Delineating the Relevant Market from Census Industry Classifications

• Date: 01 September 1977
• Published date: 01 September 1977
• Author: Raymond L. Raab
• DOI: 10.1177/0003603X7702200306
• Subject Matter: Article

DELINEATING

THE

RELEVANT

MARKET

FROM

CENSUS

INDUSTRY

CLASSIFICAnONS

by

RAYMOND

L.

RAAB·

INTRODUCTION

The

establishment

of

the

relevant

market

is

of

critical

importance

to

the

courts

in

determinating

the

verdict

in

merger and monopolization

cases.

Numerous

conceptual problems

exist

even

when

the

issues

are

limited

to

the

product market

alone.

Section

I

of

this

paper

investigates

and

clarifies

the

criteria

of

product

delineating,

specifically

the

roles

of

supply and

demand

cross

elasticity.

Section

II

introduces

the

redistribution

technique

for

evaluating

the

Census

Industrial

Classification

(SIC) system. This approach

is

based upon

the

presumption

that

all

five-digit

product

classes

aggregated

into

a good

four-

digit

industry

should experience

similar

relative

growth

or

decline

in

sales

for

a

particular

region

between census

periods.

A

descriptive

statistic

is

developed which

describes

the

similarity

of

the

product

classes

in

reacting

to

temporal changes

in

the

determinants

of

demand.

This

similarity

is

expected

to

be

greater

for

good

industry

classification.

The remaining

sections

deal

with

the

application

of

the

technique.

Sec-

tion

III

examines

several

possible

economic

relationships

which could

distort

the

interpretation

of

the

results.

Section

IV compares

the

quality

*Department

of

Economics,

University

of Minnesota, Duluth, Minnesota

621

622

THE

ANTITRUST

BULLETIN

of

two samples

of

industry

classifications

and

uses

these

samples

as

probable

limits

of

"best"

and

"worst"

classifications.

Section

V

applies

the

tech-

nique

to

the

improvement

of

several

classifications.

Finally,

Section

VI

illustrates

the

nature

of

many

of

the

census'

industry

classifications

and

discusses

some

of

the

practical

problems

which make

industry

delineation

difficult.

I.

DISAGREEMENT

WITH

INDUSTRY

CLASSIFICATION

CRITERIA

A

theoretical

industry

has

been

defined

in

several

ways.

Joe

S. Bain

considers

a

theoretical

industry

as

a

group

of

products

available

to

a

common

group

of

buyers,

which

are

close

substitutes

for

that

group

of

buyers

and

relatively

distant

substitutes

for

all

products

outside

the

industry.2

This

criterion

for

a

theoretical

industry

relies

upon

the

existence

of

a

high

de-

mand

cross

elasticity

between

the

products

sold

by

the

industry.

Here demand

cross

elasticity,

the

elasticity

of

one

seller's

product

with

respect

to

anA

other

seller's

price,3

measures

the

substitutability

of

products

in

consump-

tion.

This

definition

of

industry

points

to

the

"criteria"

adopted

by

the

Court

in

the

Cellophane

case.

4

Maxwell R.

Conklin

and

Harold

T.

Goldstein

define

an

industry

in

a more

operational

context.

They

define

a

census

industry

as

a

group

of

establish-

ments

primarily

engaged

in

the

same

or

similar

lines

of

economic

activity.5

2.

Industrial

Organization

(New

York:

Wile~

and

Sons,

1968),

p.

124.

3. See

Nicholas

Kaldor's

initial

reference

to

demand

cross

elasticity

in

"Market

Imperfections

and

Excess

Capacity",

Economica,

1935,

p.

35.

4. U.S. v ,

E.!.

Du

Pont

de Nemours &ce., 351 M.S. 311.

5.

Business

Concentration

and

Price

Policy,

National

Bureau

of

Economic

Research,

(Princeton

University

Press,

1955),

p.

18.