FTC Staff Comment Before the Federal Communications Commission In the Matter of Review of the Commissions Regulations Governing Television Broadcasting

• . t ~
Before the
FEDERAL COMMUNICATIONS COMMISSION
Washington, 	D.C.
V920020
COMMISSION AUTHORIZED
In The Hatter of
Review of 	the Commission's
Regulations 	Governing Television
Broadcasting
)
)
)
)
)
)
)
MM Docket No. 	91-221
Comments 	of 	the 	Staff of
the 	Bureau of 	Economics 	of
the 	Federal Trade Commission*
September 	24, 	1992
* 	These 	comments 	represent 	the 	views 	of 	the 	staff 	of 	the
Bureau of Economics 	of the Federal Trade Commission. 	They are 	not
necessarily 	the 	views 	of 	the 	Commission 	itself 	or 	any 	individual
Commissioner. 	Inquiries 	regarding 	these 	comments 	should 	be
directed 	to 	Michael 	Vita, 	Bureau 	of 	Economics 	(202-326-3493).

Executive Summary
The 	staff 	of 	the 	Bureau 	of 	Economics 	of 	the 	Federal 	Trade
Commission presents 	comments, 	from 	the 	perspective 	of 	economic
efficiency, 	competition, 	and 	antitrust 	enforcement, 	on 	the
Federal 	Communications 	Commission's 	proposal 	to 	relax 	certain 	of
its 	rules 	regulating ownership 	of 	television 	stations. 	This
reply 	comment 	does 	not 	address 	other 	issues, 	such 	as 	the
relationship 	between diversity of 	ownership 	and 	diversity of
editorial 	viewpoint, 	that may 	be 	important 	to 	the 	FCC. 	The
comment 	finds 	that 	the 	proposals 	under 	consideration may promote
market 	efficiency without 	posing a 	threat 	to 	competition.
The 	"duopoly 	rule" 	that 	now bans 	ownership 	of 	nearby
television 	stations 	could be 	appropriate 	if 	the 	net 	effect 	of
such 	combinations 	is 	likely 	to 	be 	anticompetitive, 	and 	the 	costs
of 	case-by-case 	evaluation 	are 	likely 	to 	exceed 	the 	benefits 	from
allowing 	those 	combinations 	having 	positive 	net 	effects. 	But 	our
analysis 	suggests 	that 	there 	could be 	combinations 	that 	might 	be
viewed 	as 	competitively unobjectionable, 	when 	judged by 	the
standards 	of 	the 	1992 	Department 	of 	Justice 	and 	Federal 	Trade
Commission Horizontal 	Merger Guidelines, 	and 	that 	could 	lead 	to
efficiencies. 	In addition, 	economic 	theory 	provides 	no 	basis 	for
concluding that 	permitting 	some 	such combinations 	would 	reduce
program variety. 	Relaxing 	the 	blanket 	rule 	prohibiting 	"duopoly"
ownership may 	thus 	be 	appropriate.
The 	FCC may wish 	to 	consider, 	as 	an 	alternative 	to 	the
proposals 	to 	relax 	the 	rule 	regulating ownership 	practices, 	using
case-by-case 	analysis 	of 	market 	conditions 	when considering
license 	assignment 	or 	transfer 	applications. 	The 	elements 	of
this 	analysis 	could be 	similar 	to 	those 	that 	the 	FCC 	already
applies 	to 	requests 	for 	waiver 	of 	the 	radio-television 	cross-
ownership 	rule, 	which 	in 	turn 	are 	similar 	to 	elements 	of 	the
Horizontal 	Merger Guidelines 	analysis. 	The 	FCC 	might 	also 	wish
to 	consider 	a 	rule 	that, 	like 	the 	Horizontal 	Merger Guidelines,
adopted 	a 	"safe-harbor" 	approach, 	defining 	conditions 	in which
combination 	is 	unlikely 	to 	be 	challenged. 	In 	antitrust
enforcement 	of 	the 	merger 	laws, 	the 	combination of 	safe-harbor
standards 	and 	case-by-case 	review 	is 	workable 	because most 	merger
transactions 	do 	not 	warrant 	challenge. 	We 	do 	not 	know 	the
relative 	costs 	to 	the 	FCC 	of 	case-by-case 	review compared 	to
rule-based 	regulation of 	combinations, 	and 	thus 	we 	express 	no
opinion 	on which would 	be 	best 	here.
Relaxing 	the 	national 	ownership 	rules 	for 	television would
also 	not 	appear 	to 	pose 	a 	threat 	to 	competition. 	Because
television 	stations 	compete 	almost 	entirely 	in 	local 	markets, 	the
competitive 	effects 	of 	multiple 	station ownership 	are 	best
understood 	by 	analyzing 	the 	impact 	of 	such 	ownership 	in
individual 	markets. 	Imposing 	national 	ownership 	limits 	is 	not
i

likely 	to 	increase 	the variety 	of 	programming 	available 	to
viewers, 	or 	to 	protect 	competition 	in 	advertising markets. 	But
such 	limits 	may 	prevent 	firms 	from 	realizing certain efficiencies
from 	group 	ownership.
The 	rules 	regulating 	television-radio 	cross-ownership may
also 	be 	relaxed without 	posing a 	threat 	to 	competition. 	In
particular, 	the 	FCC might 	consider waivers 	of 	its 	rules, 	using
its 	present 	waiver 	standards 	or the 	analytical 	framework 	of 	the
Horizontal 	Merger Guidelines, 	even 	in 	areas 	other 	than 	the 	top
television markets.
ii

Before the
FEDERAL COMMUNICATIONS COMMISSION
Washington, 	D.C.
In The Hatter of
Review 	of 	the Commission's
Regulations 	Governing Television
Broadcasting
)
)
)
)
)
)
)
MM Docket No. 	91-221
Comments 	of 	the 	Staff of
the 	Bureau of 	Economics 	of
the 	Federal 	Trade Commission1
September 	24, 	1992
I. 	Introduction
The 	staff 	of 	the 	Federal 	Trade 	Commission's 	Bureau 	of
Economics 	is 	pleased 	to 	respond 	to 	this 	notice 	of 	proposed
rulemaking 	("NPRM") 	through which 	the 	Federal 	Communications
Commission 	(FCC) 	seeks 	comments 	on 	proposals 	to 	relax 	certain
limitations 	on 	television 	station ownership.2 	The 	FCC 	is
considering 	increasing 	the 	number 	of 	stations 	nationwide 	that may
1 	These 	comments 	represent 	the 	views 	of 	the 	staff 	of 	the
Bureau of 	Economics 	of the 	Federal Trade Commission. 	They are not
necessarily 	the 	views 	of 	the 	Commission 	itself 	or 	any 	individual
Commissioner. 	Inquiries 	regarding 	these 	comments 	should 	be
directed 	to 	Michael 	Vita, 	Bureau 	of 	Economics 	(202-326-3493).
2 This 	reply 	comment 	addresses 	issues 	relating 	to 	economic
efficiency, 	competition, 	and the enforcement of the antitrust laws.
It does 	not address 	other issues, 	such as 	the relationship between
diversity of 	ownership 	and 	diversity 	of 	editorial 	viewpoint, 	that
may 	be 	important 	to 	the 	FCC.
1

be 	held 	in common 	ownership,3 	relaxing 	the 	prohibition 	against
owning more 	than 	one 	television 	station 	serving a 	single 	area,
and 	relaxing 	the 	prohibition 	against 	owning 	a 	television-radio
combination 	in 	a 	single 	area. 4 This 	comment 	provides 	general
observations 	about 	possible 	economic 	benefits 	from 	relaxing 	the
multiple 	ownership 	limitations 	and presents 	data 	to 	help 	assess
the 	consequences 	of 	permitting common 	ownership 	of 	more 	than 	one
local 	station.
Section 	III 	of 	this 	comment 	summarizes 	the 	current 	FCC
ownership 	limitations 	and 	the 	proposed 	changes. 	Section 	IV
discusses 	the 	types 	of 	efficiencies 	that 	might 	be 	associated 	with
joint 	ownership 	of 	broadcast 	facilities. 	Section V addresses
issues 	at 	the 	local 	scale, 	such 	as 	the 	creation of 	market 	power
in 	advertising markets 	and 	the 	effect 	of 	market 	structure 	on
programming variety. 	Section VI 	discusses 	the 	ownership 	limits
at 	the 	national 	scale. 	Section VII 	addresses 	radio-television
cross-ownership. 	Section VIII 	presents 	our conclusions.
II. 	Expertise of 	the 	Staff of 	the 	Federal 	Trade Commission
The 	FTC 	is 	an 	independent 	regulatory 	agency 	responsible 	for
preventing 	unfair methods 	of 	competition and 	unfair 	or 	deceptive
acts 	or 	practices. 5 In 	response 	to 	requests 	by 	federal, 	state,
and 	local 	government 	bodies, 	the 	staff 	of 	the 	FTC 	often 	analyzes
3 NPRM, 	! 	11.
4 NPRM, 	!! 	18-20.
5 	15 	U.S.C. 	§§ 	41-49.
2

regulatory 	or 	legislative 	proposals 	that may 	affect 	competition
or 	the 	efficiency of 	the 	economy. 	In 	the 	course 	of 	this 	work, 	as
well 	as 	in 	antitrust 	and consumer 	protection research, 	nonpublic
investigations, 	and 	litigation, 	the 	staff 	applies 	established
principles 	and 	recent 	developments 	in economics 	to 	competition
and 	consumer 	protection matters. 6
The 	staff 	of 	the 	FTC 	has 	commented on a 	number 	of 	issues
involving 	the 	FCC 	and 	its 	responsibilities, 	beginning with 	a 	1924
report 	to 	the 	House 	of 	Representatives 	about 	competition 	issues
in 	radio. 7 More 	recently 	the 	staff 	has 	submitted comments 	to
the 	FCC 	on 	radio 	ownership 	rules 	and 	policies;8 competition,
rate 	deregulation, 	and 	cable 	television 	service;9 common
ownership of 	cable 	systems 	and 	national 	television 	networks;10
the 	"must 	carry" 	rules 	applied 	to 	cable 	television 	systems;l1
the 	rules 	requiring 	licenses 	to 	be 	held 	for 	three 	years 	before
being 	transferred;12 	network 	ownership of 	financial 	interests·
6 See, 	~, Mathios and Rogers, 	The 	Impact of State Price and
Entry 	Regulation 	on 	Intrastate 	Long 	Distance 	Telephone 	Rates,
Bureau 	of 	Economics 	Staff 	Report 	to 	the 	Federal 	Trade 	Commission,
November 	1988.
7 	Federal 	Trade 	Commission, 	Report 	on 	the 	Radio 	Industry,
GPO, 	1924.
8 Reply 	Comments 	of 	the 	Staff 	of 	the 	Bureau 	of 	Economics 	of
the 	Federal 	Trade 	Commission, 	In 	re 	Revision 	of 	Radio 	Rules 	and
Policies, 	MM No. 	91-140, 	September 	5, 	1991.
9 MM No. 	89-600.
10 CT No. 	82-434.
11 MM No. 	90-4.
12 BC 	No. 	81-897.
3

and 	syndication rights;13 	spectrum 	allocation 	and 	standards 	for
digital 	audio 	broadcasting,14 	and 	the 	regulation of 	"900"
telephone 	number 	services. 15
III. 	The FCC's 	Current 	and 	Proposed Ownership Rules
The 	FCC's 	national 	multiple 	ownership 	rule 	limits 	the 	number
and 	total 	audience 	"reach" 	of 	television 	stations 	in which 	an
individual 	(or 	a 	single 	entity) 	may 	hold 	an 	attributable
interest. 16 	The 	current 	limits, 	set 	in 	1985, 	are 	a 	maximum 	of
12 	television 	stations 	and 	maximum 	total 	reach 	of 	25 	percent 	of
national 	television 	households. 	The 	related 	"duopoly" 	rule,17
dating 	from 	1964, 	prohibits 	holding 	attributable 	interests 	in 	two
or more 	commercial 	television 	stations 	with overlapping Grade 	B
13 MM 	Nos. 	82-345 	and 	90-162.
14 GEN 	No. 	90-357.
15 	CC 	No. 	91-65.
16 The 	group ownership 	rule 	defines 	"national 	audience 	reach"
as 	"the total number of television households in the Arbitron area-
of-dominant-influence 	(ADI) 	markets 	in which the relevant stations
are located, 	divided by the total national television households as
measured 	by 	the ADI 	data." 	47 	C.F.R. 	§ 	73.3555 	(d)(3)(i).
17 The term 	"duopoly" means different things 	in communications
law 	and 	in 	antitrust 	law 	and 	economics. 	Communications 	law 	uses
the 	term 	to 	describe 	holding 	an 	ownership 	interest 	in 	two
facilities 	in the same area. 	In antitrust usage, 	a 	"duopoly" would
be a 	market 	in which there were only two 	suppliers. 	The antitrust
term 	implies 	both 	that 	there 	are 	no 	additional 	suppliers 	(here,
stations), 	and 	that 	the 	service 	and 	the 	area 	comprise 	antitrust
markets. 	See 	p. 	15 	below 	for 	discussion 	of 	defining 	antitrust
markets.
4

contours. 18 Finally, 	combinations 	of 	radio 	and 	television
stations 	in 	the 	same market 	generally 	are 	prohibited;19 	however,
FCC 	rules 	provide 	for 	waiving 	this 	prohibition under 	certain
circumstances. 2o
The 	NPRM 	describes 	the 	recent, 	substantial 	growth 	in 	the
variety of 	communications 	sources 	available 	to consumers. 	These
include 	not 	only broadcast 	and 	cable 	television, 	but 	also
wireless 	cable, 	low-power 	television, 	home 	satellite 	receivers,
video 	and 	audio 	recordings, 	and, 	soon, 	direct 	satellite 	audio 	and
video 	services 	with 	digital 	audio. 21 The NPRM 	notes 	that 	this
multiplicity of 	sources 	poses 	a 	substantial 	competitive 	challenge
to 	television 	broadcasters 	in 	seeking viewers. 22 The 	NPRM
further 	notes 	that 	profitability of 	television broadcasters 	is
18 According to the Television & Cable Factbook 	(1988 	ed., 	p.
A-14), 	with 	"Grade B" 	service, 	"the quality of picture is 	expected
to 	be 	satisfactory 	to 	the 	median 	observer 	at 	least 	90 	percent 	of
the time 	for 	at 	least 50 	percent 	of the receiving 	locations 	within
the 	contour, 	in 	the 	absence 	of 	interfering 	and 	co-channel 	and
adjacent 	channel 	signals." 	With 	"Grade A" 	service, 	"satisfactory
service is expected at 	least 90 percent of the time 	for at 	least 	70
percent of the receiving 	locations." 	Thus, 	the Grade A contour 	is
smaller 	and 	contained within 	the 	Grade 	B contour.
19 The 	term 	"market" 	has 	a 	technical 	meaning 	in 	anti trust 	law
and 	economics, 	which 	may 	differ 	from 	its 	usage 	in 	communications
practice. 	What 	communications 	practitioners 	may 	consider 	a
"market" 	might, 	or might not, 	be considered an antitrust 	"market."
See 	p. 	15 	below 	for 	explanation 	of 	the 	antitrust 	analytical
approach 	to 	market 	definition.
20 NPRM, 	! 	22.
21 NPRM, 	!! 	3-7.
22 NPRM, 	!! 	5-7. 	Some 	of 	these 	sources 	do 	not 	compete 	with
local 	broadcasters 	in 	the 	sale 	of 	advertising 	time 	in 	particular
local 	areas. 	The 	issue of 	advertising competition is 	addressed 	in
greater 	detail 	in § 	V, 	below.
5

poor 	and expresses 	the 	FCC's 	concern 	that 	the 	ownership
dispersion 	required by 	its 	existing 	rules 	might 	undermine 	some
stations' 	competitive viability.23 	Thus, 	the 	FCC 	is 	considering
whether 	relaxing 	its 	ownership 	limitations 	could, 	by 	promoting
efficiency, 	enable 	broadcasters 	to 	compete more 	effectively, 	to
the 	benefit 	of 	television 	audiences. 24
Concerning 	specifically 	the 	"duopoly" 	rule's 	limits 	on 	local
multiple 	ownership, 	the 	FCC 	has 	asked whether 	overlap 	should be
based 	on 	the Grade 	A contour, 	rather 	than 	the 	Grade 	B contour.
The 	FCC 	also 	seeks 	comment 	on 	permitting combinations 	of 	stations
with overlapping Grade A contours 	where 	at 	least 	one 	is 	a 	UHF
station 	and 	at 	least 	six 	independently-owned 	television 	stations
would 	remain 	in 	the 	market. 	The 	FCC 	also 	has 	asked whether 	only
UHF-UHF 	combinations 	should be 	permitted. 25
With 	respect 	to 	the 	national 	multiple 	ownership 	limitations,
the 	FCC 	is 	considering 	several 	options. 26 It 	has 	proposed
raising 	both 	the 	number 	of 	stations 	and 	national 	reach 	limits,
offering two 	specific 	proposals: 	(1) 	increasing 	the 	number 	of
stations 	limit 	from 	12 	to 	20 	or 	24, 	and 	the 	audience 	reach 	limit
from 	25 	to 	35 	percent; 	or 	(2) 	increasing 	the 	number 	of 	stations
23 	NPRM, 	! 	7.
24 	NPRM, 	! 	7. 	The 	NPRM 	notes 	(! 	11) 	that 	the 	FCC 	wishes 	to
ensure 	that regulation does 	not 	"impede the competitive ability of
broadcast 	television 	stations 	or 	preclude 	them 	from 	taking
advantage 	of 	certain economic 	efficiencies."
25 NPRM, 	!! 	18-20.
26 NPRM, 	!! 	11-13.
6

limit 	to 	18, 	and 	the 	audience 	reach 	limit 	to 	30 	percent. 	It 	has
also 	asked 	for 	comment 	on 	increasing 	only 	the 	number 	of 	stations
limit, 	while 	retaining 	the 	25 	percent 	audience 	reach 	limit.
Lastly, 	the 	FCC 	has 	requested comment 	on eliminating 	the
restrictions 	on 	local 	"cross-ownership" 	of 	television 	and 	radio
stations 	and 	instead basing 	its 	restrictions 	on ownership 	rules
for 	each broadcast 	service 	considered 	separately. 	As
alternatives, 	the 	FCC 	is 	seeking comment 	on permitting 	holding
interests 	in one AM, 	one 	FM, 	and one 	television 	station 	in 	a
market, 	and 	on 	allowing 	cross-ownership of 	television 	and AM
stations.
IV. 	Economic 	Efficiencies 	from Common Ownership
The 	competitive 	process 	generally 	rewards 	efficiency 	in ways
that 	ultimately 	benefit 	consumers. 	In competitive markets, 	the
profit 	incentive 	tends 	to 	displace 	inefficient 	ownership 	patterns
with ones 	that 	are 	more 	efficient 	and more 	profitable. 	One
aspect 	of 	greater efficiency 	is 	more 	cost 	effective 	use 	of 	labor,
capital, 	and 	other 	inputs. 	Another 	is 	improvement 	or 	innovation
in 	the 	array of 	goods 	or 	services 	provided. 	Differences 	in
ownership 	patterns 	can be 	related 	to 	differences 	in 	operating
cost 	effectiveness 	or 	in product 	quality. 	If 	common ownership 	of
two 	productive 	facilities, 	such 	as 	two 	broadcast 	stations, 	is
more 	efficient 	than 	separate 	ownership, 	then 	the 	facilities 	will
tend 	to 	be 	commonly owned, 	absent 	legal 	or 	regulatory
restrictions 	on common ownership.
7

If 	multiple 	ownership 	leads 	to 	cost 	savings, 	the 	result
could be 	significant benefits 	to 	consumers. 	For 	example, 	savings
may 	be 	invested 	in producing higher quality programming. 27 In
addition, 	the 	prospect 	of 	savings 	from 	multiple 	ownership, 	if
significant, 	might 	encourage 	construction of 	new broadcast
facilities 	that 	could enhance 	program variety, 	as 	well 	as 	prevent
the 	exit 	of 	stations 	that 	are 	(or would 	become) 	unprofitable 	if
compelled 	to 	continue 	operating on 	a 	stand-alone 	basis.
Efficiencies 	that 	can 	result 	from 	combining 	local 	broadcast
station operations, 	which 	have 	been 	studied 	as 	they 	apply 	to
combinations 	of 	radio 	stations, 	may 	include 	cost 	savings 	in
administration 	and 	overhead, 	promotion, 	equipment, 	and
programming. 	These 	were 	described 	in our Reply 	Comments 	on 	the
FCC's 	1991 	NPRM 	on 	radio 	station ownership.28 	For 	example, 	the
engineering 	staff 	required 	for 	two 	commonly-owned 	stations 	may 	be
smaller 	than 	the 	sum of 	the 	staffs 	required by 	separate 	stations.
In 	our 	radio 	rules 	Reply 	Comments, 	we 	cited 	empirical 	research
which 	suggested that 	joint 	operation of 	local 	radio 	stations
27 For 	evidence 	that 	competition 	among 	broadcast 	outlets
resul ts 	in 	greater 	expenditures 	on 	programming, 	see 	Fournier,
Nonprice 	Competition 	and 	the 	Dissipation 	of 	Rents 	From 	Television
Regulation, 	51 	S. 	Econ. 	J. 	754-65 	(1985).
28 Supra, 	note 	8. 	In 	previous 	proceedings, 	Provision 	of
Improvements and Benefits to the AM Radio Broadcast Service, MM No.
87-7 	the National Association of Broadcasters and other commenters
also described the kinds 	of cost 	savings 	that may flow 	from common
ownership 	of 	same-market 	radio 	stations. 	See 	Appendix 	A 	to
Comments 	of 	the 	National 	Association of 	Broadcasters; 	Comments 	of
the National Association of Broadcasters at Appendix C; 	Comments of
CBS, 	Inc,. 	at 	9-12; 	Comments 	of 	Capital 	Cities (ABC, 	Inc., 	at
Appendix C.
8

could 	generate 	savings 	of 	up 	to 	35 	percent 	on 	salary 	expenses 	and
up 	to 	18 	percent 	on 	technical 	equipment 	costS. 29 An 	additional
empirical 	analysis, 	contained 	in 	the 	Appendix 	to 	that 	Reply
Comment, 	implied 	that 	joint ownership of 	an AM 	and 	an 	FM 	station
in 	the 	same 	local 	market 	can create 	substantial 	efficiencies. 30
Also, 	the 	fact 	that 	nearly 	60 	percent 	of 	all 	radio 	stations 	are
in AM-FM combinations 	strongly 	suggests 	that 	common ownership
results 	in efficiencies.
Of 	course, 	these 	empirical 	results 	apply 	directly 	only 	to
the 	combination of 	an AM 	and 	an 	FM 	station 	in 	a 	local 	market, 	and
not 	to 	combinations 	of 	television 	stations. 	Since 	same-market
television combinations 	have 	been 	prohibited, 	direct 	empirical
29 	See 	Comments 	of 	the 	National 	Association 	of 	Broadcasters,
MM No. 	91-140, 	August 5, 	1991, 	p. 	20. 	These estimates are based on
comparisons 	of 	the 	reported 	costs 	of 	commonly-owned 	AM-FM 	radio
stations and the sum of the reported costs of independent AM and FM
stations. 	.
30 The 	Reply Comments 	Appendix contained a 	study conducted by
FTC 	economists 	Keith 	Anderson 	and 	John 	Woodbury. 	This 	study
compared 	the 	price 	that 	would 	be 	paid 	for 	an AM-FM 	combination 	in
the 	same market 	with 	the 	sum of 	the 	prices 	that 	would 	be 	paid 	for
the 	same 	stations 	if 	they 	were 	independently 	owned 	and 	operated.
If efficiencies result from combined operation, 	the stations' 	value
ought 	to be greater when sold as 	a 	combination than when sold 	(and
when 	expected 	to 	remain 	operated) 	as 	stand-alones. 	Anderson 	and
Woodbury 	found 	that 	the 	average 	price 	paid 	for 	an 	existing 	AM-FM
combination would exceed by 	about 	20 	percent 	the 	price 	that 	would
be paid 	for 	the 	same 	stations 	if those 	stations were operated 	(and
were expected to continue to operate) 	on a 	stand-alone basis. 	This
difference was statistically significant at the one percent 	level.
To 	ensure 	that 	the 	price 	differences 	were 	not 	attributable 	to
factors 	other 	than 	combination 	versus 	stand-alone 	ownership,
regression 	analysis 	was 	used 	to 	control 	for 	other 	possible
differences, 	including differences 	in 	the 	level 	of 	competition 	in
local 	radio 	markets. 	The 	study 	used 	market 	concentration 	as 	a
proxy 	measure 	(albeit, 	an 	imperfect 	one) 	of 	the 	level 	of
competition.
9

identification of 	efficiencies 	that 	might 	be 	associated with 	such
mb · 	t' 	.. 	. bl 	31	co 	1na 	10ns 	1S 	1mposs1 	e. 	The 	efficiencies 	from 	such
combinations 	may 	turn 	out 	to 	be 	different 	from 	those 	estimated
for 	AM-FM 	radio 	combinations. 32 Local 	television combinations
may yield 	similar efficiencies, 	if 	they could enjoy 	the 	same
benefits 	from 	joint 	operation, 	such 	as 	lower equipment 	and
personnel 	costs.
Although 	there 	is 	no 	direct 	evidence 	of 	efficiencies 	from
common 	ownership of 	television 	stations 	with overlapping 	signals,
there 	has 	been 	some 	research on 	the 	effects 	of 	common 	ownership
of 	non-overlapping 	stations. 	Certain 	types 	of 	efficiencies, 	such
as 	savings 	from 	shared 	technical 	facilities, 	would 	not 	be
available 	to 	non-proximate 	stations, 	but 	television 	stations 	that
are 	in 	common 	ownership 	groups 	could 	receive 	other 	benefits, 	such
as 	efficiencies 	in program 	selection, 	acquisition, 	and
production. 	If 	so, 	group-owned 	stations 	might 	be 	expected 	to
receive 	higher 	audience 	ratings 	than 	non-group 	stations. 	This
hypothesis 	was 	tested 	by 	Parkman, 	who 	examined 	the 	impact 	of
group 	ownership 	on 	the 	ratings 	of 	local 	news 	broadcasts. 33
31 	For 	the 	same 	reason, 	we 	have 	no 	direct 	evidence 	on 	any
possible anticompetitive effects 	from allowing common ownership of
television 	stations 	with overlapping 	signals.
32 The anticompetitive effects from combining local television
stations 	also 	could differ 	from 	those 	associated with AM-FM 	radio
station combinations.
33 	See 	Parkman, 	The Effect 	of 	Television Station Ownership 	on
Local 	News 	Ratings, 	64 	Rev. 	Econ. 	& Stat. 	289-95 	(1982).
10

Controlling 	for 	other 	factors,34 	he 	found 	a 	positive 	and
statistically 	significant 	relationship between 	group ownership
and 	local 	news 	ratings. 	Parkman 	interpreted this 	finding 	as
consistent 	with 	the 	proposition that 	"group owners 	were 	able 	to
adapt 	to 	changes 	in 	the 	production techniques 	of 	local 	TV 	news
programming 	better than 	other owners. 11
35
v. 	Local 	Ownership Limits
1. 	Market 	Power 	in Advertising Markets
Broadcast 	television 	stations 	are 	in 	the 	business 	of 	selling
air 	time 	to 	advertisers. 	In deciding whether 	to 	relax 	the 	rule
against 	owning 	stations 	with 	overlapping 	signal 	contours, 	an
important 	competitive 	concern 	is 	whether 	common 	ownership 	of
same-market 	broadcast 	facilities 	would 	tend 	to 	create 	or 	enhance
market 	power 	in 	local 	advertising markets.
When 	assessing whether 	a 	proposed merger 	or 	acquisition
should be challenged, 	the 	FTC 	and 	Department 	of 	Justice 	("DOJ")
d 	1 	. 	. 	f· 	t 	36	con 	uct 	an 	ana YS1S 	1n 	1ve 	sages. 	This 	analysis 	is
34 	These 	other 	factors 	included 	each 	station's 	other
characteristics 	(e. g. , 	VHF 	or 	UHF, 	network 	affiliate 	or
independent): 	the 	number 	and 	type 	of 	other 	stations 	in 	the 	local
market; 	whether that 	station or 	its 	rivals owned a 	local newspaper
or 	radio 	station; 	the 	age 	of 	the 	station; 	and 	the 	time 	zone 	in
which 	the 	station operated.
35 	Parkman, 	supra 	note 	33, 	at 	294.
36 	The 	principal 	federal 	antitrust 	law 	concerning 	mergers 	is
the 	Clayton 	Act. 	Section 	7 	of 	the 	Clayton 	Act 	exempts 	from 	its
coverage 	broadcast 	license 	transfers 	and 	assignments 	that 	receive
FCC 	approval. 	In 	principle, 	the 	Federal 	Trade 	Commission Act 	and
the 	Sherman 	Act 	can 	also 	apply 	to 	mergers 	and 	acquisitions, 	and
(continued ... )
11

described 	in 	the 	1992 	Department 	of 	Justice 	and 	Federal 	Trade
Commission Horizontal 	Merger Guidelines 	("Merger Guidelines"). 37
The 	first 	stage 	is 	to 	assess 	whether 	the 	proposed 	transaction
would 	significantly 	increase 	concentration, 	and 	result 	in 	a 	high
level 	of 	concentration, 	in a 	properly 	defined 	and measured
antitrust 	market. 	Second, 	the 	agencies 	assess 	whether 	the
transaction would 	raise 	concern 	about 	potential 	anticompetitive
effects, 	given what 	is 	known 	about 	market 	concentration and 	other
market-specific 	factors 	affecting competitive 	behavior. 	Third,
the 	agencies 	assess 	whether 	entry 	into 	the 	relevant 	market 	would
be 	timely, 	likely, 	and 	sufficient 	either 	to 	deter 	or 	counteract
the 	competitive 	effects 	of 	concern. 	Fourth, 	the 	agencies 	ask
whether 	there 	are 	efficiencies 	associated with 	the 	transaction
that 	could 	not 	be 	achieved 	through 	other means 	that 	pose 	less 	of
a 	threat 	to 	competition. 	Last, 	the 	agencies 	ask whether, 	but 	for
the 	transaction, 	either 	of 	the 	parties 	to 	it 	would 	fail, 	causing
their 	productive 	assets 	to 	exit 	the 	market. 	If 	the 	problem were
determining whether 	any particular combination of 	nearby
television 	stations 	would 	raise 	competitive concerns 	under 	the
antitrust 	laws, 	it would 	be 	addressed 	by 	applying 	this 	five-step
analysis.
36 ( ••• continued)
neither 	contains 	an 	exemption 	from 	its 	coverage 	for 	combinations
involving 	broadcast 	licenses. 	However, 	because 	the 	FCC 	has
exercised 	its 	regulatory 	authority over 	broadcaster 	combinations,
there 	has 	been 	no 	occasion 	to 	test 	the 	application 	of 	the 	Sherman
Act 	or 	the 	FTC Act 	to 	these 	transactions.
37 See Merger Guidelines, 	57 Fed. 	Reg. 	41,552 	(Sept. 	10, 	1992),
! 	0.2.
12

A similar 	analysis 	could be 	applied 	to 	the 	problem before
the 	FCC, 	which 	is 	addressing 	general 	rules 	to 	govern
combinations. 	The 	problem, 	in 	terms 	of 	competition policy, 	could
be 	restated 	as 	determining 	how many 	possible 	combinations 	there
might 	be 	that 	would be 	unobjectionable 	under Merger 	Guidelines
criteria. 	If 	there 	were 	few 	-- 	that 	is, 	if 	for 	most 	potential
transactions, 	significant 	concentration 	increases 	in 	relevant
antitrust 	markets 	and 	other 	factors 	imply 	that 	the 	effect 	may 	be
substantially 	to 	lessen 	competition, 	and 	there 	was 	little
probability of 	achieving otherwise 	unattainable 	efficiencies
then 	there 	would 	be 	little 	competition-policy rationale 	for
relaxing 	the 	local 	ownership 	rules. 38 But 	if 	a 	large 	proportion
of 	potential 	transactions 	would 	not 	violate 	antitrust 	standards,
it 	could be 	efficient 	to 	relax 	the 	absolute 	prohibition. 	Which
particular method 	to 	choose, 	whether 	a 	relaxed version 	of 	the
present 	rule, 	a 	safe-harbor 	approach, 	or case-by-case 	review
using 	procedures 	and 	principles 	like 	those 	applied 	in other
antitrust 	markets, 	would depend 	on 	the 	balance 	of 	costs,
including 	administrative 	costs, 	and 	benefits.
To 	assess 	some 	of 	the 	potential 	benefits 	and 	costs 	of
relaxing 	the contour 	overlap 	rule, 	it will 	be 	useful 	to 	follow
the 	steps 	in 	the 	Merger Guidelines' 	analysis. 	The 	first 	step 	is
to 	delineate 	properly the 	product 	and 	geographic 	markets 	in which
television 	stations 	compete, 	and 	then measure 	concentration 	in
38 In this 	situation, 	the rules 	could be efficient, 	since they
would 	avoid 	the 	costs 	of 	case-by-case 	adjudication.
13

these 	markets. 39 Under 	the 	Merger Guidelines, 	defining 	the
product market 	starts 	with 	asking whether 	a 	hypothetical
monopolist 	seller 	of 	some 	product 	(or 	service) 	would 	find 	it
profitable 	to 	impose 	a 	"small 	but 	significant 	and 	nontransitory"
price 	increase. 4o Here, 	the 	question might 	be whether 	a
monopolist 	seller 	of 	broadcast 	television 	advertising 	time 	would
find 	it 	profitable 	to 	impose 	such a 	price 	increase 	on
advertisers. 	If 	the 	answer 	is 	"no," 	because 	advertisers 	would
switch 	to 	other 	advertising media 	in 	response 	to 	this 	attempted
price 	increase, 	then 	the 	product market 	would 	be 	expanded 	to
include 	the 	next-best 	substitute 	for 	broadcast 	television
advertising, 	and 	the 	price 	increase 	question would 	be 	posed
again. 	This 	process 	is 	repeated 	until 	a 	set 	of 	services 	is 	found
for 	which 	such a 	price 	increase 	by 	a 	hypothetical 	monopolist
would 	be 	profitable.
The 	boundaries 	of 	the 	relevant 	geographic 	market 	are
delineated 	similarly. 	To 	illustrate, 	suppose 	that 	"broadcast
39 We 	note 	that 	concentration 	figures 	in 	the 	FCC 	staff 	report
entered 	into 	the 	record 	for 	this 	docket 	("Overview 	of 	the
Television Industry," Federal Communications Commission, Policy and
Rules 	Division, 	Mass 	Media 	Bureau, 	March 	1992) 	("Television
Overview") 	understate 	concentration 	in 	local 	broadcast 	television
advertising. 	The 	FCC 	computes 	a 	Herfindahl-Hirschman 	Index 	(HHI)
of 	187 	using 	data 	on 	delivered 	audiences. 	Because 	this 	index 	is
computed 	on 	a 	nationwide 	basis, 	it 	does 	not 	reflect 	the 	local
nature 	of 	broadcast 	competition.
40 The 	"small 	but 	significant" 	increase 	is 	generally taken 	to
be 	five 	percent. 	The 	following 	discussion 	assumes 	that 	the
hypothetical 	monopolist 	cannot 	price 	discriminate 	among 	different
buyers 	of 	the 	product. 	If 	such 	discrimination 	is 	possible, 	the
market 	definition 	procedure 	must 	be 	modified. 	See 	Merger
Guidelines, 	i 	1.12.
14

television 	advertising" 	constituted the 	relevant 	product market
for 	analyzing a 	merger 	between 	two 	stations 	in 	area A. 	One 	would
then 	ask whether 	a 	monopolist 	of 	television 	broadcast 	advertising
in 	area A could profitably 	raise 	prices 	by 	a 	small 	but
significant 	and 	nontransitory 	amount 	(again, 	usually 	five
percent), 	or whether 	its 	attempt 	would 	be 	defeated 	by 	advertisers
switching 	to 	television 	stations 	in 	nearby 	area 	B. 	If 	the 	price
increase would be 	profitable, 	then 	area A would 	constitute 	the
relevant 	geographic 	market. 	If 	not, 	the 	geographic 	market 	would
be 	expanded 	to 	include 	television 	advertising 	in 	area 	B, 	and 	the
question 	repeated. 	This 	procedure would continue 	until 	an 	area
was 	found 	in which a 	monopolist 	could profitably raise 	broadcast
television 	advertising 	prices. 	That 	area 	would 	be 	the 	relevant
geographic 	market.
Ideally, 	the 	first 	step 	in 	appraising 	the 	likely
consequences 	of 	relaxing 	the 	FCC 	ownership 	rules 	would 	be
defining 	all 	of 	the 	relevant 	product 	and 	geographic 	markets 	in
which broadcast 	television 	stations 	compete, 	then measuring
concentration 	in 	them. 	To 	do 	so 	rigorously would exceed 	our 	time
and 	resource 	constraints. 	However, 	available 	data 	on 	broadcast
television viewing 	behavior 	do 	allow 	us 	to 	make 	several 	general
observations 	relevant 	to 	market 	definition 	and 	concentration. 	We
examine Arbitron data 	on 	broadcast 	television 	ratings 	for 	the
most 	recent 	year 	for 	which we 	could obtain data, 	1988. 	Arbitron
computes, 	for 	each 	station, 	the 	share 	of 	"total 	households 	using
15

television" 	in 	the 	station's 	AD! 	tuned 	to 	that 	station. 41 	These
percentages 	can then be 	aggregated 	across 	all 	of 	the 	stations 	in
an AD!.
The 	sum of 	the 	audience 	shares 	of 	the 	stations 	in each AD!
is, 	on 	average, 	far 	less 	than 	100 	percent. 	The 	average 	is 	66
percent,42 with values 	ranging 	from 	a 	low of 	15 	percent 	to 	a
high 	of 	93 	percent. 	This 	means 	that many 	households 	using
television 	are 	not 	watching 	the 	broadcast 	stations 	in 	their AD!.
This 	shortfall 	results 	from 	some 	combination of 	(1) 	"out-of-
market" 	viewing 	(such 	as 	Washington-area 	households 	watching
Baltimore 	stations), 	and 	(2) 	cable 	programming viewing. 	Thus,
there 	are 	two 	implications 	relevant 	to 	market 	definition. 	First,
to 	the 	extent 	that 	cable-only viewing 	is 	substantial, 	the 	market-
definition process 	might 	have 	to 	consider 	advertising 	sold by
cable 	networks 	and 	cable 	systems 	in 	addition 	to 	advertising 	sold
by 	broadcast 	television 	stations. 43 Second, 	to 	the 	extent 	that
"out-of-market" 	viewing 	is 	substantial, 	the 	relevant 	geographic
market 	in which 	the 	typical 	television 	station competes 	might 	be
larger 	than 	the 	station's 	AD!.
41 For example, 	if 	a 	station has 	a 	10 percent 	share 	in a 	given
time 	slot, 	this 	means 	that 	10 	percent 	of 	the 	households 	viewing
television 	in 	that 	station's 	AD!, 	during 	that 	time 	period, 	were
tuned 	to 	that 	station.
42 The 	standard deviation 	is 	17 	percent.
43 Note that an apparent premise of the proposed re-institution
of 	cable 	"must-carry" 	rules, 	as 	well 	as 	of 	the 	radio-television
cross-ownership 	ban, 	is 	the 	existence 	of 	competition 	between
broadcast 	television 	and 	these 	other media.
16

Other evidence 	suggests 	how much viewers 	are 	turning 	to
cable-only viewing. 	The Arbitron data 	show 	that, 	on 	average,
broadcast 	stations 	experienced viewership 	declines 	among
households 	in 	their ADls 	between 	1984 	and 	1988. 44 Data 	in 	a
recent 	FCC 	staff 	report 	suggest 	that 	the 	growth of 	cable 	viewing
probably 	accounts 	for 	much of 	the 	decline 	in 	local 	broadcasters'
audience 	share. 	Of 	the 	over 	90 	percent 	of 	all 	households 	that
are 	passed by 	cable, 	over 	60 	percent 	subscribe. 45 In 	1989-90,
almost 	40 	percent 	of 	cable 	households' 	total 	television viewing
time 	was 	spent watching cable 	network 	programming, 	up 	from 	24
percent 	in 	1984-85. 46 Measured 	as 	a 	percentage 	of 	all
television 	viewing 	(i.e., 	cable 	households 	plus 	noncable
households), 	the 	share 	of 	cable 	networks 	in 	total 	viewing
increased 	from 	14 	percent 	in 	1984-85 	to 	26 	percent 	in 	1989-90. 47
The 	regression 	analysis 	in 	the 	Appendix 	also 	suggests 	that
cable 	television 	growth 	has 	been 	an 	important 	determinant 	of 	the
declines 	in 	broadcast 	shares. 	We 	estimate 	that 	every 	one
percentage 	point 	increase 	in 	the 	total 	number of 	households
passed by cable 	in 	an ADI 	was 	associated 	with 	a 	one-half
44 	On 	average, 	"in-market" 	stations 	lost 	about 	three 	percent
of 	total 	households 	using television in their ADI 	between 1984 	and
1988. 	The 	loss 	was 	largest 	in 	the 	top 	20 	ADls, 	where 	the 	decline
averaged over 	five 	percent.
45 See 	FCC 	staff 	"Television Overview," 	supra 	note 	39.
46 See Setzer and Levy, 	"Broadcast Television in a Multichannel
Marketplace," 	FCC Office of Plans 	and Policy Working Paper No. 	26,
June 	1991, 	Table 	6.
47 Id.
17

percentage 	point 	reduction 	in 	the 	viewer 	shares 	of 	the 	local
stations. 48
This 	information 	on changing viewing patterns 	does 	not, 	by
itself, 	demonstrate 	that 	broadcast 	television 	advertising 	is 	not
a 	relevant 	antitrust 	product market 	under 	the 	Merger Guidelines
analysis. 	Viewership 	shares 	do 	not 	translate 	directly 	into
advertising market 	shares. 	Despite 	the 	increase 	in 	cable
viewing, 	cable 	television 	still 	accounts 	for 	only a 	small 	share
of 	total 	video 	advertising; 	in 	1990, 	cable 	networks 	accounted 	for
about 	5 	percent 	of 	total 	television 	advertising, 	and 	non-network
cable 	advertising, 	about 	1.5 	percent. 49 Thus, 	determining 	the
proper 	product market 	definition may well 	require 	examining 	other
kinds 	of 	evidence, 	such 	as 	the 	extent 	to 	which 	advertisers 	would
move 	from 	one medium 	to 	another 	in 	response 	to 	a 	change 	in
relative 	advertising 	rates. 	But 	the 	general 	data 	on viewing
patterns 	do 	suggest 	that 	television 	stations 	may compete 	in
antitrust 	markets 	that 	are 	broader 	than 	their AD!, 	and 	that
perhaps 	include 	services 	other 	than 	those 	produced 	by 	broadcast
television 	stations.
Measuring more 	precisely 	the 	boundaries 	and 	levels 	of
concentration 	of 	properly defined markets 	is 	beyond 	the 	scope 	of
this 	comment. 	To 	illustrate 	the 	scope 	of 	the 	issues, 	we 	set 	out
a 	stylized concentration 	analysis 	based 	on 	general 	information
48 	We 	can 	reject 	at 	the 	one 	percent 	significance 	level 	the
hypothesis 	that 	cable 	growth 	was 	unrelated 	to 	broadcast 	share
reductions.
49 	See 	Setzer and Levy, 	supra 	note 	46.
18

about 	the 	number 	and 	size 	distribution 	(by 	audience 	share) 	of
commercial 	broadcast 	stations 	in 	the 	212 	Arbitron ADls, 	and
discuss 	the 	relationship 	these 	data might 	bear 	on more 	relevant
concentration measures 	of 	antitrust 	markets. 	This 	discussion,
based 	as 	it 	is 	on very 	general 	data, 	does 	not 	necessarily
describe 	the 	"average" 	antitrust 	product 	and 	geographic 	market,
nor 	does 	it 	provide 	all 	of 	the 	information 	necessary 	for
determining 	the 	particular 	concentration 	levels 	in 	relevant
product 	and 	geographic 	markets.
According 	to 	the 	1992 	Television 	& Cable 	Factbook, 	the
average 	number of 	commercial 	television 	stations 	in 	an ADI 	is
about 	five; 	in 	the 	largest 	100 	ADls, 	the 	average 	is 	seven, 	but
the 	numbers 	range 	widely, 	from 	a 	low 	of 	two 	to 	a 	high 	of 	17. 	If
it 	were 	assumed 	that 	a 	seven-station ADI 	was 	a 	relevant
geographic 	market 	and 	that 	"broadcast 	television 	advertising" 	is
a 	relevant 	product market, 	and 	also 	that 	each 	station 	is 	of 	equal
competitive 	significance, 	then 	the 	change 	in Herfindahl-Hirschman
concentration 	indices 	(HHls) 	due 	to 	a 	combination of 	stations 	in
such 	a 	"market" 	could exceed 	the 	Merger Guidelines' 	threshold
concentration 	levels. 5o But 	there 	are many 	reasons 	why 	this
50 The Herfindahl-Hirschman Index would be computed as 10000 IN,
where 	N is 	the 	number of 	stations. 	Thus 	a 	merger of 	two 	stations
in a 	seven station market would increase the HHI 	from about 1410 to
about 	1835. 	Markets 	with 	post-merger 	HHls 	in 	the 	1000 	to 	1800
range are considered 	"moderately concentrated," and those over 1800
"highly 	concentrated," 	by 	the 	Merger 	Guidelines. 	Cases 	such 	as
this 	hypothetical 	that 	are 	near 	the 	thresholds 	are 	considered 	to
present 	comparable 	issues. 	Mergers 	producing 	an 	increase 	in 	the
HHI 	of 	more 	than 	100 	points 	to 	a 	level 	between 	1000 	and 	1800
"potentially 	raise 	significant 	competitive 	concerns 	depending 	on
( continued ... )
19

stylized 	result might 	provide 	a 	misleading 	indication 	of 	the
competitive 	implications.
First, 	the 	Arbitron ADls 	from 	which 	these 	numbers 	are 	drawn
may 	not 	correspond 	to 	geographic 	antitrust 	markets, 	to 	the 	extent
that 	"out-of-market" 	viewing 	is 	substantial. 51 If 	it 	is,
concentration 	statistics 	computed 	from ADI 	data might 	incorrectly
suggest 	that 	certain combinations 	are 	competitively
objectionable. 	As 	a 	hypothetical 	demonstration, 	consider 	two
identical, 	geographically 	contiguous 	ADls, 	each with 	six 	equal-
sized 	stations, 	that 	together 	constitute 	a 	relevant 	geographic
market. 	Within 	each ADI, 	the 	HHI 	would 	be 	1,667, 	in 	the 	Merger
Guidelines' 	"moderately concentrated" 	range. 	If 	each ADI 	was
(wrongly) 	regarded 	as 	a 	separate 	relevant 	geographic 	market, 	a
merger 	between two 	stations 	in 	the 	same ADI 	would 	be 	thought 	to
raise 	significant 	competitive 	concerns. 	But 	the 	pre-merger HHI
corresponding to 	the 	correct, 	broader relevant 	geographic 	market
that 	included 	both ADls 	would 	be 	only 	883, 	and 	the 	post-merger
50 ( ••• continued)
the 	factors 	set 	forth 	in 	Sections 	2-5 	of 	the 	Guidelines." 	Merger
Guidelines, 	! 	1.51. 	For 	mergers 	producing 	a 	similar 	100 	point
increase 	to 	a 	post-acquisition 	level 	over 	1800, 	a 	likely 	increase
in market power is presumed, 	but the presumption can be overcome by
these 	other 	factors. 	Sections 	2-5 	of 	the 	Merger 	Guidelines 	deal
with 	competitive 	effects, 	entry, 	efficiencies, 	and 	failing 	firms.
51 On 	average, 	only 	66 	percent of 	households 	using 	television
are 	viewing 	the 	signals 	of 	broadcast 	stations 	located 	in 	the
viewers' 	ADI. 	And even though the average number of 	stations 	even
in 	the 	large 	ADls 	is 	seven, 	more 	than 	half 	of 	all 	households 	now
receive 	ten 	or more 	over-the-air 	signals, 	compared to 	six 	signals
in 	1975. 	NPRM 	(! 	3). 	In 	1964, 	when 	the 	contour 	overlap 	rule 	was
first 	adopted, 	only 	four 	percent 	of 	all 	households 	had 	access 	to
ten 	or more 	over-the-air 	signals 	(NPRM, 	! 	17).
20

HHI 	only 	972. 52 That 	would 	be 	in 	the 	Guidelines'
"unconcentrated" 	range, 	where 	an 	antitrust 	challenge 	is 	highly
unlikely.
Using ADI 	data 	to 	assess 	market 	concentration could, 	of
course, 	also 	lead 	to 	mistaken 	inferences 	of 	the 	opposite 	sort.
For 	example, 	Arbitron defines 	separate 	ADIs 	for 	Baltimore 	and
Washington, 	D.C., 	even 	though 	the 	Grade 	B 	(and 	often 	the 	Grade A
contours) 	of 	the 	stations 	in 	the 	two 	cities 	overlap. 	It 	may 	be
that, 	by 	applying 	the 	market 	definition methods 	of 	the 	Merger
Guidelines 	to 	television 	stations 	in 	the 	Baltimore-D.C. 	area, 	the
Baltimore 	and Washington 	areas 	would 	be 	found 	to 	constitute 	a
single 	antitrust 	market. 	That 	is, 	advertisers 	might 	perceive
Baltimore 	stations 	as 	good 	substitutes 	for 	Washington 	stations
(and vice 	versa) 	for 	conveying 	advertising messages 	to 	a
t ' 	1 	d' 	53	par 	1CU 	ar 	au 	1ence. 	If 	the 	two metropolitan 	areas 	comprised
a 	single 	relevant 	market, 	then 	to 	assume, 	wrongly, 	that 	the
Arbitron ADIs 	always 	constituted 	separate 	relevant 	geographic
markets 	might 	lead 	to 	the 	incorrect 	conclusion 	that 	a 	merger
between 	a 	Baltimore 	station and 	a 	Washington 	station would 	likely
be 	of 	no 	competitive 	consequence. 54
52 This 	is 	computed 	as 	{(2/12)2 	+ 	10*(1/12)2}*10,OOO.
53 	We 	emphasize 	that 	this 	is 	a 	hypothetical 	example; 	we 	have
not 	actually conducted 	this 	analysis 	for 	the 	Baltimore-Washington
area.
54 On the other hand, 	it might be that Washington and Baltimore
stations are poor substitutes for each other, 	notwithstanding their
contour overlaps. 	If so, 	a 	merger between stations 	in the separate
( continued ... )
21

Another, 	and perhaps 	more 	important, 	reason why 	the 	stylized
calculation 	is 	misleading 	is 	that, 	contrary to 	its 	assumption
that 	each 	station 	is 	of 	equal 	competitive 	importance, 	in 	reality
viewer 	shares 	vary widely 	across 	stations. 55 It 	might 	be
appropriate, 	therefore, 	to 	compute 	HHIs 	on 	the 	basis 	of 	stations'
viewer market 	shares, 	rather 	than on 	the 	basis 	of 	the 	number 	of
stations. 	The merger of 	two 	stations, 	each with 	a 	share 	of 	one
percent, 	would 	likely be viewed 	as 	far 	less 	objectionable 	than
the 	merger 	of 	two 	stations 	each with 	a 	share 	of 	20 	percent. 56
Although 	there 	are 	likely 	to 	be 	possible 	transactions 	that 	would
increase viewer-share 	concentration 	significantly, 	others 	would
h 	. 	l' 	1 	57	C 	ange 	~t very 	~tt e.
54 ( ••• continued)
markets 	would 	be 	blocked under 	the 	current 	FCC 	rules, 	even 	though
antitrust 	analysis 	might 	find 	the 	combination 	competitively
unobjectionable.
55 	In 	our 	data 	set, 	values 	range 	from 	less 	than 	1 	percent 	to
51 	percent.
56 	The 	change 	in 	the 	HHI 	that 	results 	from 	the 	merger 	of 	two
firms 	with 	respective 	market 	shares 	of 	Sl 	and 	S2 	is 	2*Sl*S2.
57 	In 	1988, 	86 	stations 	had 	viewer 	shares 	below 	one 	percent,
and another 293 	stations 	had shares between one and seven percent;
these 	379 	stations 	comprised 	almost 	37 	percent 	of 	the 	commercial
over-the-air 	stations. 	Assuming 	these 	were 	shares 	of 	antitrust
markets, 	a 	combination 	of 	two 	stations, 	each 	with 	a 	7 	percent
share, 	would 	increase 	the 	HHI 	by 	just 	under 	100 	points; 	in 	a
"moderately concentrated" market (with post-merger HHI below 1800) ,
the 	combination 	would 	be 	considered 	"unlikely 	to 	have 	adverse
competitive 	consequences 	and 	ordinarily 	require[s] 	no 	further
analysis." 	Merger 	Guidelines, 	! 	1.51. 	There 	were 	228 	stations
with 	shares 	below 	five 	percent. 	A combination 	of 	two 	stations,
each 	with 	a 	five 	percent 	share, 	would 	increase 	the 	HHI 	by 	50
points; 	even 	in markets 	that 	are 	considered 	"highly concentrated"
(with post-merger HHI above 1800), 	mergers producing an increase in
(continued ... )
22

Finally, 	the 	significance 	of 	the 	figures 	would 	change
substantially 	if 	"broadcast 	television 	advertising" 	did 	not 	by
itself 	constitute 	a 	properly defined 	antitrust 	product market.
It 	may be 	that, 	after 	consideration of 	more 	detailed evidence,
the 	antitrust 	market 	would 	be 	found 	to 	include 	other 	services.
Other 	services 	whose 	competitive 	importance might 	be 	considered
would 	include 	advertising 	sold by cable 	networks 	and 	cable
systems, 	radio 	stations, 	or 	even 	print media. 	To 	the 	extent 	that
the 	market 	is 	found 	to 	be 	broader 	than 	broadcast 	television,
concentration 	figures 	based 	solely on ADI 	viewer 	shares 	will 	be
inaccurate.
Defining markets 	and measuring 	concentration 	are 	the 	first
steps 	in 	the 	Merger Guidelines 	analysis. 	If 	properly measured
concentration 	in properly 	defined markets 	exceeds 	the 	threshold
levels, 	the 	Merger Guidelines 	call 	next 	for 	analysis 	of 	other
market 	factors 	that 	pertain to 	competitive 	effects, 	that 	is, 	that
might 	contribute 	to 	a 	lessening 	of 	competition through
coordinated 	interaction. 58 Little 	can be 	said 	from 	direct
57 ( ••• continued)
the 	HHI 	of 	less 	than 	50 	points 	are 	considered 	unlikely 	to 	raise
significant competitive concerns. 	Id. 	Of course, 	the HHI 	increase
could 	be 	substantially 	greater 	for 	the 	combination 	of 	a 	small
station and a 	larger one. 	We do not know in how many markets 	there
might be two or more very small stations whose combination would be
unlikely 	to 	increase 	concentration significantly.
58 Merger 	Guidelines, 	S 	2. 	Antitrust 	analysis 	is 	also
concerned about possible anticompetitive unilateral conduct. 	This
discussion focuses on coordination, rather than unilateral conduct,
because 	available 	information 	suggests 	that 	broadcasting 	is
unlikely 	to 	be 	differentiated 	enough 	to 	implicate 	the 	Merger
Guidelines 	analysis of conditions 	that give rise 	to concerns 	about
unilateral 	conduct.
23

experience 	about 	how 	television 	station mergers 	might 	be 	related
to 	the 	likelihood of 	anticompetitive 	coordinated 	interaction,
because 	the 	contour overlap 	rule 	has 	prohibited 	such
transactions. 	One 	study 	attempted 	to 	examine 	the 	relationship
between 	television 	industry concentration 	(based on ADI 	data) 	and
advertising 	prices,59 	and 	found 	no 	correlation between
advertising 	prices 	and ADI 	concentration. 	These 	findings 	are
consistent 	with 	the 	possibility described 	above, 	that 	television
broadcasting 	in 	an ADI 	is 	not 	a 	well-defined 	antitrust 	market 	in
the 	first 	place. 	But 	if 	a 	study based on 	well-defined 	product
and 	geographic 	markets 	produced 	results 	like 	these, 	it 	would
confirm why, 	under 	the 	Merger Guidelines 	approach, 	measuring
concentration 	levels 	is 	only the 	first 	step 	in 	the 	analysis. 	The
Merger Guidelines 	contemplate 	further 	examination of 	the
characteristics 	of 	the 	market 	to 	identify and 	assess 	other
factors 	affecting competition.
The 	next 	step 	in 	the 	Merger Guidelines 	analysis, 	if
concentration 	increases 	and 	other 	characteristics 	of 	defined
59 	See 	Fournier 	and 	Martin, 	Does 	Government-Restricted 	Entry
Produce Market Power?: 	New Evidence From the Market for Television
Advertising, 	14 	Bell 	J. 	Econ. 	44-56 	(1983). 	Fournier 	and 	Martin
examined the relationship between measures of market concentration
and 	actual 	transaction 	prices 	for 	spot 	television 	advertising.
According to estimates obtained by one statistical method, 	ordinary
least squares, 	advertising prices were not related to some measures
of 	concentration 	(e.g., 	the 	HHI), 	and 	were 	negatively 	related 	to
others 	(e. g., 	the 	two-firm 	concentration 	ratio). 	According 	to
estimates obtained using a 	preferred statistical method 	(two-stage
least 	squares, 	to 	control 	for 	the 	possible 	endogeneity 	of 	market
concentration), 	none 	of 	the 	measures 	of 	concentration exhibited 	a
statistically 	significant 	relationship 	to 	price. 	Fournier 	and
Martin defined each 	"market" 	as 	an 	Arbitron ADI.
24

antitrust 	markets 	imply 	that 	a 	combination 	is 	likely 	to 	reduce
competition, 	is 	determining 	the 	likelihood of 	entry. 	The 	number
of 	commercial 	television 	stations 	has 	increased by over 	28
percent 	since 	1985, 	nearly 	55 	percent 	since 	1980 	(from 	734 	in
1980 	to 	1,136 	in 	1992), 	and 	by 	over 	100 	percent 	since 	the 	rule
was 	first 	adopted 	in 	1964. 60 Whether 	there 	are 	enough
unassigned broadcast 	frequencies 	to 	permit 	further 	entry by 	new
stations, 	and 	if 	so, 	whether 	entry would 	be 	"timely, 	likely, 	and
sufficient" 	to 	counteract 	potential 	competitive 	problems, 	cannot
be 	determined. 	The 	answers 	to 	these 	questions 	would 	differ
across 	geographic 	markets. 61 To 	the 	extent 	that 	entry would 	be
"timely, 	likely, 	and 	sufficient," 	the 	competitive 	concerns
surrounding 	a 	television 	station merger 	would 	be 	substantially
diminished, 	even 	in 	a 	market 	where 	post-merger 	concentration
would 	be 	high.
Next, 	the 	Merger Guidelines 	analysis 	assesses 	merger-
specific 	efficiencies. 	(In 	the 	case-by-case 	Merger Guidelines
analysis, 	the 	parties 	must 	demonstrate 	these 	efficiencies, 	and
show 	that 	they cannot 	reasonably be 	achieved 	except 	through 	the
merger.) 	Here 	again, 	there 	is 	no 	direct 	evidence 	about
efficiencies 	from 	the 	merger 	of 	nearby 	television 	stations,
60 	See 	FCC 	Staff 	"Television 	Overview," 	supra 	note 	39,
attachment 	l.
61 	It 	may 	be 	that 	the 	regulatory 	process 	governing 	broadcast
licenses, 	as 	currently 	applied, 	would 	delay 	all 	entry 	beyond 	the
point that antitrust analysis would consider sufficiently "timely."
If 	so, 	a 	response 	suggested 	by 	a 	competition 	policy 	perspective
would 	be 	to 	modify 	the 	regulatory 	process 	so 	as 	to 	speed 	the 	rate
at 	which 	new 	licenses 	can be 	issued.
25

because 	the 	rules 	have 	prohibited 	such 	transactions. 	As 	we
described 	in § 	IV, 	supra, 	however, 	research 	has 	suggested 	that
there 	are 	efficiencies 	associated with 	common 	ownership 	of 	local
radio 	stations. 	There may be 	similar 	efficiencies 	that 	could be
realized 	through 	joint 	ownership 	of 	local 	television 	stations.
The 	final 	step 	is 	assessing 	the 	possibility 	that 	the 	merger
will 	prevent 	the 	imminent 	failure 	of 	one 	of 	the 	merging 	parties.
That 	will 	of 	course 	depend on 	the 	details 	of 	particular parties'
situations.
Overall, 	our 	analysis 	suggests 	that 	there may 	be 	potential
transactions 	now barred by 	the 	contour-overlap 	rule 	that 	might 	be
viewed 	as 	competitively unobjectionable 	when 	judged by 	the
standards 	of 	the 	Merger Guidelines. 	Unless 	potential
efficiencies 	from 	such 	transactions 	are 	small, 	or 	the 	cost 	of 	a
different 	regulatory 	approach 	is 	too 	high, 	it 	appears 	consistent
with 	a 	competition policy 	analysis 	of 	the 	issue 	to 	relax 	the
television 	"duopoly" 	rule.
One way 	to 	relax 	the 	rule 	is 	to 	change 	the 	standards 	of 	the
existing 	rules 	to 	permit 	some common ownership 	of 	nearby
television 	stations. 	The 	proposals 62 include 	changing the
signal 	contour used 	for 	determining 	prohibited 	overlaps 	from 	the
Grade 	B to 	the 	Grade A; 	allowing combinations involving only UHF
stations; 	and 	allowing 	combinations 	where 	a 	certain minimum
number of 	independent 	stations 	would 	remain 	after 	the
combination. 	Each of 	these 	approaches 	offers 	the 	advantage 	of
62 NPRM 	(!1[ 	17 -2 0) •
26

low 	enforcement 	costs, 	but 	they 	entail 	two 	risks: 	first, 	that
potentially efficient 	combinations 	could 	still 	be 	thwarted, 	and
second, 	that 	potentially troublesome 	combinations 	could be
permitted.
As 	an 	alternative,63 	the 	might 	FCC 	consider 	adopting 	a
case-by-case 	approach, 	combined with 	a 	"safe 	harbor" 	for
transactions 	that 	satisfy certain market 	structure 	criteria.
This 	approach would be 	similar 	to 	that 	embodied 	in 	the 	Merger
Guidelines, 	and 	to 	what 	the 	FCC 	is 	proposing here 	for 	certain
transactions 	subject 	to 	the 	radio-television cross-ownership
rule. 64 In considering requests 	for 	waiver of 	the 	radio-
television 	cross-ownership 	rule, 	the 	FCC 	takes 	into 	account 	the
potential 	benefits 	of 	the 	combination, 	the 	types 	of 	stations
involved, 	the 	number of 	stations 	already 	owned by 	the 	applicant,
the 	financial 	difficulties 	of 	the 	station(s), 	and 	the 	competitive
nature 	of 	the 	market. 65 Generally 	speaking, 	these 	are 	similar
to 	the 	elements 	of 	the 	Merger Guidelines' 	analysis. 	The 	FCC 	may
wish 	to 	consider whether 	it would be 	appropriate 	to 	apply 	these
(or 	similar) 	criteria 	to 	television 	licensure 	issues.
63 This 	alternative was 	not 	proposed 	in 	the 	NPRM.
64 See 	§ 	VII, 	infra.
65 NPRM, 	! 	22.
27

2. 	Programming Variety
As 	in 	the 	Radio 	Rules 	and 	Policies 	rulemaking,66 	an
important 	issue 	in 	this 	proceeding 	is 	the 	impact 	of common
h · 	( 	f 	t) 	. 	t 	67	owners 	1p on program 	or 	orma 	var1e 	y. 	Because 	television
stations 	earn 	profits 	by 	selling 	time 	to 	advertisers, 	rather 	than
by 	selling programming 	directly 	to 	viewers, 	the 	relationship
among market 	structure, 	programming variety, 	and economic 	welfare
is 	not 	simple. 	Unlike 	ordinary 	goods 	and 	services, 	which 	are
sold 	to 	consumers 	at 	prices 	that 	reflect 	consumers' 	valuation 	of
these 	goods, 	broadcasters 	provide 	programs 	to 	viewers 	free 	of
charge. 	Broadcasters' 	costs 	are 	covered by 	selling 	air 	time 	to
advertisers, 	who 	care 	principally about 	audience 	size. 68
Consequently, 	broadcasters' 	profit-maximizing programming 	choices
will 	be 	determined 	by 	the 	ability 	to 	"sell" 	a 	large 	audience 	to
an 	advertiser, 	rather 	than 	by 	the 	ability 	to 	sell 	a 	highly-valued
program 	to 	an 	audience. 	Hence, 	the mix of 	programs 	that 	emerges
under 	advertiser 	support may 	differ 	from 	that 	which would 	emerge
66 	Supra, 	note 	8.
67 	As 	noted 	earlier 	(see 	note 	2, 	supra), 	we 	address 	here 	the
relationship 	between 	market 	structure 	and 	the 	degree 	of
differentiation 	in 	types 	of 	programming 	offered, 	not 	issues
relating to diversity in the range of editorial viewpoints offered.
68 This 	is 	an 	oversimplification, 	since 	advertisers 	also 	care
about 	audience 	demographics 	(See 	Setzer 	and 	Levy, 	"Broadcast
Television in a 	Multichannel Marketplace," 	FCC Office of Plans 	and
Policy Working 	Paper No. 	26, 	June 	1991, 	p. 	117). 	To 	simplify 	the
discussion, 	however, 	we 	will 	ignore 	this 	complication.
28

under 	a 	system of 	direct 	viewer payments. 69 The 	divergence
would depend 	on, 	among 	other 	things, 	the 	structure 	of 	consumer
preferences, 	whether or 	not 	channel 	capacity 	is 	limited, 	and 	the
costs 	of 	producing 	different 	types 	of 	programming.
The 	advertiser-supported 	feature 	of 	the 	system makes 	it
difficult 	to 	predict 	the 	relationship 	between market 	structure
and program variety. 	It 	is 	possible, 	for 	example, 	to 	describe
circumstances 	under which 	a 	monopolist 	provides 	a 	different
variety of 	programs, 	and will 	provide 	more 	programs 	for 	viewers
with 	specialized 	tastes, 	than would 	be 	provided 	in 	a
competitively-structured market. 70 Competing 	stations 	might 	try
69 See
Monopolistic
(1977).
Spence 	and
Competition, Bruce 	Owen,
and 	Welfare,
Television
91 	Q. 	J.
Programming,
Econ. 	103-26
70 	See 	Steiner, 	Program 	Patterns 	and 	Preferences, 	and 	the
Workability 	of 	Competition 	in 	Radio 	Broadcasting, 	66 	Q. 	J. 	Econ.
194-223 	(1954), 	for 	examples. 	For 	monopoly 	to 	provide 	greater
program variety than a competitive market, 	channel capacity must be
limited. 	This 	result 	also requires 	specific assumptions 	about the
distribution of consumer preferences and specific assumptions about
whether viewers switch to a 	less preferred station when their first
choice is unavailable, 	or, 	instead, 	choose to forgo viewing. 	Under
this 	condition, 	monopoly 	variety 	could 	exceed 	the 	competitive
variety 	because 	the 	monopolist 	would 	not 	have 	an 	incentive 	to
provide 	duplicative 	programming on different 	stations, 	since 	this
would 	simply 	shift 	viewers 	from 	one 	of 	its 	stations 	to 	another.
Rather, 	the 	monopolist would 	offer different 	programs 	directed 	to
different 	audiences 	(as 	long 	as 	each 	program 	offered 	positive
profits). 	With 	competing 	stations, 	any 	given 	station 	might 	find
that 	it 	can 	garner 	its 	largest 	audience 	by 	offering 	a 	program
similar 	to 	that 	of 	its 	rival, 	rather 	than 	by 	offering 	a 	program
that 	is 	attractive 	to 	another 	(but 	smaller) 	group. 	With 	limited
frequency capacity, 	this could result in less program variety under
competition 	than 	under 	monopoly, 	and 	the 	displacement 	of
programming that 	caters 	to 	the 	tastes 	of 	smaller 	groups.
As 	noted 	earlier, 	there 	has 	been 	substantial 	growth 	in 	the
number 	of 	broadcast 	television 	channels 	over 	time. 	Also, 	cable
channel 	capacity has 	also 	increased 	substantially. 	Nevertheless,
( continued ... )
29

to 	reach 	the 	same 	core 	audience, 	leaving others 	unserved, 	while 	a
single 	owner might 	try 	to 	program different 	stations 	to 	appeal 	to
different 	audience 	segments 	in order to 	maximize 	its 	total
audience 	size. 	This 	result, 	of 	monopoly 	producing more 	variety,
is 	more 	likely, 	the 	greater 	the 	constraints 	on 	channel 	capacity.
If 	there 	is 	excess 	capacity, 	then 	some 	program 	duplication 	is
likely, 	because 	an 	entrant 	might 	still 	find 	it 	profit-maximizing
to 	"steal" 	part 	of 	an 	incumbent's 	audience; 	however, 	complete
displacement 	of 	specialized programming 	is 	unlikely, 	because
specialized programming would 	be 	offered 	as 	long as 	some
broadcaster 	could 	profit 	from 	it.
The 	possibility 	that 	monopoly 	could 	provide more 	variety
than 	competition 	is 	only 	an 	illustration; 	theory 	does 	not
establish which market 	structure 	would 	provide 	the 	greatest
programming variety. 	It 	does 	demonstrate, 	however, 	that 	there 	is
no 	theoretical 	presumption 	that 	maximizing 	the 	dispersion 	of
station ownership will 	maximize 	program variety. 	Moreover, 	the
illustration compares 	a 	monopoly market 	structure 	and 	a
competitive 	structure. 	The 	NPRM 	does 	not 	propose, 	nor 	do 	we
advocate, 	that 	the 	FCC 	encourage monopolization 	of 	local
70 ( ••• continued)
it 	appears 	that 	total 	(cable 	plus 	broadcast) 	available 	channel
capacity 	in 	the 	typical 	market 	will 	be 	exceeded 	by 	the 	number 	of
available 	programming 	services, 	especially when 	cable 	programming
services 	are 	taken 	into 	account. 	For 	example, 	according 	to 	the
1990 FCC Cable Report, 	in 1989 	the number of domestic 	existing and
proposed 	pay 	TV 	and 	satellite 	cable 	services 	was 	reportedly 	181.
In 1989, 	the average cable system had about 40 channels. 	Report 	in
the 	Matter of 	Competition, 	Rate 	Deregulation and 	the 	Commission's
Policies 	Relating to the Provision of Cable Television Service, 	MM
Docket No. 	89-600, 	July 	31, 	1990, 	! 	43 	and 	Table 	6.
30

television markets. 	Rather, 	under 	consideration 	is 	a 	modest
relaxation of 	the 	regulations 	that 	now mandate 	ownership
fragmentation. 	Economic 	theory 	provides 	no 	basis 	for 	concluding
that 	consolidations 	that 	might 	follow 	this 	relaxation would
reduce 	program variety.
Lastly, 	and perhaps 	most 	importantly, 	the 	models 	that
underlie 	the 	discussion 	above 	ignore 	costs. 	In 	these models, 	the
only 	constraint 	on 	adding 	formats 	(that 	is, 	stations) 	is
frequency 	availability. 	In 	reality, 	of 	course, 	costs 	are 	a 	very
important 	determinant 	of 	whether 	a 	service 	can be 	offered
profitably. 	Costs 	may be much more 	important 	than 	frequency
scarcity 	in 	determining 	program variety. 	If 	relaxation 	of 	the
common ownership 	rules 	permits 	the 	attainment 	of 	greater
efficiencies, 	and 	if 	the 	prospect 	of 	these 	efficiencies, 	in 	turn,
encourages 	the 	construction 	of 	additional 	stations 	(or 	prevents
the 	exit 	of 	otherwise unprofitable 	stations), 	it 	is 	likely 	that
program variety will 	increase 	(or 	that 	the 	rate 	of 	decline 	will
be 	reduced).
VI. 	National 	Ownership Limits
As 	an 	alternative 	to 	the 	current 	12 	station-25 	percent 	reach
limitation on 	the 	number 	of 	television 	stations 	that may 	be
commonly owned, 	the 	NPRM 	suggests 	raising 	both the 	numerical
limit 	and 	the 	national 	reach 	limit. 7l
71 	NPRM, 	, 	12. 	The 	FCC 	offers 	two 	specific 	proposals: 	(1)
increasing the numerical cap to 20 	or 24 	stations while 	increasing
(continued ... )
31

Conceptually, 	it 	does 	not 	appear 	that 	the 	proposed 	limits 	on
national 	station ownership would 	address 	any well-defined
concerns 	about 	the 	possible 	exercise 	of market 	power by 	broadcast
television 	stations. 	Competition 	among 	television broadcasters,
whether 	as 	competition 	for 	viewers 	or 	as 	competition 	in
advertising markets, 	occurs 	principally on 	a 	local, 	rather 	than
national, 	level. 	The 	local 	scope 	of 	competition 	suggests 	that 	an
economic 	assessment 	of 	same-service combinations 	should examine
their 	effects 	in 	local 	antitrust 	markets. 	A combination of
stations 	that 	individually 	have 	modest 	shares 	of 	their 	respective
local 	advertising markets 	could, 	because 	those 	markets 	are
populous, 	exceed 	the 	national 	viewer 	"reach" 	threshold. 	It 	is
unclear why 	formation 	of 	such 	a 	group would 	threaten 	competition
in 	local 	television markets.
Moreover, 	as 	we 	observed 	in 	our Reply Comments 	to 	the 	Radio
Rules 	and 	Policies 	NPRM, 	audience 	"reach" 	is 	conceptually
different 	from 	audience 	share. 72 The 	proposed 	threshold 	of 	35
percent might 	raise 	antitrust 	concerns 	if 	it 	applied 	to 	shares 	of
an 	antitrust 	market. 	A market 	of 	approximately 	three 	equal-sized
71 ( ••• continued)
the 	audience 	reach 	limit 	to 	35 	percent, 	or 	(2) 	increasing 	the
numerical 	cap 	to 	18 	stations 	and 	the 	audience 	reach 	limit 	to 	30
percent. 	It 	has 	also 	asked 	for 	comment 	on 	whether 	the 	numerical
cap 	alone 	should 	be 	increased 	while 	retaining 	the 	25 	percent
audience 	limit.
72 	The 	national 	multiple 	ownership 	rule 	defines 	"national
audience 	reach" 	as 	"the 	total 	number 	of 	television 	households 	in
the Arbitron area-of-dominant-influence 	(ADI) 	markets 	in which the
relevant 	stations 	are 	located, 	divided 	by 	the 	total 	national
television 	households 	as 	measured 	by 	the 	ADI 	data." 	47 	C.F.R. 	§
73.3555 	(d) (3) (i).
32

firms 	would 	exceed 	the 	"highly concentrated" 	concentration 	level,
and 	an 	acquisition creating 	such 	a 	market 	structure would 	receive
1 . 	73
C 	ose 	scrut1ny. 	But 	audience 	reach 	and 	audience 	share 	are
quite 	different 	concepts, 	so 	an 	audience 	reach 	of 	35 	percent
would 	not 	necessarily correspond 	to 	an 	audience 	share 	of 	35
percent. 	Depending on details 	of 	population 	distribution 	and
spectrum 	allocation, 	it 	is 	conceivable 	that many more 	than 	four
different 	ownership 	groups, 	perhaps 	even 	dozens 	of 	them, 	could be
assembled 	that 	would each 	reach 	35 	percent 	of 	the 	national
audience, 	even 	though 	concentration 	in 	separate 	local 	markets
would 	remain 	low. 	What 	would 	happen 	in 	fact 	is 	unknown, 	but 	the
audience 	reach proposal 	as 	drafted does 	not 	describe 	a 	situation
that 	is 	readily 	understandable 	as 	raising 	a 	concern 	about
competition 	or efficiency.
The 	local 	scope 	of 	competition among 	television 	stations
also 	attenuates 	competitive concerns 	about 	creation 	of 	monopsony
power 	in program 	acquisition 	if 	group ownership 	rules 	are
relaxed. 	A program 	supplier 	looking 	for 	broadcast 	distribution
in 	the 	New York 	and Los 	Angeles 	areas 	will 	care 	about 	the 	degree
of 	competition to 	buy 	programs 	among 	stations 	within each 	of
these 	two 	markets, 	not 	between New York buyers 	and Los 	Angeles
buyers. 74 A unilateral 	attempt 	by 	a 	New York 	station to 	reduce
73 	Department 	of 	Justice 	and 	Federal 	Trade 	Commission 	Merger
Guidelines, 	! 	1.51(C).
74 	We 	are 	thus 	assuming 	that 	geographic 	markets 	for 	sales 	of
programming 	to 	individual 	stations 	are 	local. 	Even 	if 	this 	is
incorrect, 	the 	level 	of 	concentration 	in 	a 	national 	market 	is 	far
(continued ..• )
33

the 	price 	below the 	competitive 	level 	will 	be 	defeated 	if 	there
are 	other competing bidders 	for 	the 	New York 	area 	rights 	to 	the
program. 	Competition among 	New York 	area 	stations 	would be
unaffected 	if 	the 	owner 	of 	one 	of 	them 	also 	owned 	a 	Los 	Angeles
station. 	Further, 	any 	attempt 	by 	an 	owner 	of 	stations 	in 	each
area 	to 	pay 	less 	than 	the 	competitive 	price 	for 	the 	rights 	in
both 	areas 	combined would be 	defeated 	if 	competing 	sources 	of
distribution 	services 	existed 	in each 	area. 75 Even 	if 	competing
sources 	of 	distribution were 	lacking--for 	example, 	if 	the 	Los
Angeles 	station did 	not 	face 	effective 	local 	competition--the
degree 	of 	market 	power 	possessed would 	be 	unaffected 	by 	the 	fact
of 	joint 	ownership with 	a 	station 	in 	another 	area. 76
74 ( ••• continued)
too low to warrant the continued existence of the national multiple
ownership 	rules 	in 	their 	current 	form. 	As 	noted 	earlier, 	the
national 	HHI 	for 	the 	television 	industry 	ranges 	from 	187 	to 	229,
depending 	upon 	which 	measure 	of 	station 	output 	is 	employed. 	See
"Television Overview," 	supra 	note 	39, 	at 	p. 	4.
75 There 	is 	little 	empirical 	support 	for 	the 	proposition that
group 	stations 	pay 	less 	than 	nongroup 	stations 	for 	programming.
The FCC's 	Network Inquiry Special Staff Report 	analyzed the prices
paid 	for 	programming 	by 	group 	and 	non group 	station 	owners.
Controlling for 	other factors 	(including the degree of competition
for 	programs), 	the 	staff 	found 	that 	the 	price 	paid 	per 	viewer-
minute 	was 	higher, 	not 	lower, 	when 	the 	purchaser 	was 	owned 	by 	a
large group. 	See Besen and Johnson, 	"Regulation of Media Ownership
by the Federal Communications Commission," 	RAND Corporation Report
No. 	R-3206-MF, 	December 	1984, 	pp. 	15-16.
76 Besen and Johnson, 	supra note 75, 	at 18-19, 	provide a 	useful
example to illustrate the theoretical principles that lead to these
results. 	Consider a 	group 	owner, 	with 	stations 	in 	three 	separate
television 	markets, 	bidding 	for 	a 	program 	to 	be 	shown 	in 	each
market. 	Suppose 	this 	group owner 	is 	willing 	to 	pay up 	to 	$100 	in
market 	1, 	$50 	in 	market 	2, 	and 	$20 	in 	market 	3, 	while 	some 	rival
station 	in each market would pay 	$90, 	$40, 	and 	$30, 	respectively.
If 	the 	program 	can 	be 	sold 	to 	the 	highest 	valued 	user 	in 	each
(continued ... )
34

Concern 	about 	local 	television 	stations' 	exercise 	of 	either
monopoly 	power 	in 	advertising or monopsony 	power over programs 	is
likely 	only 	in connection with 	joint 	ownership of 	local 	stations,
the 	issues 	discussed 	in 	Section V, 	supra. 	Relaxing 	the 	national
ownership 	limits 	would 	not 	appear 	to 	pose 	a 	threat 	to
competition.
VII. 	The Television-Radio Cross-ownership Rule
The 	FCC's 	rules 	prohibit 	holding 	attributable 	ownership
interests 	in both a 	radio 	station 	and 	a 	television 	station 	in 	the
same 	market. 77 Since 	1989, 	however, 	the 	FCC 	has 	entertained
waivers 	of 	this 	rule 	for 	combinations 	in 	a 	top 	25 	television
76 ( ••• continued)
market, 	the 	group 	owner 	would 	acquire 	the 	exhibition 	rights 	in
markets 	1 and 2 	(paying slightly more than $90 + $40 = $130), 	while
a 	rival would acquire the rights 	for market 3 	(paying slightly more
than 	$20). 	The 	program 	producer's 	revenues 	will 	be 	slightly more
than 	$150, 	the 	group 	owner's 	surplus 	will 	be 	just 	under 	$20, 	and
the 	rival 	station's 	surplus will 	be 	just under 	$10. 	Total 	surplus
will 	thus 	be 	just under 	$180.
Suppose, 	instead, 	that 	the 	group 	owner 	tried 	to 	acquire 	the
exhibition 	rights 	for 	all 	three 	markets, 	without 	outbidding 	its
rivals 	in market 3. 	If the effort succeeded, 	the total surplus, 	to
be divided between the producer and the group owner, 	would be 	$170
(= 	$100 	+$50 	+ 20). 	The 	group owner's 	surplus would be 	$170 	- 	P,
where $P is the amount the group owner would pay. 	But the producer
could 	now 	offer a 	deal 	that 	would 	increase 	total 	surplus 	and make
everyone 	individually better 	off, 	including the 	group 	owner. 	The
producer 	could 	buy 	back the 	rights 	to 	market 	3 	for 	($20 	+ 	d) 	and
sell 	them 	to 	a 	another 	station 	there 	for 	($20 	+ 	e), 	where 	e 	> 	d.
The group owner's 	surplus would 	increase by d, 	the rival 	station's
surplus 	would 	increase 	by 	$10 	- 	e 	(= 	$30 	- 	[$20 	+ 	el), 	the
producer's 	surplus 	would 	increase 	by 	e 	- 	d, 	total 	surplus 	would
increase by 	$10, 	and allocative efficiency would be restored. 	The
program 	would 	be 	allocated 	to 	its 	highest 	valued 	user 	in 	each
market.
77 	Section 47 	C.F.R. 	§ 	73.3555(b).
35

market, 	if 	30 	separately owned, 	operated, 	and 	controlled
broadcast 	licensees 	would 	remain 	after 	the 	combination. 78 The
FCC 	also 	considers 	requests 	for 	waivers 	if 	the 	request 	involves 	a
"failed" 	station. 	Other 	requests 	for 	waivers 	are 	considered
according 	to 	criteria 	such 	as 	potential 	benefits 	of 	the
combination, 	the 	types 	of 	stations 	involved, 	and 	the 	competitive
nature 	of 	the 	market. 79
From 	the 	perspective 	of 	competition policy, 	"cross-
ownership" 	could cause 	problems 	if 	television 	and 	radio 	stations
compete 	in 	the 	same 	antitrust 	market. 	But 	concentration 	in 	a
market 	defined 	so 	broadly will 	often be 	far 	too 	low 	for
combinations 	to 	raise 	substantial 	antitrust 	concerns. 	Even 	in
small 	markets, 	audiences 	can 	receive many 	over-the-air 	television
and 	radio 	signals. 	In our Reply 	Comments 	to 	the 	Radio 	Rules 	and
Policies 	NPRM we 	observed 	that 	local 	radio 	ownership 	is 	typically
not 	highly 	concentrated. 8o Only 	about 	14 	percent 	of 	radio
78 According 	to 	the 	NPRM 	(! 	22), 	it 	is 	official 	FCC 	policy 	to
"look 	favorably" 	upon 	waiver 	requests 	when 	these 	conditions 	are
satisfied.
79 	See 	NPRM, 	! 	22.
80 	The 	National 	Association 	of 	Broadcasters 	noted 	in 	its
comments 	in 	the 	1987 	Radio 	Rules 	and 	Policies 	proceeding 	that 	in
47.9 	percent 	of 	local 	broadcasting 	markets, 	Herfindahl-Hirschman
indexes 	(HHIS) 	were below 1,000, 	based on a 	market definition that
includes 	only 	AM 	and 	FM 	radio 	stations. 	Under 	the 	Department 	of
Justice 	and 	Federal 	Trade 	Commission 	Merger 	Guidelines, 	such
markets 	are 	considered 	"unconcentrated." 	In 	most 	other 	local
markets 	(38.6 	percent), 	the 	HHIs 	were 	between 	1,000 	and 	1,800, 	a
range 	of 	values 	that 	the 	Merger 	Guidelines 	consider 	"moderately
concentrated". 	(See 	"An 	Updated 	Examination 	of 	Market
Concentration 	in 	Radio 	Markets," 	filed 	as 	Appendix 	E 	to 	the 	1987
Comments 	of 	the 	NAB.) 	The 	NAB 	also 	noted 	that 	a 	variety 	of 	other
(continued ... )
36

markets 	appear 	to 	be 	in 	the 	"highly concentrated" 	range 	(about 	35
out 	of 	259 	markets). 	We 	have 	not 	estimated concentration
statistics 	for 	combined 	"radio 	and 	television" 	markets, 	but 	we
believe 	that 	concentration 	in 	combined markets 	may well 	be 	lower.
Unless 	audiences 	are much 	larger 	and 	concentration much 	greater
for 	television 	than 	for 	radio, 	concentration 	in combined markets
would 	likely be 	lower 	than 	concentration 	calculated 	for 	"radio
only" 	markets. 81
One 	of 	the 	FCC's 	proposals 	is 	to 	codify 	the 	present 	top-25
market 	waiver 	criteria 	and extend 	their 	application 	to 	all
markets 	where 	30 	independent voices 	would 	remain 	after 	the
combination. 82 	In essence, 	the 	FCC 	would be 	creating a 	"safe
harbor" 	for 	certain types 	of 	transactions. 	A 	"safe 	harbor"
approach would 	resemble 	in principle 	the 	Merger Guidelines'
approach. 	Under 	the 	government's 	merger 	enforcement 	standards,
mergers 	in 	"unconcentrated" 	markets 	(where 	the 	post-merger HHI 	is
below 	1,000) 	are 	considered unlikely 	to 	have 	adverse 	competitive
consequences 	and 	ordinarily do 	not 	require 	further 	analysis.
This 	threshold 	applies 	to 	any 	antitrust 	market, 	regardless 	of
absolute 	size.
80 ( ••• continued)
media 	serve 	local 	markets, 	and 	that 	in 	many 	cases 	the 	number 	of
such outlets 	is 	considerable. 	(See 	"An Analysis 	of 	Media 	Outlets
by 	Market", 	filed 	as 	Appendix 	B 	to 	the 	Comments 	of 	the 	National
Association of 	Broadcasters, 	MM 	No. 	91-140, 	August 	5, 	1991.)
81 	In addition, 	new 	station entry, 	where 	technically 	feasible
and authorized, 	would also tend to constrain any attempted exercise
of 	market 	power.
82 NPRM, 	! 	28.
37

As 	the 	FCC 	suggests, 	delineating 	appropriate 	"safe 	harbors"
can produce 	social 	benefits. 	Enforcement 	resources 	can be
concentrated where 	anticompetitive 	activities 	are 	most 	likely.
Costs 	of 	compliance 	are 	reduced 	too, 	when 	regulated 	firms 	have
clearer 	guidance 	on enforcement 	standards 	and procedures.
Consequently, 	there 	is 	likely considerable merit 	in 	the 	FCC's
proposal.
Recommending 	a 	regulatory 	"safe 	harbor" 	does 	not 	necessarily
imply that 	all 	transactions 	failing 	the 	"safe 	harbor" 	criteria
would 	be 	objectionable. 	For example, 	when 	enforcing the
antitrust 	laws, 	the 	FTC 	and DOJ may decline 	to 	challenge mergers
occurring 	in 	"moderately concentrated" 	markets 	(where 	the 	post-
merger HHI 	is 	between 	1,000 	and 	1,800) 	because 	further 	analysis
reveals 	that 	the 	transaction would be 	unlikely 	to 	reduce
competition. 	The 	same may 	be 	true 	for 	transactions 	involving
television 	and 	radio 	stations. 	Accordingly, 	we 	suggest 	that 	the
FCC 	continue 	to 	consider waiver 	applications 	even when 	the 	"safe
harbor" 	conditions 	are 	not 	met. 	To 	the 	extent 	that 	the 	FCC 	is
concerned 	about 	competitive 	problems 	from 	such 	combinations, 	we
believe 	that 	the 	analytical 	framework 	outlined 	in 	the 	Merger
Guidelines 	provides 	a 	useful 	method 	for 	identifying 	potentially
troublesome 	transactions.
38

VIII. 	Conclusion
FCC 	rules 	ban holding 	attributable 	ownership 	interests 	in
television 	stations 	whose 	signals 	overlap, 	limit 	the 	number 	and
audience 	reach 	of 	commonly-owned non-overlapping 	television
stations, 	and 	limit 	owning 	geographically proximate 	television
and 	radio 	stations. 	From 	the 	perspective 	of 	competition policy,
the 	blanket ban on 	local 	television multiple 	ownership 	and 	the
restrictive 	policy towards 	radio-television cross-ownership 	could
be 	appropriate 	if 	the 	net 	effect 	of 	such combinations 	was 	likely
to 	be 	anticompetitive, 	and 	if 	the 	costs 	of 	case-by-case
evaluation 	are 	likely 	to 	exceed 	the 	benefits 	from 	allowing 	those
combinations 	having positive 	net 	effects. 	But 	applying 	the 	kind
of 	analysis 	used 	in 	antitrust 	enforcement 	to 	the 	competition
issues 	in 	local 	television 	and 	radio 	markets 	suggests 	there 	may
be 	little 	basis 	for 	such 	a 	strong presumption 	that 	combinations
in 	proximate 	locations 	would 	reduce 	programming variety 	or 	create
market 	power in 	advertising markets. 	Moreover, 	it 	may 	be 	that
combination could 	lead 	to 	substantial 	efficiencies, 	similar 	to
those 	demonstrated 	in other 	broadcast 	situations. 	Thus, 	it may
be 	appropriate, 	and 	consistent 	with 	the 	concerns 	of 	competition
policy, 	to 	relax 	the 	current 	blanket 	prohibitions. 	The 	FCC 	might
consider, 	in 	addition 	to 	the 	ways 	it 	has 	proposed 	to 	relax 	the
existing 	rules, 	using more 	case-by-case 	analysis.
For 	similar reasons, 	competition policy would 	not 	bar
relaxing 	the 	rules 	applying 	to 	television 	station ownership
39

nationwide. 83 Because 	television 	stations 	compete 	in 	local
markets, 	the 	competitive 	effects 	of 	multiple 	station ownership
are 	best 	understood 	by analyzing 	the 	impact 	of 	such ownership 	in
individual 	markets. 	National 	ownership 	limits 	are 	unlikely 	to
increase 	the 	variety of 	programming 	available 	to 	local 	viewers 	or
to 	protect 	competition 	in 	advertising markets. 	But 	the 	limits
may 	prevent 	firms 	from 	realizing certain efficiencies 	that 	derive
from 	owning more 	than 	one 	station.
83 	We 	reiterate 	that 	this 	reply comment 	has 	addressed 	issues
relating 	to 	economic 	efficiency, 	competition, 	and 	the 	enforcement
of 	the 	antitrust 	laws. 	It has 	not addressed other issues, 	such as
the 	relationship 	between 	diversity 	of 	ownership 	and 	diversity 	of
editorial 	viewpoint, 	that 	may be 	important 	to 	the 	FCC.
40

Appendix
To 	examine 	the 	impact 	of 	cable 	television on broadcast
stations' 	audience 	shares, 	we 	constructed 	a 	data 	set 	consisting
of 	the 	following 	variables:
1. 	SHAREi 	= sum of 	audience 	shares 	of 	broadcast 	stations 	in
ADI 	i 	in 	the 	9:00 	a.m 	to 	midnight 	time 	slot. 	The 	audience
share 	of 	each 	station 	is 	the 	percentage 	of 	households 	using
television 	tuned 	to 	that 	station.
2. 	CABLEi 	= percentage 	of 	homes 	passed by cable 	in ADI 	i.
3. 	HUTi 	= percentage 	of 	television 	households 	using
television 	in ADI 	i 	during 	9:00 	a.m 	- 	midnight 	time 	slot.
4. 	NUMSTAi 	= number 	of 	broadcast 	television 	stations 	in ADI
i.
5. 	RPCINC i 	= real 	per 	capita 	income 	in ADI 	i.
6. 	RLPCSLi 	= real 	per 	capita 	retail 	sales 	in ADI 	i.
7. 	POP_HHi 	= average 	number 	of 	persons 	per 	household 	in ADI
i.
Each of 	these 	variables 	was 	created 	for 	1988 	and 	1984 	for
each 	of 	the 	212 	Arbitron ADls. 	We 	then pooled 	the 	data 84 for
both years 	(thus 	creating a 	data 	set 	of 	424 	observations 	on 	7
variables), 	and 	regressed 	SHARE 	on 	the 	remaining variables. 	To
control 	the 	possibility 	of 	heteroskedastic 	disturbances, 	we
employed weighted 	least 	squares 	(using 	the 	number of 	households
84 	We 	conducted 	a 	Chow 	test 	to 	assess 	whether 	the 	parameter
restrictions 	implied 	by 	this 	pooling 	were 	valid. 	The 	results 	of
this 	test 	led 	us 	to 	accept 	the 	null 	hypothesis 	that 	the
restrictions 	held.
41

in each ADI 	as 	the 	weight).85 	We 	obtained 	the 	following
results:
Source I 	SS 	df 	HS 	Number of obs .. 	424
---------+------------------------------ 	F{ 	6, 	417)" 	132.09
Hodel
Residual
44600.8195 	6 	7433.46992
23467.2177 	417 	56.2763014
Prob > F
R-square
.. 	0.0000
= 0.6552
---------+------------------------------ 	Adj R-square 	= 0.6503
Total I 	68068.0372 	423 	160.917346 	Root HSE 	= 7.5018
The 	results 	of 	this 	regression 	suggest 	that, 	holding 	other
factors 	constant, 	each one 	percentage 	point 	increase 	in 	the
number 	of 	homes 	passed by cable 	in 	an ADI 	is 	associated with 	a
half-percentage 	point 	decrease 	in 	the 	aggregate 	audience 	share 	of
the 	broadcast 	stations 	located 	in 	that 	ADI. 86 We 	can 	reject 	the
85 	The 	ordinary 	least 	squares 	parameter 	estimates 	were 	quite
similar 	to 	those 	obtained using weighted 	least 	squares.
86 The 	sample correlation coefficient between CABLE 	and 	SHARE
is 	-0.64.
42

hypothesis 	of 	no 	relationship 	between 	the 	two 	variables 	at 	the
one 	percent 	level 	of 	statistical 	significance. 87
87 	If SHARE 	is 	also a 	determinant of CABLE, 	then this 	ordinary
least 	squares 	coefficient might 	fail 	to 	estimate 	consistently 	the
true 	relationship 	between 	CABLE 	and 	SHARE. 	To 	obtain 	a
statistically 	consistent 	parameter 	estimate 	under 	such 	a
circumstance 	would 	require 	one 	to 	find 	some 	statistically
predetermined 	factor 	that 	influences 	CABLE, 	but 	not 	SHARE. 	It
seems 	likely, 	however, 	that 	any 	factor 	that 	influences 	CABLE 	will
also 	influence 	SHARE.
43