn't try to layout

As soon as you start laying out rules, you will

set up a situation where some case that those unique

facts drove you to a perfectly intelligent decision to

challenge it is not consistent with that rule, and you

will have to eat this in court, and those are the cases

that you lose.

MR. SCHILDKRAUT:

I think maybe sometimes you

should lose some of those cases, that if we really need

rules that give us guidance, that give courts guidance,

and we have a universal vocabulary.

all those things.

We can talk about

MS. CREIGHTON:

Can I try a reconciliation?

MR. SHELANSKI:

Yes, terrific.

MS. CREIGHTON:

15 seconds because I think

really going back to the point about that the District

Courts that are the standard.

I think it can't seem to

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

the Judge like you're just making stuff up, so as long

as there's enough constraint in there, that it doesn't

look like you're just pulling things out of the thin

air, I think you could find a medium ground between

them.

MR. SHELANSKI:

I think with those two policies

as parameters, I think there is some middle ground when

you talk about specifying your analytic framework, but

maybe not locking in presumptions or rules too tightly.

There is middle ground there.

With that, I would like to thank there excellent

panel for their remarks this morning, and we will have a

shortened break.

to you out there, that shouldn't be too painful but

let's come back in about five minutes and start with our

entry panel.

Given the wonderful luxuries provided

(Whereupon, a brief recess was taken.)

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

PANEL 3:

ENTRY.

Moderator:

economics

PANELISTS:

GEORGE S. CARY, Partner, Cleary, Gottlieb, Steen &

Hamilton, LLP

MARGARET GUERIN-CALVERT, Vice Chairman and Senior

Managing Director, Compass Lexicon

JOHN E. KWOKA, JR., Neal F. Finnegan Distinguished

Professor of Economics, Northeastern University

JOSHUA D. WRIGHT, Associate Professor of Law, George

Mason University School of Law

HOWARD SHELANSKI, Deputy Director, Bureau of

MR. SHELANSKI:

Welcome back to our final panel

for the morning, and we'll then have a lunch break.

This is the panel on market entry, and again we are

fortunate to have a really distinguished panel.

Immediately to my left is Josh Wright, another FTC

veteran.

professor at George Mason University and has been a

valuable contributor, and he is also the prime mover

behind the truth on the market blog, and a prime mover

in some antitrust topics.

Josh is a lawyer and an economist and is a

Thank you.

Immediately to the left of Josh is John Kwoka.

John is a veteran of both the FTC and the DOJ, if I have

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

100

that right, and he's currently the Neal Finnegan

Distinguished Professor of Economics at Northeastern

University.

professor at George Washington University here, and also

over the years, along with Larry White, has produced our

periodic volumes of the Antitrust Revolution.

very glad John could come down from Boston.

For a long time before that, he was a

We're

To John's left is Meg Guerin-Calvert, a well

known economist to most of us, and Meg is currently

president of Compass Lexicon, one of the leading

economic consultancies in the world.

Finally to her left, Josh Wright's uncle, but we

won't presume any collusion in your remarks, is George

Cary, long time and distinguished antitrust partner at

Cleary Gottlieb, and also an alum, we're glad to say, of

the Federal Trade Commission.

So thank you all for coming here and giving us

your thoughts, and I would like to jump in I think with

a difficult and broad question, and I'm going to direct

this first to Meg because I know this is something she's

thought about.

How should the Horizontal Merger Guidelines take

into account the several ways that entry might factor

into merger review?

role of uncommitted entrants as market participants in

The current guidelines address the

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

101

Section 1, but then defer the supply side consideration

of committed entrants until Section 3.

distinction make sense and is there a better way to

address entry comprehensively in merger review and in

the guidelines?

MS. GUERIN-CALVERT:

Does this

Thanks, Howard.

I think

just as an overall comment, I think one of the things

that is of interest on this panel is the idea that the

overall approach to the entry question is something that

does belong in the guidelines, and that should be fully

incorporated and unified as best as possible to take

into consideration the competitive effects of the merger

going forward and taken into account the analytical

principles that are embodied in the current guidelines.

I think your question raises something more than

organizational issues as to whether or not certain

concepts should stay at Section 1 and others at Section

3, but really whether we want to affirm or reaffirm the

concepts of uncommitted and committed entry and what it

is that we could best do to have a unified theory.

have a couple of thoughts.

First, I think that the concepts underlying --

whether one agrees with the words or not, the concepts

do remain very relevant and deserve continuance in the

guidelines, probably albeit in a little built different

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

102

organization form.

I think what the uncommitted entry

concept provides is a means to consider the evaluation

of entry, in particular where there are low or no sunk

costs through either repositioning or expansion or other

modes of entry.

I think it has a great deal of empirical

relevance for a broad range of industries in which you

have competitive constraints, both pre and post merger,

that may arise largely from either nearby firms or

nearby products.

I also think that the concept of committed

entry, which really is the embodiment of the timely,

likely, sufficient test, in its emphasis on evaluation

of the importance of scale and sunk costs relative to

the market are the cornerstone of modern economic theory

on entry as is uncommitted and deserves a space.

I want to just mention one particular thing that

I think as currently configured particularly makes the

evaluation of uncommitted entry very difficult.

where it is right now in the market definition section

and the market participation section, it essentially

says if you have a set of circumstances where you're

able to identify firms or products that could move in,

they are, as they said, hypothesized to be in the

market, and then implicit in that is some notion that

I think

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

103

the consequence of that entry and expansion should be

incorporated in HHIs and shares or some measure of

elasticities.

I think as we all know, that is extremely

difficult to do, and I think empirically, it has tended

perhaps to be given less weight or done less frequently.

I think in particular, historical loss data, diversion

analyses, share analyses have a very difficult time

capturing that fully.

captured fully, and it is embodied there.

cases it's more difficult.

I think in some cases it has been

In other

I think a lot of the models that we have in many

industries make it very difficult to take into

consideration the full effect of repositioning and

uncommitted entry.

So given that, what might we do?

What I would

tee up for discussion is it would be useful to have a

unified theory of entry that is fully part of the

competitive effects analysis that takes into

consideration uncommitted and committed entry in some

form of a synthesized entry section, and that

particularly looks at what I think embodied in the

guidelines is a two part test now:

have relatively low sunk costs, there's less importance

perhaps in looking at the economies of scale part, and

Where it is that you

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

104

easier perhaps to identify that the constraints exist

post merger.

One can look at repositioning and expansion and

I think evaluate that, and that way it will, I think,

appropriately give a lot of weight in the entry analysis

in the application to looking at those cases where you

have either high sunk costs or very large economies of

scale that are required relative to the market.

So that's where I think a unified approach that

would give perhaps more weight and more analytic support

for how to embody the uncommitted entry concept would

improve the guidance provided by the guidelines.

MR. SHELANSKI:

remark.

Thank you for those very helpful

George, would you like to follow-up on that?

MR. CARY:

Yeah.

I generally agree with the

concept of a unified exploration of entry in the

competitive effects analysis.

starting points.

point of view, defining a product market from the demand

side as the guidelines currently do is the right choice.

Not including the supply side elements and market

definition I think makes sense.

I guess I would make two

First I, think that from a demand side

To move the supply side consideration into

product market definition I think will create a little

bit of confusion and also could potentially lead to some

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

105

ad hoc decision making at the agencies which may not be

particularly helpful.

current division between demand and supply is a good

general framework.

So, I think maintaining the

Secondly, I think that the current merger

guidelines explication of timely, likely and sufficient

on entry is a good format.

think it's worked well, and at least I'm not familiar

with any economic literature that would undermine that,

so I think maintaining that makes an awful lot of sense.

It's a good structure.

The one place where I would differ slightly I

think with Meg is that ultimately the question of

product market definition feeds into the question of

concentration:

looking at whether you consider uncommitted entry as

part of the market participation, the question to me

becomes whether there's a metric that allows you to feed

that in rationally and sensibly into the market

concentration numbers.

What are the market shares?

So in

So, for example, if the issue where defining

market shares is current sales in the market, if that's

the relevant metric, then it doesn't seem to me to make

a lot of sense to try to cobble on top of that

uncommitted entry, which currently has zero market share

and try to factor that in.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

106

On the other hand, if we are talking about truly

uncommitted entry, and if you're talking about a market

where capacity is the relevant measure of concentration

because people can move in and out of products very

easily, say a chemical factory that's got a particular

set of facilities, and what they're doing is they're

shifting the ratio of one input chemical to another

input chemical.

kind of uncommitted entry ought to be considered part of

the market.

share and fed into the competitive analysis that way.

If capacity is the metric, then that

It ought to be calculated in the market

So in short, I guess, if there's a way to

quantify consistent with how you're quantifying

the participation of existing market participants, those

who are uncommitted entrants in a real sense, I would

include them in the calculation.

would wait and look at the impact when you're looking at

the competitive effects.

MR. SHELANSKI:

If there is not, I

John, do you have some thoughts

to follow on that?

MR. KWOKA:

both Meg and George on this.

where we need to be concerned about uncommitted entry

are sufficiently few that it's quite possible to address

them in the market definition section with a notation

Yes, I agree in large measure with

I think that the instances

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

107

that where there is swing capacity freely, quickly,

flexibly adaptable to the product in question, that that

should be taken into account.

That would be as far as I think that section on

market definition really needs to go in addressing that.

I think the remainder of the concern about entry ought

to be integrated into the later section where committed

entry is now discussed in much greater measure.

I think that the present placement of the

discussion of uncommitted entry invites seemingly the

agencies and outside counsel and consultants to sort of

scour around for possibly flexible capacity to determine

market shares and to do calculations of concentration on

that basis, and I think that's really not, as a

practical matter, either necessary nor is it

administratively a good use of resources at that very

early juncture in the process.

So I think postponing that discussion until the

entire matter of entry arises more naturally is both for

practical purposes as well as on the economics a good

idea.

I also would offer the suggestion, I think you

may have asked two questions, Howard, or at least I

heard two questions in your initial commentary.

part of this is:

One

How should the guidelines take into

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

108

account the various ways that entry might factor into

merger review?

of entry into a later section is appropriate, but I

would broaden that section to be sure that it

encompasses supply response more generally.

I think that integrating the discussion

I think that there are grounds for an

integrative analysis of all of the ways that supply

response may thwart or undermine a perspective price

increase from a merger, and in particular what I have in

mind is the fact that while the '92 guidelines performed

a useful service in focusing attention on entry issues,

elevated entry to an important place in the analysis,

the '92 guidelines also downgraded something which I

think is equally important, and that is the role of

potential competition.

competition or entry as a defense to an otherwise

problematic merger, that's in many ways the thrust of

many commentary, but I think the issue that I have in

mind is where an incumbent firm actually acquires a firm

deciding to enter, a constraining outside firm.

Not so much potential

While potential competition had been part of the

1982 and '84 guidelines, and certainly it continues to

play an important part in the UK, EU, Canadian, Japanese

guidelines, that has really vanished, as has our

elimination here in the '92 guidelines.

I think that

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

109

the considerations of these issues has not ended, the

Google Double Click merger, the Hopsira pharmaceutical

arrangements, mergers in the airline industry dating

back ten years to United USAir and more recently Delta

Northwest, and even as of 18 hours ago, the DOJ consent

in Ticketmaster LiveNation, all raise these issues of

the role of potential entry.

There is no longer any explicit mention of it in

the guidelines, and I would be happy to discuss -- I

don't want to take too much time right now, but I would

be happy to discuss I think the further reasons why that

deserves to be re-introduced as an explicit part of any

guidelines revision, and I also have some suggestions as

to how that might be done.

MR. SHELANSKI:

Josh, do you want to pick up

either on the initial discussion that George and Meg

sparked or also addresses John's point for a broader

supply side analysis of an accomplished's repositioning

and potential competition?

MR. WRIGHT:

I will try to do a little bit of

both, with a lower degree of difficulty now that the

lights are back on.

So one of the things I hear emerging from the

first three comments is something that I agree with

wholeheartedly, which is whatever we're going to say

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

110

about entry, a unified theory I think was the term that

was used, a unified theory that talks about supply side

responses generally, that talks about I think retaining

this distinction between committed and uncommitted entry

as a conceptual matter is fine.

I mean, they're obviously important economic

differences for how we think about entry with respect to

its ability to constrain prices between uncommitted and

committed entries, so I think conceptually, it's

perfectly fine to retain that discussion, and whatever

unified theory on entry that we might have might include

in the guidelines.

Repositioning I think is also something that

could be included in that sort of section to make it a

little bit more clear how the agencies are evaluating

issues of repositioning.

that will become increasingly important.

I think that that's an issue

I think such a unified approach to entry that is

a little bit more clear on how the agencies are

approaching the ultimate question of how the supply side

responses are either counteracting competitive effects

or constraining the ability to raise prices will I think

ameliorate some of the problems that would arise if

there are some who support moving supply side into

market definition.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

111

I agree with George that I think this probably

produces a little bit more confusion than it's worth,

but that's conditional on having a little bit more of a

comprehensive unified section on entry that reflects

Agency practice.

My last comment, along those lines, while I do

think it's valuable to think about this distinction

between committed and uncommitted entry, and as a

conceptual matter, to the extent that this reflects

Agency thinking is valuable to include in the

guidelines.

As currently written, I think it would be

desirable to deemphasize that distinction.

line distinction now I think, to my knowledge, doesn't

reflect Agency practice.

disagree, that the Agency thinking about these issues is

consistent with that bright line distinction.

The bright

I don't think, and others may

There's a spectrum of sunk costs for entrants,

and we really don't know things like how much sunk costs

are enough to hang our hat on that sort of bright line

distinction, so I think something more reflective of

economic thinking and Agency practice would be to write

up that conceptual distinction in a way that's

consistent with the idea that this is a spectrum and

that we think about these different types of entry in

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

112

different ways.

MR. SHELANSKI:

Meg, did you want to follow-up

on that?

MS. GUERIN-CALVERT:

I was going to emphasize

one clarification maybe that I thought of as George was

speaking.

now is, in a way, conceptually where it belongs because

in essence, you are saying that it is feasible with the

hypothetical price increase that you would draw in the

appropriate capacity, to use John's phrase, and the

appropriate constraint.

I think that where uncommitted entry is right

I think where the difficulty is is really more

so in execution, which is how is it, to go to George's

point, that you actually try to measure what that supply

response is and what its influence is, and I think the

importance of really trying to keep that concept, even

if you move it into a unified theory, is not to all of a

sudden raise the threshold such that you are putting a

greater burden on having to evaluate that, but maybe

just figuring out how better to take it into

consideration.

The reason why I think it's important is I would

differ some with John in that I think where the economic

literature supports is that there is, in a great of

industry, a very large amount of repositioning of nearby

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

113

firms and nearby products, and so the concept of

uncommitted entry is one that resonates in a lot of

industries.

research has shown that if you are positioned at one or

other end points, you do have an influence on pricing,

which is consistent with uncommitted entry, and also

with potential entry theories.

In airlines, for example, I think a lot of

So I do think it's important to try to figure

out how best to articulate both the economics as well as

what the principles are as to how the tests are going to

be applied.

MR. SHELANSKI:

MR. CARY:

George?

Yeah, I think the point that Meg just

made and following up on Josh's point of the continuum,

the reality is that most of the deals that come through

agencies these days are differentiated product deals,

and many of the deals, when they're reviewed, go back to

this question:

relatively narrow space?

long will it take for other firms to reposition?

Will there be unilateral effect in a

And the question becomes:

That's a continuing question.

How

It's not a

question of:

Is there or is there not a sunk cost?

is a question of:

to reposition a consumer product based on advertising

money?

What are the sunk costs for example

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

114

That can be large.

It can be small.

It can be

It can be positive.

It's very unlikely to be a

zero.

binary kind of decision, which to me suggests again that

you move it back into the competitive effects entry

analysis and evaluate it as part of the competitive

dynamic rather than treating it upfront, especially when

you're using unilateral effects analysis in the narrower

market focus there.

MR. SHELANSKI:

MR. KWOKA:

MR. SHELANSKI:

MR. KWOKA:

I would like to --

May I just add something?

Sure.

I don't so much disagree I think in

principle with what Meg says, but the problem here, of

course, as in so many places in the guidelines is that

there is a continuum, and the guidelines seek to draw

cutoffs and make arbitrary distinctions.

It certainly is true that there is both capacity

and repositioning that can be brought online and becomes

relevant in a short period of time.

course of uncommitted entry doesn't allow for time at

all.

available.

The pure term of

It does mean that it's virtually instantaneously

The minute you move away from that polar

extreme, then you're into the continuum, the question

becomes a practical one it seems to me of whether it's

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

115

more administratively feasible bringing the matter to a

resolution more quickly to deal with it upfront or to

postpone it later.

My own view is that unless the capacity is

virtually instantaneously available, and I submit that

that is -- it's clear when that's true and it doesn't

happen all that often, then I would argue for postponing

the issue of repositioning and capacity availability

where the latter requires some time and effort, to the

later point where supply responses are more fully

accommodated.

I certainly agree with Meg, however, fully that

in the airline case where one end point of a route is

served by another firm represents a potential entrant,

that it is precisely that case, which as Meg knows, she

and I have talked about this, that I have investigated

and others in some research which look at the USAir

Piedmont merger, it's now 20 years ago, but it was a

very good example of where two carriers merge where in

some very minute routes they represented two incumbents.

There we know what's supposed to happen, and

indeed it did.

Prices rose by maybe 10 or 12 percent,

but the more interesting empirical exercise was to look

at routes where one of the two was an incumbent and the

other served an end point but not the route itself.

For The Record, Inc.

(301) 870-8025 - www.ftrinc.net - (800) 921-5555

116

So there's no change in concentration measured

among incumbents, and if one looked only at markets

where concentration changed as a result of the merger,

one would ignore all those routes.

to be quite numerous because these are network carriers,

and they intersect and overlap in lots of different

ways.

statistically significant price increase of about 60

percent as great as where the two firms were incumbents.

They would turn out

There was indeed a price increase, a

It seems to me that where two firms meet each

other in that fashion, whether they happen to be both

incumbents in other markets or not, a pure and potential

competition merger is simply the case where there is no

change in concentration measured by incumbents, but

someone

This text is long and has been trimmed here. Open the source document for the complete record.

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.