DEPARTMENT OF JUSTICE

Agency decision

Ask Donna

What actually matters in this document.

Text

FEDERAL TRADE COMMISSION

MERGER WORKSHOP

DAY THREE

THURSDAY, FEBRUARY 19, 2004

9:00 a.m.

FTC Conference Center

601 New Jersey Avenue, N.W.

Washington, D.C.

Reported by: Rita M. Hemphill, C.V.R.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

PARTICIPANTS:

Panel I:

Efficiencies/Dynamic Analysis/

Integrated Analysis

Moderators:

Alden F. Abbott and Mary T. Coleman(FTC)

Panel:

J. Mark Gidley (White & Case)

Ilene Knable Gotts (Wachtell, Lipton,

Rosen & Katz)

William J. Kolasky (Wilmer Cutler

Pickering)

Robert Pitofsky (Georgetown University)

David T. Scheffman (LECG)

Joseph J. Simons (Paul, Weiss, Rifkind,

Wharton & Garrison)

Vincent Verouden (EC/DG Comp.)

Panel II:

Economists and Lawyers Roundtable

Moderator:

R. Hewitt Pate (DOJ)

Panel:

William Baer (Arnold & Porter)

Jonathan B. Baker (American University

Wayne D. "Dale" Collins (Shearman &

Sterling) James Loftis (Gibson, Dunn &

Crutcher) James F. Rill (Howrey & Simon)

Daniel L. Rubinfeld (University of

California, Berkeley)

Robert D. Willig (Princeton University)

Dennis W. Carlton (Univ. of Chicago)

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

C O N T E N T S

PAGE

Welcome and Introduction

Panel I:

Efficiencies, Dynamic Analysis

Integrated Analysis

Panel II:

Economists and Lawyers Roundtable

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

118

P R O C E E D I N G S

MR. ABBOTT:

Good morning.

Welcome to the

third day of a Joint FTC/DOJ Workshop on Merger

Enforcement.

Policy and Coordination in the FTC's Bureau of

Competition.

I'm Alden Abbott, Associate Director for

I am joined here by the co-moderator for our

panel, Dr. Mary Coleman, Deputy Director of the FTC's

Bureau of Economics.

Thus far, workshop panels have focused on

discrete parts of the Guidelines:

concentration, competitive effects and entry.

commentators, however, have noted that it would be a

mistake to view individual Guidelines provisions in

isolation.

Market definition,

Several

However, the implications of such statements

that the Guidelines provisions should be viewed

holistically, to use the New Age term, with simultaneous

consideration of different factors that enter into a

Guidelines analysis, have not been developed.

The aim of this morning's panel, entitled

Efficiencies, Dynamic Analysis and Integrated Analysis,

is to explore what it means to carry out such a holistic

or integrated analysis.

efficiencies and dynamic considerations, with particular

The panel will also focus on

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

attention to their role in the overall competitive

assessment of a proposed merger.

We are fortunate to have a true all star cast

assigned to assist us in carrying out our daunting task.

Their academic and professional laurels are so

impressive, we could take up the entire morning

recounting them.

will refrain from doing so.

affiliations.

Given the time constraints, however, I

But I will note their key

Our first speaker will be Dr. David Scheffman,

who is a recidivist, already having served on the panel,

former Director of the FTC's Bureau of Competition,

currently a director at LECG and Adjunct Professor at the

Owen Graduate School of Management at Vanderbilt

University.

Dave, I hope, will tell us what this integrated

approach is all about and help dispel the fog and give us

a clear sky.

Simons, former Director of the FTC's Bureau of

Competition, and currently co-chair of the Antitrust

Group at Paul, Weiss, Rifkind, Wharton & Garrison.

Dave will be followed in order by Joe

Joe will be followed, again in order, by Mark

Gidley, co-head of White & Case's antitrust group, and a

former Acting Assistant Attorney General and Deputy

Assistant Attorney General for Antitrust.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Mark will be followed by Ilene Gotts, a partner

at Wachtell, Lipton, Rosen & Katz, who has held various

ABA Antitrust Section leadership positions, including

membership on the section counsel currently.

Bill Kolasky, co-chair of Wilmer, Cutler &

Pickering's antitrust and competition practice group, and

a former Deputy Assistant Attorney General for antitrust,

will follow Ilene.

Then we're saving our superstar for near the

end, the Honorable Robert Pitofsky, former FTC Chairman

and Dean of Georgetown Law School, currently a professor

at Georgetown and of counsel at Arnold & Porter.

will give us the big picture to put everything together

for us, we hope.

And he

And then our panel will end on an international

comparative note with a presentation by Dr. Vincent

Verouden, who has been an economist at the European

Commission's DG competition since 2000, and Dr. Verouden

has already spoken, and we're delighted to have his

comparative perspective.

enlighten us on EC Merger Guidelines issues.

And he will undoubtedly

After our seven panelists have spoken, we will

have a brief 10-minute break followed by a discussion and

question-and-answer session among the presenters and

moderators.

Given the fact that at least some of these

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

speakers I know are not shy, Mary and I look forward to a

lively and thought provoking session.

So, Dave, will you lead off, please?

DR. SCHEFFMAN:

Yes.

I'll try and be a little

thought provoking or obnoxious as I usually am, rather

than just talk -- and I will talk about integrated

analysis, but I'm going to first talk about efficiencies

-- can we get this slide up so everyone can see?

I thought we'd give a report card since I left

the agency not that long ago and talk about how I think

the agencies are doing -- obviously, I have much more

experience at the FTC -- with respect to efficiencies.

It's hard to tell, because you don't see many

tough cases with efficiencies compared, say, to the '80s.

I think that's because the outside believes that

efficiencies aren't given much weight and they advise

their clients not to try them.

well-known problem.

That's well known and a

With all respect to Chairman Pitofsky, who I

have the highest regard for, baby foods was not a helpful

development, and I'm sure he will respond.

involved in the case, and maybe I misunderstood the

facts, but it's not surprising that it's not most

economists' favorite case.

in favor of efficiencies, I don't think, although I can't

I wasn't

Nor is it a favorite of those

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

say that, because Bob did promulgate, you know, was an

instigator of the '97 revision and does believe in

efficiencies.

say.

So I'd be interested in what he has to

There are some good things about baby food, I

think, such as it's good to have a clear decision that

when three to two is a real three to two.

high hurdle, and I think most economists in the

mainstream would agree that that's where the really high

hurdle should be.

said, and he worked on baby food, his concern with baby

food was that it wasn't really three to two.

really like one to two or 3.1 to two, 2.1 to two.

It is a very

I think as Chairman Muris has often

It was

What I've heard from the parties, but I don't

know a lot about this, is the ex poste story is not a

pretty one.

retrospectives on hospitals.

some of these cases where efficiencies were a significant

issue and look at what happens.

I know the FTC is engaged in various

I urged them to look at

My favorite case, one in which I was an expert

soon after leaving the FTC and was one of the FTC's and

Ann Malester's best cases, was the tank ammo case.

was the two to one defense merger case, a big victory for

the FTC.

exactly opposite from what the FTC said it would.

This

Everyone has known after that it turned out

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

And so

I think my view of the track record, from anecdotal

evidence, on how the agencies have treated cases that had

serious efficiency claims is not good.

the research has yet to be done.

But, you know,

However, the record is not as bad on

efficiencies as all that.

Efficiencies are important.

They're more important than people and counselors think

they are, but not in the Guideline’s sense.

that much more the case now than it was in the '80s, is

that the agencies rely on customer opinions.

industrial mergers in which you have a relatively small

number of sophisticated customers, even in a pretty

concentrated merger, if the customers say we're not

concerned, it's unusual that the agencies will challenge.

That has locked into it efficiencies and other

considerations.

quite comfortable the agencies almost always get right.

So in those cases I'm not so worried about the

One thing is

And so in

I think those are the mergers which I'm

efficiencies.

suppliers can't convince the customers of the

efficiencies when they have big, sophisticated customers,

then they haven't fulfilled their burden and the merger

is likely to be challenged and it probably should be.

The real problems are in the cases, all the

Another way you could say it, if the

cases where you don't have a relatively small number of

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

sophisticated representative customers -- the oil

industry, branded products mergers, supermarkets, et

cetera.

of customers, where you don't really have sophisticated

customers to speak for the benefits or potential costs of

the deal.

think what we've done in oil for 20 years, and it's been

going on a long time, in the way efficiencies have been

treated has really been quite counterproductive.

everyone, including FTC staff, believes that there have

been substantial efficiencies gained from a lot of the

oil mergers.

not given much weight in oil merger enforcement.

still, as it has been for 20 years, largely a structural

enforcement policy.

Those are cases where you have middlemen or lots

I think that's where the real problem is.

I think

Nonetheless, the efficiencies are usually

It's

I will say, the other way that efficiencies

count is that, it affects remedies, which I don’t think

is really recognized.

that, well, actually this is a good deal, in a general

sense not in specific Guideline sense, and are crafting

remedies that may impact the achievement of the benefits

in the deal, you'll see in lots of consents kind of

exotic, flexible consents at the agencies.

cases, my view of that of what's going on is the agencies

are crafting things to alleviate the competitive problems

When staff and the agencies think

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

In some

but in a way that allows the potential benefits of the

deals to go forward.

But efficiencies can be very important.

Efficiencies sometimes, are part of the reason why you

don't get a second request.

in which it's really just a structural case where you

could go forward because it’s a five to four merger, but

where you don't have complaining customers.

got a good story about why the merger is taking place,

that can be part of the reason why either you don't get a

second request or why a case is closed.

example of an interesting case on efficiencies is drug

wholesalers II, which was an interesting situation where

the Commission had a prima facie case to be able to block

the merger that came back second time, and for lots of

complicated reasons which I don't fully understand and I

was there near the end –- the Commission did not block

it.

Commission staff that there were benefits of the mergers,

that just having two mergers in that industry going from

four to two in the case that was litigated was a no go

proposition.

part of the reason why drug wholesalers II matter was

cleared.

And in close cases, in cases

If you've

And a good

But certainly part of it was a belief within the

The efficiencies were clearly a significant

And I'll get to the integrated analysis, along

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

with the argument and belief that the merger really

wasn't going to be anticompetitive anyway, in part

because it was going to be efficient.

strengthen the smaller competitor.

Okay.

It was going to

How do the agencies actually do things?

Well, they do it really the way the Guidelines say, which

is part of the problem.

in which if you've got a case where the staff has a

pretty strong belief that the matter is anticompetitive,

there's no efficiency is going to turn that round.

really has to do with the stronger the belief by the

staff, and their decisions are usually ratified, and

obviously the belief by the ultimate decision maker about

the likely anticompetitive effects of the deal, the less

weight efficiencies get.

There really is a sliding scale

So efficiencies really are in play in the gray

area where you've got a case where there isn't a strong

belief and basis for believing the merger is

anticompetitive.

This is the reason why we need to get more

integrated analysis because the way the efficiencies

actually get treated in the more difficult cases, it's

not really the proper way.

reasons why the merger might not be anticompetitive is

because it's efficient.

That is, you know, one of the

Talk about that in a minute.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

What's happened with the Merger Guidelines, I

think it's probably more at the FTC, because the FTC has

litigated the efficiencies provisions probably more than

DOJ, is unfortunately is the staff builds a prima facie

case that the parties can't win on the Merger Guidelines

efficiencies checklist.

merger-specific, it's not variable cost.

leads that effort for the financial analysts, and he does

a very good job on that, that's his job in a way for the

client.

to show that, you know, they're not going to be able to

get through the Guidelines efficiencies checklist.

It's not cognizable, it's not

Gabe Dagen

But if we have to go to court we have to be able

The problem with that is that gets the focus on

the efficiencies on litigation and disproving the

efficiencies.

at whether there are some real efficiencies here, folks?

And that was something I tried to do at the FTC, with

mixed success.

The question is, is anyone really looking

There aren't procedures and incentives really

to look for real efficiencies, at least within the FTC.

I don't know about the DOJ.

happen, but it happens sort of depending on which

staffers you get on a case.

what the staff is mainly looking, and that's partly on

the basis for the client, to be able to show that you can

I don't say it doesn't

Because otherwise, again,

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

disprove the efficiency claims of the parties if you have

to go to court.

And with the emphasis on disproving

efficiencies, not surprisingly, there's not a lot of

emphasis on finding efficiencies.

good merger?

efficiencies under the Guidelines test, is this really a

good deal, okay?

But is this really a

Even if we maybe could disprove the

And we are proud of transparency, and this is a

real benefit of the transparency that we tried to

increase, the Commission tried to increase.

But this is

the part where it's been least successful.

In my

experience, there's the least communication between the

parties and the staff about the efficiencies claim.

happens is, and I don't think the outside understands

this, the staff looks at the documents, deposes the

people, gets enough so that they think they could

disprove the efficiencies claim under the Guidelines.

And there's usually nothing comes back over the net from

the parties.

what's happened, right?

facie case that your efficiencies don't count.

What

It's like the other side doesn't even know

Your opponent has made a prima

Now, what we found in a couple of instances

where we actually required some transparency between the

staff and the parties, is that the staff didn't always

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

have it right, particularly on whether the efficiency was

merger-specific or what would happen but for the merger.

But the efficiency arguments, often really don't get

tested.

It's a fault of the staff in not always being open, but

the parties have to push.

And that's a fault of the parties, in part.

They really have to be serious -- they have to

say, okay, what is your basis for thinking that this is

not merger specific, that something's going to happen

independent of the merger, et cetera.

to test that to understand what the basis of the staff's

opinion is in order to come back with and answer.

is the one situation in which the parties actually have

information that the staff doesn’t, which is usually not

the case on competitive effects.

get this information.

to engage the staff and to the extent they can, request,

demand transparency.

conclusion this efficiency isn't merger specific, et

cetera?

that the basis isn't there, even though the staff has

good reason to believe what they believe, they might not

have the facts right.

The parties have

This

But the staff doesn't

So the parties really have

What is staff’s real basis for it’s

Because when you press on it, you might find

Okay.

roundtable.

We learned a lot from the efficiencies

I don't know that it's had any effect, but I

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

think we learned, you know, the merger consulting

companies have certainly not helped the consideration of

efficiencies with all the articles arguing that the

typical merger is not, quote, "successful."

think -- I don't know that it came through in the

roundtable, but Paul Pautler has a good paper on that

issue, and I've looked at lots of literature, and it's

important to understand what that literature means.

think that literature is right, but you have to be

careful about what you think it means for what we do in

antitrust.

Well, I

That is, clearly the leading reason why mergers

aren't successful is because the acquirer pays too much.

That's not an antitrust issue.

efficient merger.

winner's curse sort of thing.

why mergers are not “successful.”

reason.

You can have a perfectly

They may have paid too much.

It's the

That's the primary reason

That's the main

But another important reason, which is related

to what we actually do, has been recognized in recent

years. Another important reason why mergers are not

successful is that the mergers lose revenue that they

didn't expect.

business.

it might be because it's anticompetitive or could be

That is, they lose customers and

In a horizontal merger, why?

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Well,

anticompetitive, but it could be because the customers

actually react adversely to the merger for other reasons.

That's related to reliance on customer opinions and

things like that, so that reason is important.

fit with what we do.

reemphasizes how important sophisticated customer

opinions are.

It does

It does indicate how -- it

The roundtable clearly indicates, other things

equal, that horizontal mergers are more likely to be

successful and efficient, you know, if there is “fit” --

all the stuff about fit and being in a similar business,

da, da, da.

been on that, and that goes back to the Scherer and

Ravenscraft papers.

The literature is very clear and always has

The other thing which is true, if you listen

carefully, and it's that straightforward cost savings are

generally realized.

to the shareholders, what they report to the 10-K, what

they report to the street, you know, they say we're going

to reach these cost reductions.

much more than on average these days, they do, if those

are costs savings which aren't pie in the sky but

standard sort of consolidation savings.

savings that we worry about, cost reductions we worry

about, in horizontal mergers, they're, you know, you need

If you look at what companies report

And on average, in fact,

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

So the cost

a basis for believing that they're there, obviously, and

how they're going to achieve them.

do so -- you can have pretty high confidence if you have

some basis that they're going to actually be realized.

But they're going to

The other thing from the efficiencies

roundtable is that planning and implementation is really

important.

That's a leading reason why mergers aren't

successful.

That has something to do with what we do.

Gun jumping is a problem despite -- I don't think that

the efficiencies roundtable was successful in explicating

that issue.

significant issue.

planning as probably even they could do if they didn't

have such conservative counseling.

what the agencies should be looking at in terms of some

evidence of serious planning of how the merger is going

to be implemented to believe that the efficiencies are

going to be realized.

Gun jumping I continue to believe is a

Okay.

It is why companies can't do as much

And it also indicates

How do efficiencies fit into the

analysis?

This gets to the integrated approach.

Efficiencies are related to -- a merger that's in a five

to four industry without unilateral effects is

significant -- one point of the merger is to become

significantly more efficient, and not dominant.

almost a prima facie case for economists.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

This is

Why isn't the

industry going to be more competitive?

not be.

not.

analysis.

a merger through, where both staffs at both agencies,

lawyers and economists, say, well, yes, that's a pretty

good efficiency story is where you've got a manufacturing

merger with a combination of batch and continuous

production processes, so you can do lots of interesting

things in terms of getting more output out of the same

facilities, et cetera.

it's an industry where there is not a viable theory that

you're going to significantly reduce capacity utilization

because the costs of reduced capacity utilization are too

small, or too large.

that's not batch and continuous.

I mean, it might

But you need a pretty convincing story about why

So efficiencies obviously impact the competitive

The ideal case, and one I think, that will get

You can get higher capacity if

This is like the oil industry, but

Then you've got, well, how is the merger going

to be anticompetitive?

more almost surely as a result of the merger.

to say you might not have other reasons to believe that

the merger is problematic, but you should have some

strong reasons to believe, say, in a five to four merger

why this merger is going to be problematic.

efficiencies are important.

The party is going to produce

That's not

Let me begin with the integrated analysis.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

First what we have to remember -- and this is not the way

the law works at all, and it's not the way the

enforcement agencies really work in practice, is that we

define markets in antitrust and then we do the analysis

after that.

Well, not all markets are alike, and the real

basis of the market is not equally strong.

This is not

recognized.

clearly.

sometimes the reason why the courts reject cases based on

market definition is for squishy reasons that probably

don't have to do with market definition.

things are all related.

market definition will usually spill over into

competitive effects.

example, and Joe will talk about this more in a minute.

The markets are taken as given to us

This is the way the courts decide, although

But these

In particular, weaknesses in

That was true in cruises, for

All these things have to fit together.

The

weaknesses in market definition are important in

competitive effects analysis.

outside don't understand is that you're never going to

win an argument with the staff on market definition in a

hard case.

will start talking about, well, if this is your market,

given the defects in that market definition, this is why

the competitive effects that you're worried about are not

I think what people on the

It's not that you should give it up, but you

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

going to occur.

Okay.

What efficiencies should count?

I have

long believed –- going back to the '80s and the papers

that were written at the FTC, like Fred Johnson's paper -

- that this whole idea of passthrough incremental costs

and passthrough is a whole red herring.

should be important in how the agencies think about

things -- it's of some relevance, but it's not the key

issue.

efficiencies is really the sliding scale.

I don't think it

The key issue in the way the agencies look at

Joe Simons is going to talk more about a more

sophisticated, in my view sensible approach.

fixed costs count?

about that.

efficiencies roundtable about exactly why anyone who

works with real world companies and looks at their pro

formas include fixed costs.

lots of companies, do they count their fixed costs in

their pro formas and in their decisions?

do.

that, and other people will talk more about that.

Bill Kolasky will talk

Dave Painter has a nice presentation at the

I teach MBAs and work with

Of course they

There are good economic, sound economic reasons for

Of course.

Should

Joe Simons will talk more about the integrated

approach.

Let me finally talk about dynamic competition,

which hasn't been so prominent lately.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

It is an

important issue, and it's going to come back.

I think

it's very interesting to look at what happened to

dot.com.

dot.com mergers that, you know, if the agencies wouldn't

have been so busy when the mergers happened, they would

have stopped them.

three to two mergers that looked, you know,

problematic based on the way the agencies look at things.

They were let go just because the mergers were

There were all sorts of consolidations in

There were a lot of two to one and

pretty

small and the agencies were very busy.

present dynamic competition issues.

these issues in Monster/Hot Jobs, which I think was

probably a good case but presented some very difficult

issues.

competition issues, because the industry is based on IP

and in dynamic economy.

these, and the agencies really don't know how to analyze

them, in my opinion.

time.

Those mergers

The Commission faced

We'll see other deals involving these dynamic

We're going to see more of

And I've more than used my

Thank you.

MR. ABBOTT:

Thank you, Dave.

You've certainly

given us a lot of food for thought and digestion.

now Joe Simons will tell us -- give us a chart which will

explain all future efficiency analyses and solve our

problems, we hope.

MR. SIMONS:

And

Well, that's a little ambitious, I

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

think.

putting this panel together.

hard work, and I just want to thank them for that and for

inviting me to appear here today.

First, I just want to thank Alden and Mary for

I know they put in a lot of

What I'm going to do, what I have in mind

really is to present what I think is actually a fairly

simple and straightforward way of doing an integrated

approach to analysis of anticompetitive effects and

efficiencies in mergers.

And here's what I have in mind.

Let me be very specific about the first

principle applied.

I think that's really important.

you don't know what it is you're looking for, it's kind

of hard to find it.

important, and it is prohibiting mergers that reduce

consumer welfare.

competitive effects and efficiencies.

exercise basically is to make a judgment or prediction

about the overall effects of a merger over a reasonably

foreseeable period of time, two, three, five, years

something like that.

So the first principle is really

That principle applies equally to

And the ultimate

Now, you know, based on what Dave said just a

few minutes ago, obviously this is not an easy thing to

do in practice.

you're moving in and you know where you're supposed to be

going, then you have a better chance of getting there.

But at least if you know what direction

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Even if you can't get really close with the

tools at hand that you have today, it's better to create

a framework and to develop the tools over time.

that has really been shown to work with respect to the

Guidelines that were issued in 1982.

who are old enough to remember, when those Guidelines

first came out, there was a hue and cry that those

Guidelines were way too theoretic, particularly the

market definition paradigm, too theoretic, completely

nonoperational.

I think

For those of you

Today, with the advancements that we've seen in

merger analysis, I would say that the market definition

paradigm is probably the most practical tool in all of

antitrust, and when it started out, it was nothing.

the only improvement area is fairly theoretical, but I

think it can be used in practice, at least as a tool is a

type of a sensitivity analysis to see how things fit in

and the relative importance.

So purely from a theoretical point of view

then, the way to determine whether the overall effect of

a merger is to reduce competition, or reduce consumer

welfare, is to perform what I refer to as a risk-

adjusted, net present value calculation.

what we do is we estimate the magnitude of any price

effect, and by "price effect," I'm including qualityFor The Record, Inc.

Waldorf, Maryland

(301)870-8025

In other words,

adjusted price, innovation, et cetera.

magnitude of that price effect, the probability that it's

going to be realized, its timing and its duration, and

you do the same for efficiencies.

estimate the magnitude of the efficiencies, their likely

effect on price, the likelihood that those efficiencies

are going to be realized, their timing, and their

duration.

the expected benefits to consumers are over time, and you

make a net present value calculation.

You estimate the

So that is, you

And then you see what the expected costs and

So whether the merger is challenged or not

depends on whether the NPV is positive or negative for

the consumers.

It's fairly straightforward.

It's a tool

that's used every day in the business world.

And I have

an illustration if we can put that slide up.

(Slide.)

Okay.

This is just a spreadsheet, and it

involves the following example.

with a potential merger of to widget producers and we

conclude as follows.

witnesses, documents, third parties, everybody, we

conclude that the market is widgets with an 80 percent

probability.

Suppose we are presented

After a lengthy investigation of

We conclude that entry will not occur for two

years, also with an 80 percent probability, and we

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

conclude that the anticompetitive effects given the

market definition and the entry conclusions, are that

we're expecting a 10 percent rise in price for the first

two years, and we expect that within 80 percent

probability.

On the efficiency side, we are expecting that

marginal costs will decline and impact price by 2

percent.

probability and that begins in year two and continues

through year 5.

We're expecting that with a 70 percent

We concluded that pecuniary costs would decline

and impact price by 1 percent, with a 70 percent

probability beginning in year one and continuing through

year 5.

decline and impact price by 1 percent with a 70 percent

probability, that beginning in year three, and continuing

through year 5.

Then we also concluded that fixed costs would

All right.

So these assumptions are all

summarized on the spreadsheet, which performs the net

present value calculation for that flow of positive and

negative benefits to the consumer that result from this

hypothetical transaction.

It shows that even though the merger is

projected to raise price by 10 percent for two years, the

net projected effect on consumers is actually positive.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

So we'll just through this.

If you look on the left-hand

side, you see we have competitive effects, market

definition, entry, anticompetitive effects, all with

probabilities of .8.

so the probabilities add up, and the total probability is

51 percent.

that by the probability and you get an expected harm to

the consumer of 5.1 percent, and that appears over the

first two years so you see the columns on the right, year

one has a negative 5.1 and year two has a negative 5.1,

and then years three, four, and five shows zeroes.

In efficiencies the same thing.

Those are determined by each other,

The potential harm is 10 percent.

Multiply

The marginal

cost probability is 70 percent.

benefit and the same for fixed cost.

2 percent from marginal costs and pecuniary benefits is 1

and fixed costs is two, and then the expected value of

those benefits given the risk, and then the columns to

the right show how those play out over time, given the

assumptions in the hypothetical.

The same for pecuniary

The expect harm is

And then just adding up the total effects down

the columns for year one, two, three, four, 5, you see

that the expected effect in year one is a 4.4 percent

increase in price; for year two is a 3 percent increase

in price; and then years three, four and 5, the benefits

are to the consumers 3.5 percent each year.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

And when you do a net present value

calculation, you see at the bottom there in the lower

left-hand corner, if you can see that, it's positively

slightly about 1 percent, or .68.

One thing more, I took a discount rate of 10

-- I basically just picked that out of the air.

Obviously that has a significant impact on whether the

result is positive or negative.

hard about, you know, what the discount should be, except

for the fact that it probably relates to the market in

question and who the consumers are.

are going to have different discount rates applicable to

them probably.

I haven't thought real

Different consumers

I don't mean by putting this chart up here and

having, you know, somewhat precise numbers like 5.1, to

suggest that, you know, we should do a calculation with a

great degree of mathematical precision.

have the tools to do that, at least not yet.

We don't really

On the other hand, what I think this allows you

to do is to see in a broad way what's going on, to be

cognizant of the ultimate purpose of this exercise, and

perhaps most importantly, be transparent about the

assumptions that we're making in the analysis.

being transparent, I think we can expose some

inconsistencies in what we may be doing and not realizing

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

And by

some flaws, and I think we can also provide incentives to

develop new techniques that will make this kind of

framework more practical.

Among other things, I think what this approach

does or this structure does is help to define what

efficiencies are cognizable, how to evaluate them, and

more significantly, how to weigh them, which I think is

something that's seriously missing now.

And just as importantly, it does the same thing

for competitive effects.

antitrust community about, well, what's the standard for

proving efficiencies?

to, well, what's the standard for proving competitive

effects?

And if you look at this type of analysis, you'll see that

they directly relate to each other.

There's been this debate in the

But we've never really tied that

And do the two relate to each other at all?

So it's not only the size of the competitive

effect that's significant, it's the likelihood that it's

going to occur and the duration.

how long, how high, how much passthrough does there have

to be to make the efficiencies offset those perceived

harms?

That then impacts well,

So I think what this type of analysis

demonstrates really is the larger, the more likely, and

the longer the adverse competitive effect is, the larger,

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

the more likely, and the longer must be the offsetting

efficiency effects, and the weighting is then determined

by the NPV calculation.

One of the things that is demonstrated by this

type of example is fixed costs.

If you look at the fixed

costs, you see that they're occurring by assumption here

in years three through 5, and what it shows is they

really can be determinative, and they shouldn't be

ignored or treated with the back of the hand, which I

think is the tendency now.

I mean, basically, what would happen is, people

would say yeah, we got these big fixed cost savings, and

the agency folks will say, yeah, that's really nice and

maybe you do, but, you know, those really get little

weight.

There's really no mechanism to kind of figure out how

important they are.

with that.

Well, you know, how much weight should they get?

And so, I think this really helps

And the other thing that this kind of focuses

your attention on -- and Dave made a reference to this --

is the probabilities of the anticompetitive effect,

particularly if you're talking about analysis that's kind

of like compartmentalized.

So we do a market definition, then we do a

competitive effects analysis, assuming that market

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

definition, and there's some risk associated with each of

those, and I think that kind of tends to get lost, that

once we conclude the market is X, we're making kind of

like a subconscious assumption that it's X with 100

percent probability, and that's really not the case.

So I think this type of analysis kind of helps

to expose what's really being assumed, and then the

magnitude of the effects versus the efficiencies. And I

think something like this would really help get us over

the hump that we find ourselves at now.

today basically is when somebody comes in with a merger,

they do their efficiencies analysis, the parties do, and

as Dave said, it goes over the net to the staff, and it

really never comes back.

really pitch it that much, and I think that's true for

the following reason.

What happens

And I think the parties don't

I think they intuitively understand that it's

not going to save an otherwise anticompetitive merger.

By the same token, they don’t understand that if the

staff has some serious doubts about the confidence of,

you know, their projection of an anticompetitive effect,

the fact that the merger has efficiencies associated with

it will make the staff feel more comfortable about not

challenging the merger.

that kind of a circumstance to go back and forth over the

And so there's really no need in

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

net.

The problem, though, is that the efficiencies

really are playing a very little role in the merger

investigation, and I think that's something that needs to

be rectified.

know, can provide some context in which to do that.

And hopefully, this type of structure, you

Thanks.

MR. ABBOTT:

Thank you, Joe, for that succinct

presentation.

effort, as I say, to rank the probabilities of the

different factors going into net consumer welfare

analysis in sort of a simple manner.

that inspires some of our speakers.

And it's the first time I've seen such an

So we'll see if

Now we turn to Mark Gidley, who is borrowing

from Voltaire and Professor Pangloss.

I know he has a

paper for distribution you may want to pick up outside at

the entrance entitled "Misuse of the `Merger-Specific'

Requirement:

Best of All Possible Worlds."

Merger Analysis is Not the Search for the

Mark, on that literary note, please proceed.

MR. GIDLEY:

Thank you very much, Alden.

And

let me say, first it's good to be amongst you.

I see a

lot of old friends.

five merger cases since I left the DOJ, with some success

and with some failure, and I may acknowledge that as I go

I've had the pleasure of litigating

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

through my remarks.

I would also say that I come down on the side

of Spinoza on the issue posed by Candide, but I won't

expound on that today.

I do turn out to be an optimist.

My main point in citing Professor Pangloss's

oft-attributed comment is that I don't contend, and I

don't think most people contend, but unfortunately I

think this is the way it plays out sometimes in the

conference rooms of both agencies, that merger analysis

is the search for the best sort of Platonic ideal

transaction that the parties could ever enter into.

Human beings will never get there, and I think that I

really come down on the side of an analysis that's very

similar to what Joe is proposing:

the pro and anticompetitive effects with a serious

consideration of efficiency.

A net assessment of

Segueing into what I'd like to address for a

few minutes today, I want to talk about the 1997

Guidelines and the emphasis on merger specificity, which

is really to my mind one of the new hallmarks of the 1997

Section four.

And I was struck a little bit by our earlier

panelist, David Scheffman. In my own experience and maybe

in my own conference rooms as we prepare for meetings

with the government, we've tended to see Section four of

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

the 1997 Guidelines as something of an obstacle course

out of the Marine Corps.

argument and your beautiful spreadsheet that actually

analyzed the merger on going out five or 7 years, the

nail has to go through the rope that's hanging over the

net, and it's got to go over the 30-foot wall of merger

specificity.

that to become cognizable.

they're just efficiencies, and they're either likely or

they're unlikely.

to time bring to the agencies efficiencies that aren't

real world or that they don't tell their businesspeople

or that they didn't get board approval for, and all of

that is relevant.

You know, your efficiency

It's got to go meet this, it's got to meet

And in the real world,

And undoubtedly, people will from time

But what I'd like to focus today on is the

language of merger specificity, and in my own work on

this paper, I found that the actual language of the 1997

Guidelines is not the way we practitioners or the

government officials use the phrase "merger-specific."

In general, the phrase "merger-specific" is

really used as an epithet.

You might have very good

efficiencies that your deal really will cause, that will

really cause these efficiencies.

well, they're not merger-specific.

with the back of the hand, that invocation phrase,

But somebody will say,

And that's it.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

And

oftentimes even private practitioners back-off bona fide

efficiencies.

ourselves today in the discussion.

I hope to back-off of that phrase

I start by acknowledging my own sinfulness in

words.

1992 Guidelines, and I think we really punted on

efficiencies.

evidence, but we really didn't analyze efficiencies.

of our efforts and brainpower -- and there was a lot of

brainpower applied -- was on competitive effects, and I

think that was probably all the bandwidth the human

beings at the agencies had in 1992.

have the bandwidth to take on efficiencies.

I sat at the Justice Department when we did the

We took on our clear and convincing

All

We really didn't

So my compliments to Chairman Pitofsky and

others for in 1997 trying to explicate what efficiencies

we're going to recognize.

1997 Guidelines were a good evolution, but I think

they're a stopping point on the journey.

seven years of experience, particularly with merger-

specificity, we can really say what's worked and what's

not worked.

So I think to that extent, the

And now, with

I have a slide that's just the language, Alden,

of paragraph three from Section four of the Guidelines

that I thought I would just put up.

Because as I prepared my remarks, I actually

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

found language in there that I haven't been using as a

private practitioner.

may start doing so.

(Slide.)

It's up.

believe in an oral tradition.

And I'll warn the agencies that I

I'll read it for those of you who

Paragraph three says:

The Agency will consider only those

efficiencies likely to be accomplished with the

proposed merger and unlikely to be accomplished

in the absence of either the proposed merger or

another means having comparable anticompetitive

effects. These are termed merger-specific

efficiencies.

practical in the business situation faced by

the merging firms will be considered in making

this determination; the Agency will not insist

upon a less restrictive alternative that is

merely theoretical.

Only alternatives that are

And if you just take a look at this first sentence, it

seems to me, just doing a little bit of jurisprudence on

the first sentence, there are really two elements.

is that the efficiency that we're going to talk about in

Section four is caused by the merger.

that "merger-caused efficiencies".

So I would call

But there appears to be a subset of

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

One

efficiencies that are merger-intrinsic or not merger-

intrinsic.

unique to the merger, we'll ignore it.

in that sense that most of the time when people say with

the back of the hand, well your efficiencies are not

merger-specific, you could go buy another company and get

the same purchasing efficiencies.

that I think has devolved into a violation of the last

sentence, which is that the agency will not insist on a

less restrictive alternative that is merely theoretical.

In other words, if the efficiency is not

And I think it's

And it's that element

In my experience in the last seven years,

oftentimes people in a conference room will posit a

theoretical alternative and no one will debate whether

it's practical or not, or whether the managements could

ever really go there, or the time difference between

abandoning this merger and the time cost of trying to do

something else.

say do a production joint venture, and there are a lot of

reasons for why a 50-50 production joint venture starting

today from the time that we're talking in the conference

room, could be three to five years away from realizing a

fraction of the same synergies and cost savings that a

merger can efficiently realize maybe in a matter of

months or weeks, depending on the closing date.

For instance, in a merger people will

So there's a distinction that I draw on in the

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

paper between merger-caused efficiencies and merger-

intrinsic efficiencies.

hopefully clear, I reject the merger-intrinsic concept.

I don't see it as necessarily a useful concept, because I

think in practice, what happens is, you wind up battling

against theoretical alternatives, and that that last

sentence in the Guidlines, which is a good, laudable

sentence in practice has not materialized in the

conference rooms at either agency.

And just so I'm provocative and

It's also interesting to me that the sentence

that reads, “only alternatives that are practical in the

business situation faced by the merging parties.”

phrase has not been quoted by any of the cases decided

since 1997 that have dealt with efficiencies.

don't discuss it, and there is no clear allocation of

who's got the burden.

That

The courts

In other words, if the merging parties can

demonstrate likely efficiencies, shouldn't the agency

have to say there exists a practical alternative, and

it's X, and actually produce some evidence about a

practical alternative?

But that's not what's done today.

Now where did we get this notion of a less

restrictive alternative?

It's imported from joint

venture law where we're trying to figure out what

restraints are reasonable.

But the courts in joint

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

venture law, in a decision that I like very much, the

American Motor Inns case, have really discredited the

notion of trying to find the least restrictive

alternative, and I go through that in the paper.

I think that the problem with the way merger

specificity gets interpreted today in the conference room

of the agencies is that it really does devolve into,

well, that's just not merger-specific.

of-the-hand kind of statement rather than a real debate

over any practical alternative that someone would

propose.

And it's a back-

Let me turn to, I think, what is the typical

back-of-the-hand speculation, which is why don’t you do a

production joint venture?

panelists that have worked in the private sector with

parties that have done 50-50 joint ventures could

probably tell you, maybe but for the attorney-client

privilege, about the horror stories of trying to advise

clients on 50-50 joint ventures.

highlight what I think in general led to the skepticism

of them.

And probably any of the

And I'll just try to

I think the first is, typically joint ventures

that are set up 50-50 between two competitors have the

problem of being orphans.

of the management of the joint venture, the joint venture

Because there's a 50-50 split

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

is really an orphan.

It's not really owned or dominated

by one firm.

experience, they have these large meetings that really

become management-by-committee, rather than a straight

linear model.

conceive of, well, they could come up with all kinds of

tie-breaking mechanisms, but in the real world, that's an

issue.

They wind up having, at least in my

That may seem trivial, and we might all

The second is, I think, the formation of 50-50

joint ventures takes two, three, four times what it takes

to put together a merger.

simple, very direct, very linear.

I know there are complications.

target actually takes over the acquirer in the long run,

but in general, they don't suffer from the lack of a

pyramidal structure.

of the parties to the joint venture are over time going

to have to contribute intellectual capital and real

capital to the joint venture.

reluctance to do that if it's a 50-50 joint venture.

Now I'm mindful that you could have a 60-40

Mergers in general are very

I own you.

That's it.

I know sometimes the

And third, in a joint venture, both

And there tends to be a

joint venture and other ways that might somehow get

around this, but I think that those are real world

issues.

agencies, oftentimes the time difference between the

Sitting in the conference room at one of the

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

merger that you have today, the bird in the hand, versus

two in the bush, can be quite extreme.

Now one thing that I will fault the merging

parties and their lawyers for is, I don't think we make

good enough use of the Guidelines' 1997 language about

timing, and we've got that on the next slide.

Just the next page, David.

(Slide.)

Footnote 35 says "If a merger affects not

whether but only when an efficiency would be achieved,

only the timing advantage is a merger-specific

efficiency."

I think most of the time, practitioners and the

staffs at the agencies trying to build a case only see

the word "only" and they kind of blow by the fact that

the 1997 revision at least acknowledges the very

important element of time, and I think it relates to a

lot of what Joe Simons was saying earlier.

There are huge timing advantages, maybe of two,

three and four years, and those time advantages can add

up to in some mergers tens of millions or hundreds of

millions of dollars in difference between cost savings

between different practical alternatives.

I'll also credit the current Commission for the

Novazyme decision, which is a recent decision where

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Chairman Muris in his statement made a very good and

reasoned decision of LRA analysis, Less Restrictive

Alternative analysis, and stressed the efficiencies that

were the bird in the hand rather than going for two in

the bush.

I would like to talk briefly about some of the

decisional case law, only some of which I have the scars

from.

talk a minute about Staples.

illustration of a really flat out war between the merging

parties and the agency with the judge giving us the

benefit of his courtroom for a week.

wonderful thing.

I will, simply because Chairman Pitofsky is here,

Staples is a good

It was really a

And Staples probably had one of the biggest

battles over efficiency.

I think there really were

serious efficiencies, and the staffs that I've talked to

over the last seven years -- I can't believe it's been

seven years since Staples -- have acknowledged that the

efficiency arguments in Staples were very serious, very

compelling and kept people up late at night.

fact, the 1997 Guidelines came out during some of the

agency consideration of Staples.

And in

One thing I would point out, and again, maybe

the fault comes back to me for not focusing on footnote

35, one of the arguments that appears in Judge Hogan's

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

opinion is, well, Staples and Office Depot are going to

continue to grow organically, and the agencies told us in

their deliberations, well, Staples will double over the

next seven years.

But the argument we made, but we didn't make it

forcefully enough for Judge Hogan, is, yeah, but you get

the merger efficiencies today.

seven years.

to very, very large numbers.

probably should have beaten more on footnote 35.

we were just shocked that Guidelines that came out in

April would be used by a judge in June.

of the learning experiences of Staples.

You don't have to wait

And seven years times these numbers add up

And it's interesting.

I think

But that was one

Staples is an important decision on

efficiencies simply because it really, I think, creates

the modern era where efficiencies are serious.

that's one of the good things from the 1997 Guidelines,

is the agencies are really taking on efficiencies and

saying, at least in the text of the Guidelines, they're

important, and certain efficiencies we'll credit.

My criticism is that we're putting the

I think

efficiencies through too much of an obstacle course and

not comparing the likely post-merger world with the

likely world without the merger.

Tenet Health is another case where the case was

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

reversed on efficiencies.

about in the paper, and I'll let you read the paper, is I

have a brief discussion of the Heinz baby food decision.

I think it's quite possible that this case ultimately

will live in antitrust infamy, maybe much more so than

any other case that I've seen.

The other case that I talk

The production efficiencies were quite

considerable.

The competitive effects, I think, might

have been very questionable.

I will hasten to add, I

reviewed none of the evidence.

I wasn't the decision

maker, so I won't put myself in the shoes of the decision

maker.

seems that the Circuit opinion is very cavalier about the

efficiencies.

efficiency improvement in total variable manufacturing

cost, which would be stunning, is considered a

triviality.

notion that you could have a 2,200 basis point

improvement in manufacturing efficiency, and that that

would just be cast aside as, well, that's not that

significant, or that's not merger-specific, you can do

something else is stunning.

But just analyzing the D.C. Circuit's opinion, it

And, for instance, a 22.3 percent

I think, to any business executive, the

There's another passage on distribution and

logistics where the point is made by the court, well,

Heinz has very efficient distribution.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Why can't Beech-

Nut have very efficient distribution?

evidence produced by the opponents of the merger as to

what practical steps could have been done Beech-Nut to

improve the efficiency.

have been on the staff and on the Commission to do that.

But there's no

And I think that the onus should

And there have been decisions that have been

very favorable to what I call merger-caused efficiencies.

I'll let you read those in the paper.

which predates the 1997 Guidelines revisions.

Butterworth, and a 1990 case with the DOJ, the Country

Lake Foods case.

University Health,

In those decisions, what I liked from those

decisions is that the court really seems to get down to

the bottom line, just the effect of the merger versus the

nonmerger world, rather than using hypothetical

alternatives in a less restrictive alternative way, which

is, I think, what the phrase "merger-specific" has

devolved into.

Now where do we go from here?

It seems to me

that where we are today is that we really are doing less

restrictive alternative analysis even though we say that

we're not.

sometimes work that if somebody postulates a theoretical

alternative, that's really the end of the discussion, or

at least the end of the discussion that you experience.

Really, in the conference rooms, it can

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

A couple of practical things that I would do.

First, I think that I would put the onus on the staff to

come up with practical alternatives.

investment bankers out there.

business gurus out there.

solid, practical alternative, let's get testimony about

it.

That has not occurred in any merger case that I've been

involved in, and certainly not in any merger trial that

I've seen.

There are

There are management and

If there really is such a

Let's test that hypothesis with actual evidence.

I think second, ultimately we are going to in

the next revision of the Guidelines or the next step in

the evolution of considering efficiencies, I think we'll

move more away from whether or not an efficiency is

merger-intrinsic, and we'll really look at whether the

merger causes the efficiency, and we'll especially look

for whether that merger-caused efficiency feeds back into

competitive effects.

The net assessment of the merger comes in

Section two, and those efficiencies that directly improve

competitives and rivalry in the industry should already

be considered today in the Section two analysis.

Let me just conclude with a couple of where we

go from here observations.

I think the first is, we

private practitioners who are on the outside, I think

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

need to push efficiencies.

taken criticism that the treatment in the conference

rooms of efficiencies has led attorneys to focus much

less on the real world efficiencies and synergies and

styles of management.

I think it's a very well

I think a second observation I would have is,

that our economy continues to move to a service and

virtual economy.

like Dell, really are virtual companies.

and really excel at production logistics and distribution

rather than physical manufacture of products.

think that has certain implications for merger-

specificity.

Even the companies that sell products

They acquire

And I

For instance, the 1997 Guidelines take a dim

view of certain logistics and purchasing synergies.

think that's unwarranted in today's economy.

economy that's dominated by firms like Wal-Mart, I think

the Wal-Marts of the world need to be challenged by

smaller firms that may actually get themselves larger by

combining.

In an

And so those are some of my observations, and I

leave to you my paper and Voltaire and Spinoza.

MR. ABBOTT:

Okay.

Thank you for that literary

and philosophical set of insights, Mark.

going from literary and philosophical approach to a very,

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

And now we're

very meaty paper by Ilene Gotts.

her slides.

number of issues raised by Section four of the

guidelines.

Efficiencies in Integrated Merger Analysis."

although there's a lot to be covered, as Ilene was

reminding me, she's a New Yorker, and she can talk fast.

Ilene?

slides.

here.

And she's going to give us insight on a

Her slides are entitled "The Role of

MS. GOTTS:

I've been going through

And

And I'm going to trust Dave with my

We're going to test his technical capabilities

I will handle this very fast.

And there's also

a paper that I worked on.

One of the advantages of the

business downturn last year was it allowed me to take

some time to read some of the legal, economic and

business literature that was out there to get a sense on

how the rest of the world is viewing efficiencies and

then I compared it a little bit to the Merger Guidelines.

The Merger Guidelines state that certain types

of efficiencies are more likely to be cognizable and

substantial than others.

towards production efficiencies, that these are thought

to be cognizable and substantial.

There's a real clear bias

Innovation efficiencies are thought to be

substantial, but less verifiable.

And when you get into

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

things like procurement, management and capital cost

efficiencies, these are thought to be less likely to be

merger-specific or substantial, which as you'll see as I

go through this, is almost directly opposite the business

literature that's out there.

When you look at why deals are announced,

they're very much in this last category of procurement,

management and capital cost efficiencies.

some of the presumptions that we build in throughout to

be skeptical about these efficiencies seem to be contrary

to the obligations that directors and management have

under Sarbanes-Oxley.

careful about what they publicly announce as

efficiencies, because if they're wrong at the end of the

day, they lose their jobs.

They have the other liability and problems.

And indeed,

If anything, they have to be very

They have shareholder suits.

So, you know, that's just the backdrop from

which I come from as I look at this.

What kind of efficiencies should be recognized?

Although the agencies might have more experience dealing

with certain types of efficiencies, such as when we look

at productive efficiencies, other types of efficiencies

should not be excluded or handicapped on a generic basis.

And I think some of what we've seen develop in

our understanding has actually come to fruition when you

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

look at the EU Guidelines, which don't seem to draw as

much of a distinction -- we'll probably hear a little bit

more about this from Vincent -- in that it says that you

consider any substantiated efficiency claim in the

overall analysis of the merger.

Productive efficiencies, as we said, are the

least controversial.

One aspect that remains

controversial is, I think it should include fixed costs,

not just variable cost.

especially when you start talking about distressed

industries where the whole industry is not producing

well, these efficiencies are very important.

have an effect by just increasing cashflow, making it

more possible to get significant nonprice benefits to

consumers, such as to fund innovation.

Because at some point in time,

They can

And in marketplaces that are changing, either

they're distressed and you need to realize it, or where

market definition is changing, convergence markets, high

technology markets, these fixed cost savings are just

critical to the bottom line and to whether or not the

great new products of the future will be realized.

Also, distribution and promotional

efficiencies.

The 1997 revisions are silent on these.

The global staff report, which is 1995 -- which is

wonderful to read, by the way.

It is just full of

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

information -- unfortunately find that these types of

efficiencies are less likely to be substantial and often

likely to be difficult to assess.

Just because distribution and promotional

efficiencies are less likely to be substantial and might

be hard to assess doesn't mean that they don't count.

should try to, wherever we can, to factor it in.

Dave mentioned that when you look at the

quirkier remedies -- I think that would be the way I

would phrase what you were saying, and hopefully I'm not

mischaracterizing it -- when they carve out various

things that are novel, that's because there's been an

effort to save efficiencies.

I would make one other note about when I look

at remedies.

what I've seen from Exxon/Mobil or even in the General

Mills/Pillsbury thing, I have found that the staff is

thinking about distributional efficiencies in what

they're demanding I'd make sure is in the divestiture

package.

in that case they think it's for real, why isn't it for

real when as a merger party I assert it?

should be consistent in the approach we take and we

should recognize these.

When I've been in here on Nestle/Dryers or

So if it's something that I have to put in, and

To me, we

Dynamic or innovative efficiencies.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

The 1997

revisions indicate that claims relating to research and

development are potentially substantial but are generally

less susceptible to verification and may be the result of

anticompetitive output reductions.

also acknowledges that they may be -- innovation

efficiencies may make a particularly powerful

contribution to competitive dynamics.

The global report

At the same time, when you come in, you get hit

over the head by the staff saying verify them.

consider dynamic efficiencies in the integrated merger

analysis, as Joe has noted in his talk, he's tried to

factor it in.

talking about mathematical precision in any of what we

do.

a range of assumptions, and in your gut get a sense

whether they're for real or not, whether they're

powerful, and whether they're something we should

therefore use to allow the deal to go through.

These are harder to quantify.

We should

We're not

But you can build on whatever assumptions you want,

Transactional efficiencies, the elimination of

the middleman and the double marginalization.

These are

important.

then can flow through and allow for there to be new

products to be offered.

definitely be included in the integrated merger analysis.

And then my favorite area, which is the

These are really bottom line numbers which

And, therefore, they should

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

procurement management and capital costs.

are very, very resistant to recognizing these.

in every deal I've had just about, procurement savings,

reducing the number of suppliers, going to best

practices, coming up with a way to reduce costs are

always a factor for why businesspeople are doing a deal.

Managerial savings.

1997 revisions

And yet,

The merger specificity

part of this.

The reality is, it is very hard when you

have an entrenched management because of not only the

fact that they're there, but because of labor law,

because of tax law, to just say you're out.

provides a perfect excuse for doing what you might

otherwise want to do.

it and say we're not going to recognize it.

Capital cost savings.

A merger

And, therefore, we shouldn't tilt

The G.E./Honeywell

decision from what I can see in the EU seems to view this

as a negative.

huge difference if you can somehow improve the balance

sheet a larger amount and borrow money at a lesser

amount, if you've got an industry that's going through

change or is distressed, these are very important to the

ongoing competitiveness of the company and should not be

at all thrown out as not recognized.

The reality is, again, this can make a

This gets me then to what burden of proof

should be imposed upon the parties.

Let's look quickly

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

at what is currently the case.

Merger Guidelines and the case law, you see efficiency

claims will not be considered if they are vague or

speculative or otherwise cannot be verified by reasonable

means.

When you look at the

You see statements in there about a sliding

scale that Dave was mentioning.

concentration or the greater the concerns of more burden

that's put on the parties.

The greater the

As Joe noted, and this is very important, we're

making guesses as well for market definitions and other

things.

guess or put a probability on one side it becomes the

case.

the burden shifts.

I don't quite understand why once you make a

It's per se.

It's presumptive.

And then all of

When we look at concentration analysis, there's

a lot now in the literature to suggest that there aren't

these bright lines in the rules.

industry as far as when the concentration is too much and

how competitive it's really operating.

huge presumptions on parties and tilting the sliding

scale, changing the playing field on the basis on these

sorts of things, suggests a precision with respect to the

anticompetitive effects that is unfair and can lead to

really bad Type II decisions.

It depends on the

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

So again, putting

The case law.

Heinz is going to be great.

might have lost the battle, but we might win the war

because of that case.

understand it, Heinz now is pretty much -- one of the

companies is not in the business.

effect we thought of didn't happen because, you know, of

the fact that the merger didn't go through.

company doesn't exist anymore.

a huge burden and thrown out of consideration huge,

substantial efficiencies just was wrong.

That is a bad case.

As I

The anticompetitive

But, the

And to have imposed such

When we look at the EU Guidelines, again, we

see some learning in many areas, but we still see this

tendency to say that efficiencies must be substantial

enough to counteract a merger's potential harm.

again, this suggests that there's some precision

mathematically that I don't think exists.

model goes a long way in helping us to start thinking

about how you look at this, and especially in building in

the probabilities, but the math is not there.

And

I think Joe's

So I would really suggest as a result that we

really step back and we really stop putting in so many

presumptions and tilting the balance so strongly and

maybe look at this like we do in other areas, like

Section I.

has to show there's some possibility of an

Initially you start out where the plaintiff

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

anticompetitive harm.

there's some legitimate purpose for it, i.e., like

efficiencies, and then it shifts back to the plaintiff

again to show why that's not the case.

Per se rules and presumptions built in where

they can't come back is just really wrong.

Then the parties have to show

To save some time, I would like to skip up to

slide 12, if you wouldn't mind, Dave.

(Slide.)

And basically remind everyone that when we're

talking about Clayton Section 7, we're talking about

probabilities, not possibilities, and that's really what

the standard should be here, and that when we look at

this, the other problem I have is, I find the agencies

are wonderful.

about efficiencies, and in close cases, I think they go

the right way.

shouldn't block.

You can come in and you can talk to them

Their gut says, okay, this is a deal we

But when they go to court, it's like all bars

are off.

The agency all of a sudden goes back to Brown

Shoe and Vons or anything else they can dig up with these

horrible presumptions to concentration, because they're

out to win the case.

And when you look at what the role of an agency

is, I love this seminal case, the Berger v. United States

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Government.

The government is supposed to govern

impartially.

Criminal prosecution is not that it shall

win a case but that justice shall be done.

justice is not done by going to a bunch of legalistic

presumptions and tilting the case so that Type II errors

are made, but by applying the same standards at the

agency and then going forth to the court with the exact

same standards and letting a judge weighing the evidence

and decide where the probabilities really stand.

And to me,

When we weigh efficiencies effects in declining

industries, this is what I'm up at night thinking about

these days, because I really do want to try to get it

right.

today that seem to recognize that declining industries,

these are industries where the price is really below

average total cost.

covering the variable costs, but no one is investing in

the future.

really declining and dead for a long time.

To me, there are certain things in the Guidelines

So we're basically -- price might be

This is an industry that is going to be

In my heart, there's some need in that sort of

market to look at being more receptive to arguments about

potential dynamic or innovation efficiencies, to really

be more willing to accept the fixed cost sort of

arguments, because the end of it all is, yeah, maybe in

the short term you might be limiting some competition

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

that is there, but in the long term, the potential harm

to the industry from ignoring two efficiencies and

whether consumers actually benefit consumers from

efficiencies being realized by new products being

transitioned is really great.

So I would really like to see something more

explicit in the Guidelines to recognize declining

industries and how we might look at it.

stage where I even feel I could draft what that should

be, but something should be done on this.

I'm not at the

In conclusion, I think the Guidelines should

clarify that the competition authorities will consider

all types of efficiencies as long as they are verifiable,

substantial and likely to be realized.

this idea that some count more than others.

We should stop

Efficiencies should be subject to the same

standard of proof, and that should be as clear as

evidence relating to a likelihood of anticompetitive

effects, both during the agency review and then in the

court challenge.

merger in a failing industry that we'll be even more

receptive and we'll give more weight to potential dynamic

or innovation efficiencies that could help to sustain the

industry or to transition it into the new products and

services that it needs to offer.

Then finally, that when considering a

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

I stayed within my time.

MR. ABBOTT:

Thank you, Ilene.

You certainly

did, and very provocative remarks.

And you hit a lot of

topics.

I know has given speeches and written on efficiencies,

and I'm particularly interested in hearing what he might

have to say about how lessening of competition in one

market might be justified by efficiencies in other

markets, sort of another aspect of Section IV of the

guidelines that perhaps has not received all that much

attention, at least until Bill started to emphasize it.

And now I'm pleased to turn to Bill Kolasky, who

Bill?

MR. KOLASKY:

Thank you very much, Alden.

I'm

going to focus on three main issues.

what is a cognizable efficiency?

extent should "fixed cost" -- and I put that in quotation

marks -- savings that do not reduce marginal

taken into account?

one that Alden alluded to, and that is, to what extent

should efficiencies in other markets be found to justify

a merger that reduces competition in a particular market?

The first one is

The second is, to what

costs be

And then the final subject is the

In my comments, I want to make it clear that

I'm not offering fully formed views, much less

recommendations.

discussion and debate.

What I'm trying to do is stimulate some

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

I'm going to start out by turning the criticism

not on the agencies but on us practitioners.

year and a half at the Justice Department, I think it's

fair to say that there is more serious and critical

thinking going on within the agencies about the role of

efficiencies in merger analysis than there is in the bar

at large.

Based on my

I think one of the problems is that the

agencies do not hear enough well thought out and well

presented efficiencies arguments.

they can be criticized for not taking efficiencies

sufficiently in account in their merger review, the fault

lies in us, the practitioner, not in the agencies.

I'm hoping my remarks will start us down the road of

trying to address that problem.

So to the extent that

And

Focusing first on what is a cognizable

efficiency?

While I think that the basic analytical

framework in the 1997 revisions is sound, as Ilene's

summary of it, I think, reflects, that framework is to

some extent mired in the old economy, smoke stack

industry paradigm, and I think our thinking about

efficiencies is likewise mired in that paradigm.

We try to put them cubbyholes like productive

efficiencies, innovative efficiencies, procurement,

management, purchasing.

As a very good article, in fact,

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

one of the best articles on efficiencies that I think has

appeared, by Joe Farrell and Carl Shapiro in the

Antitrust Law Journal shows, the most important

efficiencies are not production cost savings and other

forms of cost savings, but rather the synergies that come

from combining complementary assets, complementary

strengths.

To some extent, you may be able to quantify those

efficiencies, but often you are not able to quantify

them, and you have to evaluate them more qualitatively.

Let me take four or five cases that I've been

involved in to illustrate what I mean and why I think we

need to change our thinking about this.

The first one, interestingly, was a case way

back in 1983, shortly after Bill Baxter took over as head

of the Antitrust Division, involving two companies that

make rechargeable nickel cadmium batteries.

Interestingly, the market structure in that case was

virtually identical to the market structure in the baby

food merger case.

with about a 65 percent share.

with about a 20 percent share, and you had a new entrant

with maybe between a 3 and 5 percent share, which was a

foreign manufacturer, and then you had a declining, you

know, fourth company that clearly was not going to be in

You had one dominant supplier, G.E.,

You had a second company

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

the business very long.

We hired George Stigler to work with us on this

merger, and with Stigler's help, we persuaded Bill Baxter

to allow a merger of the foreign producer, and it was the

leading producer in the world of these rechargeable

batteries, to merge with the number two company in the

United States.

The efficiencies that we saw from that deal had little to

do with production cost savings, but were simply that the

foreign producer had the best technology and this number

two company in the United States had a well established

distribution network in the United States, so that the

combined firm would be able to go head-to-head with the

leading producer, G.E., much more effectively than either

firm could do separately.

A second case, interestingly the same vintage,

1979, was the Ford/Mazda merger.

This was a case where

Ford bought a 35 percent interest in the Japanese company

that makes Mazda.

Don Turner and Oliver Williamson.

The reason for that merger was that Ford was moving

toward front wheel drive cars, had very little experience

making those cars, and it was acquiring the transaxle for

those vehicles from Mazda.

about the possibility of opportunistic behavior on

Working with us on that merger were

And it was very concerned

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Mazda's part.

So they wanted an equity stake in Mazda to

protect themselves from that opportunistic behavior.

In addition, that was an era when the Japanese

car manufacturers had more modern management techniques

than the U.S. companies, and Ford believed, and I think

correctly, that the only way the Japanese manufacturers

would share that knowledge with them is if they were

going to benefit from sharing that knowledge because of

this equity relationship.

that are easy to quantify, but the illustrate the

importance of synergies from combining complementary

assets.

Again, these are not things

Fast forwarding to the late 1990s, a merger

that was reviewed by the Justice Department under Joel

Klein and was approved was the merger of the Fox and ABC

affiliates in Columbus, Ohio.

merger.

a four to three merger in the television industry.

did they allow this merger?

reasons was that we were able to persuade them that it

would provide an incentive to the companies to do more

complementary programming.

Survivor type shows going head to head, competing for the

same audience, which is what they would do as long as

they were independent, if they were commonly owned, they

This was a four to three

The Justice Department had never before allowed

Why

Well, one of the principal

Rather than having two

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

would have one Survivor type show and a new show, and

together, that would actually increase output and allow

them to sell more advertising.

the incentive to do that kind of complementary program

unless they were commonly owned.

But they wouldn't have

And in the final case which came out the other

way is the Direct TV/Echo Star merger which was reviewed

by Justice about a year ago.

argument that was being made by the parties had nothing

to do with cost savings.

these two direct broadcast satellite companies would

allow them to offer local into local program, which

neither of them would have sufficient channel capacity to

do independently.

There too the efficiency

It was that the combination of

The reason the Justice Department rejected that

efficiency was because they ultimately found that it was

not merger-specific, that because of advances in

technology, there would be sufficient channel capacity

within the next two years to offer local into local

program without the merger, and that is in fact what has

happened.

The Heinz/Beech-Nut case, I agree with all the

negative things that have been said about that decision,

is probably going to become the Procter & Gamble/Clorox

of our generation.

And it's a good case that illustrates

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

the importance of integrating efficiencies into your

competitive effects analysis.

At bottom, the reason the D.C. Circuit found that

merger unlawful was because they viewed it as a three to

two merger that was likely to lead to a greater

coordination and higher prices.

significant efficiencies, and again, there weren't just

the production cost savings that Mark referred to, but

there was also the importance of being able to combine

the strong Beech-Nut brand with the lower Heinz

production cost in order to compete more effectively

against the dominant Gerber.

They pushed aside very

The facts of that case in fact show that you

already had coordinated pricing going on, but there was

the leader-follower pricing, and Gerber basically was

able to price at a high level under the umbrella created

by the fact that its rivals were so much higher cost

producers.

have costs comparable to Gerber, there was every reason

to believe that even if they continued to coordinate as

they had been, prices would be lower, and consumers would

be better off.

that issue, or at least if they did, they didn't write

about it.

(AUDIO GAP) merger, because the rival would

The D.C. Circuit never even thought about

Let me move on quickly, because I have a

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

limited amount of time, to my second issue.

what I've illustrated by this is that we need to get away

from thinking about efficiencies just in terms of cost

savings.

combining complementary assets and whether that is going

to allow the merged firm to produce a higher quality

product and/or increase its output.

thinking just in terms of price and cost.

But I hope

Think about them principally in terms of

Let's get away from

The second issue is how much weight should be

given to fixed-cost savings.

And here the text that I

begin with is footnote 37 to the Merger Guidelines, which

provides that while savings in marginal costs will

receive the greatest weight, the agencies will also,

quote, "consider the effects of cognizable efficiencies

that have no short-term direct effect on prices in the

relevant market,” such as fixed cost savings.

Again, the way that footnote is written, and that entire

section of the Guidelines, suffers from this problem of

the old smoke stack industry paradigm and sort of a

static price theory model.

that model to a more dynamic view of competitive forces,

you recognize that in fact cost savings and other

efficiencies that do not reduce marginal costs

nevertheless are likely to benefit consumers not just in

the long term but even in the medium and short term.

As soon as you move away from

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

A good book, I think, on this subject is

William Baumol's book, The Free Market Innovation

Machine.

tried to extend this argument in an article that's going

to be published shortly in the Antitrust Bulletin

entitled "What is Competition? A Comparison of U.S. and

European Perspectives."

basically building on the intuitions of Joseph Schumpeter

is that in most new economy industries, competitive

behavior is driven primarily by innovation, not by cost.

And where that's the case, firms need to incur basically

recurring R&D expenditures.

improving and innovating in order to remain competitive.

That creates what Baumol calls a "Red Queen Game" in

which the firm needs to run as fast as it can just to

stand still.

And based on the arguments in that book, I've

And what that all shows

They need to invest in

In these markets, not surprisingly, you find

very high price/cost margins, and that's because the

firms are investing heavily, and they need to be able to

charge prices above marginal costs in order to recoup

that investment, in order to justify it.

Nevertheless, these are industries in which innovation

is rapid, and in fact price is often falling over time,

even as price cost margins remain high.

perhaps not surprisingly, since he created it along with

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Baumol suggests,

Bobby Willig, that the model, the economic model that

should be used for evaluating competition in these

industries is not the traditional price theory model, but

rather the contestability model.

basically is that what determines price in these

industries is the level at which prices will attract new

entry, and firms, even though they charge prices that are

above marginal costs, nevertheless do not have any market

power because their prices are constrained by the threat

of entry, and because they are earning only a minimal

return on capital, i.e., on their investment.

Because what he argues

In these industries, obviously, efficiencies

that help to reduce the cost of R&D and other recurring

common costs will help to drive down price over time and

lead to greater innovation and improved products, i.e.,

dynamic efficiencies, and those need to be taken into

account.

Finally, and I apologize for rushing, but there

is a lot to cover here, and I want to make sure to save

time for the discussion, to what extent can a lessening

of competition in one market be justified by efficiencies

in other markets?

Joe, I think, referred to the fact that back in

1982, everyone was skeptical as to whether the

hypothetical monopolist approach to market definition

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

would work.

Obviously, we've gained a great deal of

experience with that approach over the last 20 years. I

think as a practitioner, that one of the things that has

been most striking to me over the last five to seven

years is that as we have applied the hypothetical

monopolist market definition test more rigorously,

markets have come to be defined more and more narrowly.

Second, as we've moved away from homogeneous

commodity-type products, more and more markets are

characterized by price discrimination.

characterized by price discrimination, you could, as a

theoretical matter, define virtually each customer as a

separate market.

at least, when they analyze mergers, that is in fact how

they look at them as they try to evaluate the likely

price effects of a merger.

And in markets

And in fact, at the Justice Department

They may not define the market that way when

they go to court.

I think that's Ilene's point.

But

even then, the market definitions are much narrower today

than they were a generation ago.

the Justice Department's complaint in the First

Data/Concord merger case where they defined the relevant

market as "processing PIN-debit cards at point-of-sale."

And certainly in the similarly narrow market definition

in the merger case where Mark was successful in beating

Look, for example, at

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

back a challenge, because the judge simply didn't

understand what role price discrimination plays in

defining the relevant market.

As we move to increasingly narrow market

definitions, it becomes more and more important that the

agencies take into account out-of-market efficiencies.

Andrew Dick and I have written an article that appeared

this past year in the Antitrust Law Journal and we

describe a particular merger case that was reviewed by

the FTC while Bob was there, which I think is a perfect

illustration of how important this is.

involving two natural gas gathering and processing

systems in West Texas, right next door to George Bush's

hometown of Midland.

production -- the fields were very mature.

was declining.

less than 50 percent of capacity, as a result of which

unit costs were rising sharply.

wells in this area that were close enough to both

gathering systems to benefit from competition between

them.

processing plants in these two counties.

one sense a merger-to- monopoly.

staff understood that there were literally only a handful

of wells, maybe a dozen or so, that benefitted from the

That was a merger

That was an area in which

Production

The processing plants were running at

There were very few

The remaining customers were gathering systems or

So this was in

And yet, when the FTC

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

competition between these two systems, whereas there were

hundreds of wells that were going to benefit from the

cost savings resulting from consolidating the two

systems, they granted early termination of the second

waiting period.

The efficiencies were outside the relevant

market in which the anticompetitive effect would be felt,

and yet nevertheless, they were recognized as swamping

the potential anticompetitive effects, so the merger-to-

monopoly was allowed.

Another very good example of this, but a more

controversial one, is the Surface Transportation Board's

decision in the Union Pacific/Southern Pacific merger

case where I was one of the opponents.

a large number of routes from the Gulf Coast to the

Chicago-St. Louis area that were basically two -- or the

merger was a two to one merger.

monopoly.

convinced, however, that there were very substantial

efficiencies from combining the networks of the Southern

Pacific and Union Pacific roads that would benefit all of

their shippers on all of their lines.

order to remedy the anticompetitive problem, rather than

ordering divestitures, they structured a trackage rights

remedy that would give a third carrier the rights to use

There, there were

It was a merger-to-

The Surface Transportation Board was

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

And, therefore, in

the UPSP tracks to service these customers.

And I think

that's a good illustration of how efficiencies factor

into the choice of remedy.

But, again, the point I'm making here is that

that was a case where the agency found that the out-of-

market efficiencies were inextricably intertwined with

the negative competitive effects, and therefore tried to

structure a remedy that would fix the competitive problem

without sacrificing the efficiencies.

I think one of the things that concerns me

about the new EU draft -- EU Notice on Horizontal Mergers

-- it's now final -- is that, at least as I read it, the

European Commission does not have the same flexibility

that our agencies have to take into account those out-of-

market efficiencies.

from Vincent on how the Commission would deal with cases

like the two that I've just discussed.

And I'd be very interested to hear

Thank you.

MR. ABBOTT:

provocative remarks.

overview to Chairman Pitofsky, who was around when

Section IV of the 97 -- Section IV of the Guidelines was

revised in 1997 to give us the current treatment of

efficiencies.

Thank you, Bill.

Some more

And now I turn for a magisterial

Chairman Pitofsky was also around, of course,

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

for the Staples case and the Heinz/Beech-Nut case, which

a number of practitioners here have trashed.

sure that Chairman and Professor and Dean Pitofsky will

give us a magisterial overview and maybe some rebuttal as

well.

And I'm

Bob?

MR. PITOFSKY:

Thank you.

It's good to be back

here in this unfamiliar but familiar FTC.

is unfamiliar.

This building

You know, magisterial oversight is a little

different.

we have three or four speakers on one side saying the

Merger Guidelines should be expanded, two or three

speakers saying they should be left where they are or

contracted, and then I would stake out a middle ground.

I thought what we were going to have is that

That's not it.

Everybody who's spoken so far

has said the Merger Guidelines -- the efficiency defense

in the Merger Guidelines should be expanded, even made

primary.

It's interesting.

Historically, it was Bork,

Posner and Baxter who felt so strongly that the

efficiencies defense was a wrong idea.

think they'd have some influence today, but not on this

issue.

does have efficiency credentials, and shyly, myself, who

wrote that an efficiency defense has to be included in

And you would

And it was Areeda, Turner, Broadly, Muris, who

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

the Merger Guidelines.

And it was the Clinton enforcers

who introduced it formally into the statute.

Let me just say a few general things and then

some specific things about the cases that were brought

up.

that are imposed to successfully assert an efficiency

defense are too narrow.

There is a general thread here that the conditions

Let me start at a different place.

my intention to defend every word and every concept in

these efficiency guidelines.

Ilene's notion that declining industry aspects should be

brought into your consideration of efficiencies.

Absolutely.

that fixed and variable are unnecessarily, treated

unnecessarily differently from the point of view of the

Guidelines, I agree with that.

the kind of distinction that theoretical economic

analysis does.

associate myself with him, that the staff does listen to

efficiency claims.

debunk the claim, that's not the agency I was at.

are there and they work hard on these things.

It is not

I heard a couple of things:

I think that's a good idea.

Bill's thought

Businesspeople don't make

And Bill's suggestion, I just want to

The idea that they're only there to

All right.

Narrow conditions.

It is true, the

conditions are narrow.

The efficiency must be

substantial, merger-specific, timely, clear and

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

They

convincing evidence, can't reduce output, et cetera, et

cetera, but that's what was intended.

in a close case, efficiencies can be a tiebreaker.

idea was not that efficiencies ordinarily will allow you

to merge three to two, or two to one.

American antitrust is about.

statute, which says "tend to create a monopoly" is about.

The idea is that

The

That's not what

That's not what this

On producer surplus, I'm sort of agnostic.

the producer surplus were enormous and the

anticompetitive effect were tiny, I guess anybody would

take it into account as a matter of prosecutorial

discretion.

I'm pretty sure you would do that.

As to some of the proposals, exceptional

effort, thoughtful, creative.

be, Joe Simons and others, can we do it?

very fair point.

definition of relevant market.

to do it, and we can.

My question is going to

And you make a

A lot of people said that about the

And now we've learned how

But I'm worried about this one.

Antitrust is a

practical enterprise.

It's not the development of a

dissertation.

well, let me take as an example, the claim is that, yes,

the merger will lead to higher prices now, but down the

road, three years, four years, five years, the efficiency

will produce lower unit costs and perhaps, maybe even

And I just keep thinking to myself --

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

probably, lower prices.

the subpoena look like that tries to get at the question

of whether or not five years down the road there will be

lower prices, that the efficiencies will kick in.

And I keep wondering, what will

Look at how many times people have claimed

efficiencies and they never kicked in.

going to do with the company five years later when it

turns out that prices are still higher and the

efficiencies haven't kicked in.

company, not if it's a horizontal merger.

practical problems here.

What are you

You can't break up the

So there are

There are also technical problems.

It's

contrary to the statute, which says in any line of

commerce.

and it produces the odd situation where one set of

consumers who buy today are financing lower prices for

another set of consumers who buy later.

concern is, could you possibly do it?

It's contrary to the spirit of the Guidelines,

But my main

Let me raise a question for all of you.

Do you

believe in British Oxygen?

Do you believe the court when

it said that it will not let the plaintiff get away with

coming in and making the anticompetitive case in some

speculative way?

there will be an anticompetitive effect now, but I can

tell you eventually there will be an anticompetitive

Well, judge, I can't tell you that

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

effect, and you ought to strike down the merger.

And

what was it, the 2nd Circuit?

their ear on grounds that you can't come in here with

that kind of speculative talk and expect to block a

merger.

FTC. FTC got thrown out on

I agree with that decision.

plaintiff has a responsibility to show clearly by

convincing evidence that there will be an anticompetitive

effect in a timely manner.

for defenses.

to come in and say, well, judge, I don't know about

efficiencies right now.

up now.

the road, oh, then everybody's going to really be in

great shape.

I think the

But I think the same thing

I don't think that one should be allowed

Actually, prices are going to go

That's what makes the merger illegal.

But down

A little too speculative for me.

Ilene talked about hierarchy of efficiencies

and clearly said that, you know, maybe there are

differences, but you shouldn't disregard certain

efficiencies entirely.

I would disregard certain

efficiencies entirely.

Let me say, the best work on this

by far is about 70 or 80 pages in Areeda, in which goes

through efficiency by efficiency by efficiency and talks

about whether or not they should be taken into account to

the extent that they reverse the result in an otherwise

illegal merger.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Reduce unit cost is the clearest case.

The

Guidelines picked it up.

Generally speaking, you can prove it.

reduced unit costs are there.

about R&D efficiencies.

quarrel, because I don't think Areeda thought so highly

of that.

of the country, the economy and consumers, that R&D

efficiencies should be cognizable.

The writers have picked it up.

You can know the

The Guidelines talked

I remember that as being quite a

But I think R&D is so important to the welfare

But then these other efficiencies are more or

less savaged by Areeda.

promotional efficiencies, capital cost efficiencies.

Let's just take managerial efficiencies.

being run poorly, and you should do something about it.

Well, there are two possibilities.

company that's being run well, or you can fire the bad

manager, and like George Steinbrenner in New York, you

can hire a better third baseman.

to bring in a first rate leader or a consultant to a

second rate leader, that sort of thing.

Managerial efficiencies,

The company is

You can merge with a

You can pay the money

It seems to me that managerial efficiencies are

vague, hard to measure.

How do you measure what a really

good CEO, president, sales manager brings to a company?

What questions do you ask about that?

what a managerial efficiency is.

Hard to define

And the tradeoff

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

between managerial efficiencies and anticompetitive

effect is going to be hard to manage.

up, what are you saying?

will bring the prices down?

speculative and eventuality, which British Oxygen said we

ought not do.

If the prices go

But this manager eventually

Well then we're back to

And finally, there's a less restrictive

alternative.

managers in this country.

to get a capable manager is to merge with your competitor

or your customer or your supplier, something like that.

I don't think there's a shortage of capable

Let's see.

And I don't think the only way

Merger-to-monopoly justified by

efficiency.

I guess I've already said.

statute.

understand that a company that expands internally can

point to its own superior skill, foresight and industry.

But this isn't expansion.

question is whether or not you would allow a merger-to-

monopoly and say, but, judge, we're so efficient.

we'll lower costs.

It's not our Guidelines.

It's not our

It's speculative.

These are mergers.

And the

And

First of all, why would a monopolist do that?

If you're a monopolist and a profit maximizer, it seems

to me you wouldn't lower costs if there are high barriers

to entry.

stay in place, the agencies should take that into

If the Baumol theory is you're running hard to

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

account.

That's fair enough.

Okay.

That's fair enough.

Now, let's see.

Shots at the

Commission.

Look, baby food you say will live in infamy.

It's a tough sell and similar mild comments.

Procter & Gamble of the new generation.

(Laughter.)

Let's put this into context.

It's the

A hundred and

fifteen years, and no court has ever approved a three to

two merger unless there were low barriers to entry and a

failing company assertion.

So it was hardly a departure.

affirmed what the Federal Trade Commission said.

A hundred and fifteen years.

Unanimous Court of Appeals

Now since the deal, I'm not even sure what's

happened, but I gather Gerber has grown and one of them

has gone out of business.

Heinz sold out.

MR. PITOFSKY:

They exited the market.

Okay.

I guess I have to think

about this one.

But, you know, you play the cards you're

dealt with.

two, the two, the two smaller ones were competing with

each other, fairly and vigorously, and it seemed to me

the case was justifiable.

somebody may make me eat these words -- I don't recall

either of the merging parties coming in and saying we're

going out of business.

At the time of the merger it was three to

And I don't recall -- and

They might say we will never be

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

efficient and we'll curtail our activities.

I don't

remember anybody saying they would go out of business.

Now, does that mean the Commission will never

take a three to two, never allow a three to two?

think we've done it a fair number of times.

immediately of Boeing/McDonnell Douglas where we thought

it was two to two, because McDonnell Douglas such a weak

player.

No.

I think

Chairman Muris told me that there have been

several instances.

It's hard sometimes to make public

prosecutorial discretion decisions.

three to two mergers.

is allowing three to two mergers, and efficiencies are

the turning point.

Staples.

(Laughter.)

But there was clear evidence that prices went

up in markets characterized by fewer superstores, and as

to the efficiencies -- don't hold me to any numbers.

This is very general.

board of the company was told when they were thinking

about a merger that the efficiencies would be X, 5X.

Then when they knew they were merging, the board was told

25X.

That hurt.

But we have allowed

He is allowing -- his Commission

Efficiencies are the decisive factor.

It's been so long I forgot.

But my recollection is that the

And then when they went to court, they said 75X.

That hurt in terms of trying the case.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

And the other issue is much more complicated,

and that is whether or not purchasing efficiencies, that

is to say, you're bigger so you get better discounts from

the seller, ought to be taken into account, or is that

just transferring money from one businessperson's pocket

to another?

I grant that that is a very complicated

question.

Let me sum up by simply saying this.

American antitrust is better off by the introduction of

an efficiency defense.

the people who said you put that efficiency defense in

our Guidelines and you will destroy the ability of the

government to enforce the antitrust laws, because people

will come in with phony efficiencies and you won't know

they're phony.

on forever.

worked out that way at all.

I think

I was thinking last night of all

It will delay the matters.

Cases will go

It hasn't worked out that way.

It hasn't

Incidentally, there was a comment that courts

almost never say it's illegal, but because of the

efficiencies, I'll make it -- I'll call it legal.

believe the reason for that is the agency doesn't bring

cases that are barely illegal but with substantial

efficiencies.

shot at this, and I'm not sure they're going to get a

And therefore, the courts haven't had a

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

shot very soon, because the agencies are very sensitive

to claims of efficiency.

I think some excellent deals have gone through

by emphasizing the efficiency.

particularly true in the high tech R&D drug area, and the

country is better off for that.

competition is concerned, the efficiency section put the

spotlight where it belongs -- not on the size of the

competitor, but on the efficiency, its unit cost, the

quality of its product, the quality of its research.

I think for all those reasons, we're better off.

Thank you.

MR. ABBOTT:

I think that's

And as far as global

And

Thank you, Bob, for a very good

overview and putting things in context, as you always do

so well.

Finally we're going to turn to Dr. Vincent

Verouden of the European Commission, DG Comp.

he has honored us with his presence earlier in the

workshop, and he will be focusing on the treatment of

efficiencies in the new European Merger Guidelines.

MR. VEROUDEN:

And again,

I would like to talk about

merger analysis and the role of efficiencies in the

European Union.

Before going to the specific topic of

efficiencies, it may be good to provide some background

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

on what has been going on in recent months or years in

the EC merger control system.

As you may know, this merger control system,

which is in place since 1990, has been the subject of a

review process, a reform process you could say.

has resulted in a new merger regulation which was adopted

last month by the member states of the European Union.

It will be applicable as of the 1st of May, 2004, so in

three months' time.

And this

It has also produced Horizontal Merger

Guidelines which accompany this new merger regulation.

So much what I'm going to tell now or today is relating

to, you know, what will be the system in three months'

time.

There were a couple of issues that kind of

spurred this reform process.

issues, jurisdictional issues, but also substantive

issues.

debated in this reform process were first of all, what is

the scope of the existing dominance test that we have in

the European Union in Article II of the EC Merger

Regulation?

in merger analysis, what should be the proper role of

efficiency considerations in merger analysis?

There were procedural

The two main substantive issues that were being

And the second was the role of efficiencies

Just to say a few words about the test, because

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

that's, of course, the main tool that we have to use in

the coming years, and it's difficult to speak about

efficiencies and integrated analysis without also knowing

what your substantive test for review is.

And it has often been put in terms of the

comparison of the dominance test versus SLC test,

Substantial Lessening of Competition.

difference?

with the dominance test -- by the way, I should read out

what the dominance test is.

creates or strengthens a dominant position as a result of

which effective competition would be significantly

impeded.

dominance test.

Is there a

And the question that arose in our context

It's whether a merger

That's our current test, and it's called the

And the question was, well, does it cover all

mergers that produce anticompetitive effects?

deal with mergers that produce a company with significant

market power, but which only is, for example, number two

in the market, and, therefore, not, let's say, dominant

in the usual meaning of the word?

Does it

Anyway, this resulted after, well, quite a long

debate in a clarification of our test.

It's now called

the SIEC test.

course, and it's also, well, normal that it's pretty

similar.

It's quite similar to the SLC test, of

Namely, the test is now whether a merger leads

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

to a Significant Impediment to Effective Competition.

that's the tool that we will work with in the coming

months -- in the coming years.

I must say, by the way, that the new test that

we have, SIEC, still singles out the notion of a dominant

position as a primary example of a merger that is leading

to a significant impediment to effective competition.

So, well, so far for the background on this new

test.

merger control system.

the current situation, if you like, with the dominance

test, I think it's fair to say that efficiencies have not

received much emphasis in the past.

Let's turn to the role of efficiencies in the EC

If we look at the past or even

That is quite clear.

It's probably outside the

scope of this workshop to really expand on this, and the

reasons why and so on.

thinking that we have gone through in the European Union

quite closely resembles, I think, the development that

you have seen here in the United States over the course

of the years where efficiencies were first viewed with

quite a bit of skepticism, of course, whether they could

overturn conclusions as to the anticompetitive or pro-

competitive effect of a merger.

to having a more open attitude to efficiency

considerations.

I think that development in

So starting from there

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

And -- well, I've written down here what was

the conclusion of this internal reflection, well,

internal and external reflection process.

at some point that, well, there are compelling reasons to

give more explicit consideration to efficiencies in

merger control.

course, that efficiencies may bring more competition to

the market rather than less, if anything.

We really said

And this from the understanding, of

I've written a few reasons, maybe not necessary

to say a lot about that.

Efficiencies are a natural element, although not the

easiest in any competition analysis. If you want to

determine whether the merger will increase or reduce

competition in the market, it's kind of hard to do that

in an accurate or proper way without also looking at

efficiencies or possible efficiencies.

I think it's well understood.

And in this sense it's also in line with a more

effects-based or call it economics-based approach to

merger control.

(Slide.)

Now, the treatment of efficiencies in the new

merger regulation, I think it's good to start out by

saying that there was no need to change our test for the

sole purpose of looking at efficiencies.

the reasons in the small bullet points down there, first

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Actually, given

of all, we were always of the opinion that the existing

test in Article II of the merger regulation, even though

it had, you know, the dominance wording and so on, was a

pure competition test.

And so the moment you say that, yes,

efficiencies must be considered in merger analysis, well,

actually, they should be part of an integrated

competition analysis.

of have a separate sentence in Article II on

efficiencies.

So there was no real need to kind

I must say, by the way, that Article II

comprises a number of paragraphs, and so paragraph two

and three have this test that I just said.

of this article actually says all the factors that we

have to look at when looking at the likely impact of

mergers, and so it refers to entry barriers and buyer

power and so on, but also to the technical and economic

progress that is likely to be achieved through the

merger, provided it is to the consumer's advantage.

that is wording that has been in the test in Article II

since the very beginning.

emphasis, as I said before.

this is a sufficiently worded text for Article II.

Paragraph one

And

It hasn't received much

But we still believe that

And, however, to signal the change in emphasis

a little bit, I think there are two things that I can

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

say.

First, the new wording that we have that actually

focuses on, you know, whether there will be a significant

impediment to effective competition and does not put that

much emphasis anymore on the dominance aspect, probably

better allows for, you know, a proper inclusion of

efficiencies.

It better expresses that our test is an

effects-based competition test so that instead of looking

in an almost static way to whether post-merger the new

entity will have a dominant position, which is a bit of a

snapshot analysis post-merger, the emphasis with the new

wording of the test is more towards what will change from

pre- to post-merger.

somewhat more natural place in the new wording of the

test.

So efficiencies maybe will find a

In addition, we have what we call recitals in

our legislation, and those are kind of sentences that

precede the operational part of the regulation, and they

kind of give the intent of the legislation, and you can

find here in the text that we have put, or that the

member states have put, I must say.

So, in order to determine the impact of a

concentration on competition, it's appropriate to take

account of any substantiated and likely efficiencies put

forward by the undertakings concerned.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

It is possible that the efficiencies brought

about by the concentration counteract the effects on

competition, and, in particular, the potential harm to

consumers that it might otherwise have.

Let me turn to the treatment in the Guidelines.

The regulation, of course, is all at a rather general

level and so on, but the Guidelines, they kind of provide

the analytical approach that the Commission intends to

take in the analysis of individual cases.

As I said, the Guidelines were adopted a few

weeks ago, 30 January, and, well, if there is any

characterization, I guess, of the approach to

efficiencies in these Guidelines, I would say it's open

but cautious.

to that.

invites any substantiated efficiency claim.

Guidelines recognize that in whatever form, efficiencies

may make the new entity more competitive.

bias in that sense.

Open.

I think Ilene referred a little bit

It's open in the sense that it's, you know, it

So the

So there is no

On the other hand, of course, Guidelines must

give some guidance, so I will come to that in the next

slide.

At the same time, the approach is cautious.

that's also kind of natural.

conditions or the circumstances under which the

But the opening statement is open, so to speak.

I think

It's cautious in the

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Commission can take efficiencies into account.

to that on the next slide as well.

I'll come

But the focus lies on the ability and incentive

of the merged entity to act pro-competitively for the

benefit of consumers.

welfare objective, the Guidelines are rather explicit

about what is the relevant benchmark in the assessment of

efficiencies, and for that matter, in the assessment of

possible adverse effects of a merger.

So while this is a consumer

This consumer welfare objective actually

doesn't come falling out of the air.

our Article II that we have which refers to that we

should look at technical and economic progress, provided

that it is to the consumer's advantage.

in the regulation, and it was in the regulation already

10 years ago.

It's derived from

So it's already

It's also consistent, by the way, with other

articles or sections in our antitrust legislation.

example, Article 81, which also has this consumer

objective in the end.

For

I guess it's more for the discussion afterwards

whether the benchmark is the right one, of course, that

you can always discuss.

the Guidelines is that it spells out in a rather visible

way and, also, hopefully clear way what are the three

But a good thing, I guess, about

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

conditions for efficiencies to be taken into account in

order to clear a merger for the Commission.

So the first one -- actually, there are like

headings, and then the things are further developed.

first heading is Benefits to Consumers, and the relevant

benchmark in assessing efficiency claims is that

consumers will not be worse off as a result of the

merger.

it's pretty clear.

The

This sentence is literally in the Guidelines and

In principle, this concerns consumers in the

relevant market.

here, and attentive readers can see that the "in

principle" was not in let's say in our draft notice, our

draft Guidelines.

what this means is that the focus is not entirely and

completely on consumers in the relevant market, but it's

in principle.

You will note that words "in principle"

It is in our final Guidelines.

I think this is also part of our open approach,

given that we have not spent so much emphasis on

efficiencies in the past, I think it's better to leave

things open while giving clear directions, but leave

things open.

in the Guidelines.

I think this is more or less the exercise

Efficiencies may take various forms.

savings leading to lower prices or synergies leading to

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Cost

new or improved products.

that variable costs are more likely to be relevant than

fixed costs, but this does not exclude anything.

that sense, it's open as well.

We say in terms of guidance,

So in

Then there are two things which are very

familiar to you.

expand a lot on this.

specific, which means that they are relevant when they

are a direct consequence of the notified merger.

must be a causation there.

There is maybe not much reason to

Efficiencies must be merger-

So it

And on, you know, the set of alternatives, so

to speak, that you look at, you know, are there less

restrictive alternatives and so on.

think what we did try to do in our Guidelines is to tie

very much the relevant comparison with, okay, this is

what the merger will produce and how would it likely be

in the relevant counter factual, so in the absence of the

merger?

two situations to compare.

constrained in that respect.

On that debate, I

So that in theory should give two things to -It's not -- it's a bit

Finally, verifiability.

Well, since most of

the information is in the hands of the merging parties,

when we speak of efficiencies, I mean, it's also

incumbent upon them to provide relevant information

demonstrating that the efficiencies are merger-specific

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

and likely to be realized, and it's also upon them to

give a first shot or show to what extent consumers will

benefit.

I think I will just leave it here for the

moment and leave other points or questions to the

discussion afterwards.

MR. ABBOTT:

Thank you, Vincent, for that very

good overview and introduction for all of us to the new

European Guidelines.

I'd like to take a brief break to allow -- no

longer than 10 minutes.

We will be starting up again in

10 minutes time exactly.

So if you're late, we'll

already be going.

Thank you.

(A brief recess was taken.)

MR. ABBOTT:

If people could try and take their

seats, please.

and answer session to start out with, Mary Coleman.

Let me turn things over for the question

MS. COLEMAN:

Thank you, Alden.

I'll start off

asking a question, and I'm going to direct it towards Joe

but then more towards the panel generally and in part

working a bit off of what Chairman Pitofsky had said in

terms of this integrated approach and this sort of

formulaic approach.

You had said that, you know, it may be

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

difficult to have mathematical precision in doing this,

but it's sort of an approach that could get people

thinking about weighing the different elements of the

case.

back-of-the-envelope sense, could the agency do this kind

of approach or the parties do this sort of approach?

do you come up with probabilities?

with reasonable discount rates to actually try and get

some numbers?

But to try and get to how practical, even in a

MR. SIMONS:

How

How do you come up

Well, I think in actual practice,

I think companies are doing this already, and the staff

is actually doing it already, but it's being done

subconsciously.

efficiency claims of one of the parties, that's what

you're doing.

the chances, how much do I believe this?

what happens, or I think probably what happens mostly is

it's kind of a -- people have in their minds like a

binary type of decision or a binary way of thinking.

It's either on or off.

reality it's something, you know, largely in-between.

Anytime you evaluate the claims, the

You are saying to yourself, okay, what are

And sometimes

It's either yes or no, when in

And so, I think, you know, subconsciously, this

is being done already and the only thing I'm really

suggesting we do is to kind of think about it more

consciously, more explicitly.

And the place where I say

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

this coming in is kind of like in a sensitivity analysis.

You know, you don't have to know exactly what the chances

are that the efficiency is going to be realized or

exactly how big it's going to be.

kind of set the parameters.

analysis and see how much it makes a difference and in

what regard.

But, you know, you can

You can do a sensitivity

And I think people kind of do that

subconsciously already.

And I think if you just put it

down on paper, it just clarifies the thinking.

MR. PITOFSKY:

Can I ask?

MS. COLEMAN:

Sure.

MR. PITOFSKY:

Joe, you're right.

I think that

your spreadsheet approach and the integration approach is

very challenging and it makes a lot of sense.

your answer to my question?

side of the equation and say to the plaintiff, even if

the anticompetitive effects aren't immediate, if the

anticompetitive effects occurred down the road sometime,

that's good enough?

Oxygen which says that's not good enough?

But what's

Would you go to the other

Or would you stick with British

MR. SIMONS:

My interpretation of British

Oxygen would be that the anticompetitive effect there was

speculative.

into the future.

In other words, very, very unlikely and

If what you're talking about is

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

something that is, you know, is going to occur with some

reasonable probability, maybe in two or three years or

four years even, then I think, yeah, you should factor

that into your analysis.

calculation, and if the efficiencies occur earlier than

the anticompetitive effect and there's a price drop,

yeah, earlier, then that gets more weight than the

anticompetitive effect.

you wouldn't consider it.

MR. PITOFSKY:

MR. SIMONS:

It should go into the whole

But I don't see any reason why

On both sides?

On both sides.

I don't see any

reason why it shouldn't be done equally.

MR. SCHEFFMAN:

Let me say, can you do this?

Of course.

sophisticated and talented in what they do.

is the decision making is very ad hoc.

The FTC, both agencies are incredibly

The problem

Now, if you take an agency which has much less

talent, you know, top to bottom, like the EPA, and you

look at the -- and I'm not saying that if you look at

their actual decision making there's hot politics and all

sorts of complicated things involved -- but they're

dealing with weighty tradeoff issues of health versus

costs and other sort of things.

They have formal procedures like this to do it.

are well known procedures.

There

Government agencies do it.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Businesses do it in much less complicated and -- much

more complicated and contentious situations than we're

dealing with in antitrust.

The problem is it's very ad hoc now.

really no common agreement within the antitrust agencies

and within the staff even what the standard is for

efficiencies and how they should be weighted.

There's

So it's true that something's going on in

different people's heads, but knowing makes them

accountable for it.

Joe is proposing, well, here is the particular analysis.

You have to have an agreement.

analysis?

up with programs?

can buy consultants.

and other agencies do that all the time.

So you can't just -- I don't think

Well, what is the

How are things going to work -- can you come

Absolutely.

You can buy programs, you

We'll tell you how to do that.

Is it science?

Is it hard science?

No.

less science than the way we make decisions now?

It's just making -- it makes a more transparent and

rigorous sort of approach.

MR. KOLASKY:

EPA

Is it

No.

If I can just add one comment on

this.

I think one case that's worth following, because

it's directly relevant to this, is the G.E./Honeywell

merger case, which is now pending in the Court of First

Instance in Europe.

Because that, of course, is, in

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

fact, the central issue in that case where the, you know,

Commission's decision found that the merger would make

the combined firm more efficient, lead to lower prices,

but that would ultimately drive out rivals, after which

they would be able to raise prices.

And so it basically is the paradigm example of

this issue, and it's going to be very interesting to see

how the CFI deals with it.

MR. PITOFSKY:

Good point.

MR. KOLASKY:

And I'm sure Vincent will

probably disagree with my characterization about the

reliance on efficiencies, but.

MR. SCHEFFMAN:

Let me pick up first, because I

don't -- I said something which was -- I didn't mean, and

I think was misinterpreted.

The agencies take

efficiencies very seriously.

I don't doubt that.

you look at process-wise what happens, the only clear

guidance staff has is the case law.

But if

The lawyers know they have to be prepared to

show your client you can rebut the efficiencies cases

that the parties will put forward, particularly in the

FTC where you never know whether you have a case for sure

until the Commissioners vote.

seriously because they've been very important in Staples

and Drug Wholesaling, et cetera.

They take it very

And so there's a lot of

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

100

focus on that.

And because of the sliding scale, okay, there's

a lot less focus on, well, is the deal really good or

not, because in reality, it often doesn't count much.

counts a lot in a gray area.

matter in which the legal staff believes they've got a

strong case.

It doesn't count much in a

Even the economists in that case aren't going

to fight much about the efficiencies, because that's not

where the game is.

anticompetitive theory of the lawyers case if you

disagree with it.

still not going to win because of the sliding scale.

The game is whether you can rebut the

If you win on the efficiencies, you're

So that's the problem with the system.

the process, and if you made the process less ad hoc and

provided more guidance to staff -- this is actually how

we're going to weigh things.

we have to think about the fact that none of these things

are certain, and don't send up a memo each time where

this is absolutely the market definition; that there

isn't any doubt about it, and it's just as good as the

one last week we sent up, et cetera.

rigor, reasonable rigor in the process, we'd get better

decision making.

MS. COLEMAN:

That's

This is how you, you know,

If we have more

Vincent, did you want to?

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

101

MR. VEROUDEN:

Yes.

Just to react on Bill's

comment, of course.

really say a lot about the GE/Honeywell case.

pending in front of the court.

was broader than whether, you know, the efficiencies

would allow the new entity to lower price and then as a

result of the efficiencies, the ultimate result would be

a negative one with the other firms exiting.

Well, I'm not here in a position to

It's also

But, certainly, the issue

I think if one could say one thing, the focus

of the investigation was whether the new entity would

have both the ability and the incentive to kind of embark

upon the strategy to marginalize in some way or the other

the rivals in the markets concerned.

broader than just giving it the label “efficiency

offense.”

stage.

is going to tell.

So it's a bit

That's the only thing I want to say at this

And let's just leave the debate for what the CFI

Thank you.

MS. COLEMAN:

Another question that was brought

up by some of what Chairman Pitofsky said and by Joe's

analysis, is the time horizon over which we are looking

at, both effects and efficiencies and the net effect of

the merger and what the panel thinks about what type of

time horizon should we be looking for.

two years, which some parts of the Guidelines sort of

suggest?

Should it be longer?

You know, is it

Should it be different

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

102

for the effects versus efficiencies?

to the panel for people to comment on.

MR. GIDLEY:

And I'd open that

Well, let me field that first.

think, first of all, there is a certain amount of

inherent factual difference between industries and

between deals.

So some deals and some mergers, I think, truly

are permanent in the sense that, you know, you do --

Boeing/McDonnell Douglas, the chance of somebody else

getting into commercial aircraft manufacturing on large-

scale jets is very small.

Douglas maybe was already out by the time of the deal.

On the other hand, McDonnell

So I think it depends a little bit on the

industry.

I think in some service and retailing

industries, I think there's a more dynamic process where,

you know, mail order and internet compete more and more

for dollars.

frame is, it depends, meaning it depends on the industry,

it depends on the product, it depends on the firms and

what's likely down the road.

So I think the first observation about time

I think the second thing, and frankly, I hadn't

focused on it as a practitioner, is, from the standpoint

of time frame on efficiencies, that footnote 35

explicitly says that timing matters, I think, is under-

emphasized by the merging parties in their defenses.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

And

103

so that's an example of good text that's in the 1997

Guidelines, and that again, Chairman Pitofsky, is

something where I credit the Clinton Administration for

improving on the '92 Guidelines.

There is some explicit recognition that timing

should matter.

and CEOs and management see spreadsheets that go out five

to 7 years as just being the normal horizon for

evaluating a merger, and I think that's a data point that

we shouldn't explicitly just ignore in the analysis.

I think those of us who counsel Boards

MR. SIMONS:

I was just going to say, it really

does depend, and there are some firms that do -- are very

acquisitive, and, you know, they have histories of having

achieved cost savings.

basically whatever you can prove.

it, you can't prove it.

going to occur, you know, six years out is kind of

irrelevant if you can't prove it.

You have to have some

-- you have to have good evidence.

You have to have

evidence.

MS. GOTTS:

And for those, you know, it's

And if you can't prove

The fact that you think it's

But your model, Joe, actually takes

into account the idea that it's a longer term and

appropriately discounts it, and that's the way it should

be done.

take a while to be able to integrate the two companies

They shouldn't be thrown out because it could

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

104

and really realize those, and those are benefits to

society.

MR. SIMONS:

Right.

And the timing, the risk

adjustment should also take into account not only the

chances that -- well, I guess what's going to happen is

the longer out in time, just as a general rule, the

chances that the efficiency is going to be realized are

probably lesser.

into the calculus.

But, you know, you can just figure that

MR. ABBOTT:

Bill, you raised the issue as to

the tradeoff between producers and consumers’ surplus.

Do we have a true pure consumers’ surplus standard?

are there situations in which, say, a diminution of

consumers' surplus in one market can be traded off

against a countervailing gain in producers' surplus

that's so large, that swamps the loss in consumers'

surplus, and you would allow the merger to go ahead?

-- what standard would you apply, or is that consistent

with the case law?

How

Can you apply such a standard?

MR. KOLASKY:

I think I'd give both a lawyer's

answer and maybe a prosecutor's or economist's answer.

Certainly, under the statute, as Bob correctly points

out, the standard is whether the merger is likely to

substantially lessen competition in any line of commerce

in any section of the country.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

105

And so what you are looking at is what is the

impact going to be on consumer welfare?

going to lead to higher prices and lower output, and if

it is, then it should be found anticompetitive.

in making that determination, obviously you have to take

into account efficiencies, including those, as we've been

saying that will kick in over the long term and outweigh

any immediate impact on short-term prices.

Is this merger

I think

The harder question is, and I think producer

surplus is relevant there because, to the extent the

merger increases producer surplus, that is likely to make

this market one which, you know, may be more attractive

to entry from others because it's now suddenly a more

profitable market.

long term stimulate competition and stimulate entry.

therefore, you know, I think that's the point that was

being made by footnote 37.

And so this may actually over the

And

I think it's also very important as we think

about this to have a very clear understanding of what we

mean by "competition."

Many mergers that we might characterize as a merger-to-

monopoly because it's a two to one merger or near

monopoly because it's three to two, do not in fact lessen

competition in an economic sense.

And this goes to Bob's point.

They may reduce the number of rivals.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

But as

106

Baumol makes very clear, you can have a very competitive

market with two rivals, or actually in some cases even

with one rival, where that rival has to constantly

improve its product and innovate in order to get people

to abandon the product they bought in the past and buy

their new product.

you've got software that lasts forever and you don't

improve it so that you give people an incentive to buy

the latest version of it.

account that kind of competition with the installed base.

Your revenues are going to go away if

And we have to take into

The other point, though, is even if you have

pure producer surplus that is not going to inure to the

benefit of consumers, why wouldn't you take that into

account in exercising your prosecutorial discretion as to

whether or not you're going to bring a close case?

the potential anticompetitive effect seems to be quite

minor, if the probabilities are in doubt, as Joe was

pointing out, but you know there are going to be very

substantial producer's surplus created by the merger,

maybe that's one that you ought to just take a pass on as

a matter of prosecutorial discretion.

Yes.

MR. PITOFSKY:

A couple of reactions.

the last point, absolutely right.

little barriers to entry, there's no market power, and

therefore you don't have to worry about anticompetitive

If there are no or

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

107

effects.

that's the case, absolutely right.

It's not common, but when it's there, when

On out-of-cross-market efficiencies, let me

tell you, with the privilege of someone who was there at

the drafting.

merge, prices will go up here but we'll be better

competitors in Europe or better in Florida, or better in

California.

interpreting the statute don't allow that.

On the other hand, as a matter of prosecutorial

discretion, we kept thinking to ourselves, suppose the

advantage of one market is enormous and the disadvantage

in the other market is slight.

with a very felicitous phrase, something about

inextricably interwoven, which nobody has a clue what we

meant there.

You couldn't say, look, if you let us

The statue doesn't allow that, and the cases

So we stuck in a line

(Laughter.)

We said rarely -- I think we used the word

"rarely" twice.

It's just a question of leaving some

discretion to the Commission in the most extreme of

cases.

Now, one might say, well, if you're going to

trade off anticompetitive effects in New York for pro-

competitive effects in Florida, why won't you trade off

anticompetitive effects now for pro-competitive effects

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

108

four years from now?

question.

It's a close call, and it's a good

But I think when we're doing New York versus

Florida, it's right in front of us.

predictions, by the way.

Everything about merger

analysis is prediction.

But I think you can make a more

confident production, New York/California, as of the time

of filing the suit or letting the deal go, than you can

higher prices now, but, oh, four or five years from now,

prices will be lower.

This is all

That's my definition of speculative, uncertain,

clearly not allowed by the statute, not by the

Guidelines.

an elaboration of an efficiencies defense, and I'm still

there.

enough different that I think it throws you into a kind

of a never neverland of speculation.

The Guidelines never thought of that being

It's true, it's close to cross-market, but it's

MR. SIMONS:

Let me just respond for a second.

I could make the argument that the cross-market thing, as

you've said basically, does not -- is not permitted by

the statute.

the statute says substantial lessening of competition in

any market in any part of the country.

about time frame.

But that the temporal tradeoff is, because

It says nothing

So, I don't know, I could come out the other

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

109

way.

MR. ABBOTT:

One issue that underlies some of

the discussion is burden.

parties have to meet to prove up their efficiencies?

mean, we know they must be merger-specific, cognizable,

the buzz words, but as we've heard some critics have

talked about a chicken-and-egg problem; the fact that

poor efficiency arguments are made or developed because

parties don't think staff will treat them seriously, or

alternatively, staff doesn't spend as much time on them

because they're not well made and documented.

What is it?

What burden should the merging

And as practitioners, what

standard do you think should apply?

guidance be given by the agencies as to what degree of

specificity needs to be met or proven by the parties?

MS. GOTTS:

Should better

Actually, during the break, Marius

and I were talking a little bit about not only that, but

one other aspect of it, which Dave Scheffman in an

article touched upon.

Unfortunately, when we do deals, when you start

out a lot of times the group that is able to be in the

know about the possibility of the deal is very limited,

to a CEO, maybe a chairman of the board.

might be why, before the deal is announced, you get two X

in efficiencies that are alleged.

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

And so that

110

After the deal is published -- out there in the

public domain -- you can do a little bit more work, but

there are still limitations which we impose as antitrust

lawyers in how much the companies can get together and

really drill down in understanding how much the

efficiencies will be from a deal.

And as you get closer and closer to the deal, maybe we

loosen up a little bit, but we still are concerned,

especially in deals where they're among competitors.

And so this really adds to the difficulty in

saying the parties should have some real strong burden of

proof and that somehow it should be that you give less

credibility to the efficiencies that are discovered by

the parties after the deal is announced than what the

board considered as you went forward in doing it.

So this makes it more complex than even just

who has the burden of proof in front of the agencies.

It's also a timing issue and a difficulty issue.

One other thing.

Both the Guidelines, and the

agencies are always saying, well, the parties are in

better control of the documents and the evidence to

support it.

you have subpoena power.

deals in the industry, what other third parties have

achieved in the way of efficiencies that we don't have

That's not always the case.

Sometimes --

You can find out what other

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

111

access to.

So it really has to be not so much the burdens

of proof, but we're all trying to get to an answer here,

a sense of whether or not this deal is going to be good

or bad to consumers, and you as the agency should be

using whatever power you can, the parties should, as much

as they can within the confines of antitrust

restrictions, be trying to get you the information you

need.

MR. KOLASKY:

If I can add just one thought to

that, and this comes back to the title of this panel, and

again, this is based both in my private practice

experience and my time at the agency, and that is that I

think the private bar needs to understand better that

efficiencies really have now been integrated into the

merger analysis that they are not just a quote/unquote

"defense."

may be used to rebut inferences of anticompetitive effect

arising from other factors.

In fact, they're not a defense at all.

They

But my experience now is that, as the agencies

have focused more attention on efficiencies, if you don't

have any efficiencies but you've got a horizontal merger

between major competitors, that is actually a major

negative now.

back to that, one of the arguments that Gerte Strauss and

And in fact, in G.E./Honeywell, coming

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

112

others have made is that in that case, the parties didn't

argue any efficiencies, and that was held against them.

I think that the message that the agencies

should be sending through speeches and otherwise is, if

you have a horizontal merger between two major

competitors, you'd better have some efficiencies to talk

about.

skeptical as to whether or not your real motive is to

gain market power and increase price.

Because if you don't, we're going to be very

MR. SIMONS:

Can I say something on the burden?

MR. ABBOTT:

Sure.

MR. SIMONS:

When you raised the question about

the burden, who should have the burden of doing what, the

following example occurs to me.

a really nifty article.

old now, in which he shows, he says, okay, if you think

what's going on here is unilateral behavior a la

Bertrand, here's the cost savings, the marginal cost

savings that have to be realized in order to offset the

price effect.

It's by margin, and it varies by diversion ratio.

Greg Werden actually has

It's like seven or eight years

And he does it by -- you've got a table.

And suppose you had a situation in which, okay,

the staff thought that was the type of model that was

applicable, and the parties came in and showed, yes,

here's the marginal cost reduction that's going to take

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

113

place, and there's no effect, all right?

presented that to a court.

there.

me the court would have to conclude there was no effect.

And you

And that's all that was

What would the court do?

Well, it would seem to

And so the burden would seem to be on the

government to come back and say, yes, but this effect

could be achieved a different way, and here's the way in

which it would be achieved and here's the time period

over which it would be achieved.

So in terms of what would happen in court, I'm

just visualizing that kind of circumstance, it seems the

burden really has to be on the government.

MR. SCHEFFMAN:

Let me mention the burden, and

I've written on this.

100-page single-spaced memoranda analyzing things.

that's not to say the government should take things, you

know, whatever is proffered.

think there is, some amount of sophistication,

particularly by the financial analysts, about what you

should expect to see in a deal.

People in business don't write

Now,

But there has to be, and I

We tend to think that there's things there and

you push buttons and things happen, and that's the way

business works.

consolidation merger say we're going to get $50 million

in cost savings.

That's not true.

So you might in a

Where does that come from?

For The Record, Inc.

Waldorf, Maryland

(301)870-8025

Okay.

Well

114

it comes from some sort of analysis or explanation, but

what it is, is a goal that some manager has to meet.

least that's what it better be, or we shouldn't pay

attention to it.

And as Ilene said from Sarbanes-Oxley, if you

say it to the Street, you are going to be accountable.

So one of the management tools is saying we understand

this much about the other company.

Go achieve it.

micromanagement of how you do it.

consolidations are a savings or not, you shouldn't expect

to see a lot of detailed analysis of them.

get to the more sophisticated stuff, which I agree might

be much more important, I think you have to be able --

you know, it's reasonable for the agencies to require

more.

This is your target.

They're not going to get down into

So whether cost

Now, when you

And that may be difficult to prove.

But I do think that a very important source of

evidence is to ask, you know, show me where you've done

this before.

Have you done some other deal before?

you did in a similar circumstance.

important evidence.

including most decision makers.

It hopefully the

situation should be analogous.

And if we did more of

that, we had some of that at the FTC, if we did more of

Have you done something like this before?

Or sho

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.