FTC Staff Comment Before the North Carolina State Goals and Policy Board Concerning Certificate of Need Regulation of Health Care Facilities

V89~,)28
COMMISSION AUTHORIZED
Certificate of Need Regulation of Health Care Facilities
Prepared Statement of
Keith B. Anderson
Special Assistant to the Director
Bureau of Economics
Federal Trade Commission
Before
North Carolina State Goals and Policy Board
March 6, )989

It is a 	pleasure to be here today to discuss the available evidence on
the need for and effectiveness of Certificate of Need, or CON, regulation. 1
My name is Keith Anderson. 	I am Special Assistant to the Director of the
Bureau 	of 	Economics 	of 	the 	Federal Trade Commission. 	For the last 	ten
years, 	much 	of 	my 	professional 	work 	has 	focused 	on 	the 	effects 	of
government regulation on the functioning of markets. 2 	I am a coauthor of
one of the three studies the Bureau of Economics has published in the last
three years dealing with Certificate of Need regulation.s
To 	promote 	competition 	in 	health 	care 	markets, 	the 	Federal 	Trade
Commission and its staff have been active both in antitrust law enforcement
and 	in 	advocacy of regulatory 	reforms, including advocating the 	repeal of
Certificate of Need regulation in several other states.· 	In addition, many of
the 	Commission's antitrust 	investigations 	in 	the 	health care 	field 	focus 	on
competitive 	problems 	ti18t 	would 	be 	less 	severe 	if 	there 	were 	no 	CON
regulations. 6 	.
In these comments, I would like to do four things. 	First, I would like
to 	review 	the 	economic 	evidence 	concerning 	the 	effectiveness 	of 	CON
1 These 	comments 	represent 	the 	views 	of 	the 	staff 	of 	the 	Federal
Trade Commission's Bureau of Economics, and 	not necessarily 	those of the
Commission itself or any individual Commissioner.
2 A copy of my vita is attached.
3 Keith 	B. 	Ander~on 	and 	David 	I. 	Kass 	(1986), 	Certificate 	of 	Need
Regulation 	of 	Entry 	into 	Home 	Health 	Care: 	A 	Multi-product 	Cost 	Function
Analysis, Bureau of Economics Staff Report to the Federal Trade Commission.
The 	other studies 	are 	Monica 	Noether 	(1987), Competition 	Among Hospitals,
Federal Trade Commission, 	Bureau of Economics Staff Report; and 	Daniel
Sherman 	(1988), 	The 	Effect 	of 	State 	Certificate-of-Need 	Laws 	on 	Hospital
Costs: 	An 	Economic 	Policy 	Analysis, 	Bureau 	of 	Economics, 	Federal 	Trade
Commission.
• See, e.g., 	Letter 	to Maston 	T. 	Jacks, 	Esq., Chairman, Commission 	on
Medical Facilities and Certificate of Public Need, Commonwealth of Virginia,
from Jeffrey I. Zuckerman, Director Bureau of Competition, Federal Trade
Commission, dated August 6, 1987; Letter to The Honorable John F. Pressman
and 	The 	Honorable 	Donald 	W. Snyder, Pennsylvania 	State 	Representatives,
from 	John 	M 	Mendenhall, 	Acting 	Director, 	Cleveland 	Regional 	Office,
Federal Trade Commission, dated March 	30, 	1988; letter to The Honorable
Culver 	Kidd, Georgia State Senator, from 	Paul 	K. Davis, Director, Atlanta
Regional 	Office, 	Federal 	Trade 	Commission, 	dated 	February 	6, 	1989; and
Letter 	to 	The 	Honora;,;e 	Bernice 	labedz, 	Nebraska 	State 	Senator, 	from
Jeffrey 	I. 	Zuckerman, 	Director 	Bureau 	of 	Competition, 	Federal 	Trade
Commission, dated February 22, 1988.
5 See, e.g., Hospital Corporation of America [Chattanooga acquisitions],
106 F.T.C. 361, 489-496 (1985), affirmed 807 F.2d 	1381 (7th Cir. 1986), cert.
denied, 	107 	S.Ct. 	1975 	(1987); 	Hospital 	Corporation 	of 	America 	[Forum
acquisitions], 	106 	F.T.C. 	298 	(1985) 	(consent 	order); and 	American 	Medical
International, 104 F.T.C. I (1984).

In these comments, I would like to do four things. 	First, I would like
to 	review 	the 	economic 	evidence 	concerning 	the 	effectiveness 	of 	CON
regulation 	in 	controlling 	health 	care 	costs. 	Second, 	I 	will 	consider 	the
effects of CON regulation in limiting competition and innovation in health
care 	markets. 	Third, I will discuss changes in 	health 	care in recent years
that 	have 	increased 	the 	ability 	of 	health 	care 	markets 	to 	function 	in 	an
unregulated, competitive 	way. 	Finally, I will 	briefly consider some of the
rationales for CON regulation.
To jump ahead and provide a preview of my conclusions, almost all of
the available empirical evidence suggests that CON regulation does not help
control 	the 	costs of 	health 	care. 	Further, 	the staff of 	the 	Federal 	Trade
Commission's 	Bureau 	of, Economics 	find 	nothing 	in 	the 	issues 	that 	I 	will
discuss 	that 	leads 	us 	to 	believe 	that 	Certificate 	of 	Need 	regulation 	is
necessary to achieve any of the rationales advanced for it. 	Consumers would
most likely be better served if CON regulation were removed.
A. 	The AuUable Economic Evldeace Stroni1y Sunests That Certificate of
Need 	Reaulation 	Has 	Not 	Been 	Effective 	In 	Redudni 	tbe 	Costs 	of
Health Care.
As 	1 noted, 	in 	the 	past 	three 	years, 	the 	Bureau 	of 	Economics 	has
published 	three studies that, in whole or in part, look 	at the effectiveness
of Certificate of 	Need regulation 	in 	reducing the cost of providing health
care 	services. 	(I 	believe cost 	reduction 	is 	the 	primary 	objective 	of 	those
advocating CON regulation.) 	These studies are only three of at least sixteen
studies 	on 	the 	effects 	of 	CON 	regulation 	that 	have 	appeared 	in 	the
economics 	literature 	in 	the 	last 	fifteen 	years. 6 	Of 	these 	16 	studies, 	II
• In addition 	to the three Bureau of Economics' studies, the following
studies 	have 	been 	done: 	(1) 	David 	S. Salkever 	and 	T.W. 	Bice (1976), "The
Impact 	of 	Certificate 	of 	Need 	Controls 	on 	Hospital 	Investment," 	Milbank
Memorial 	Fund 	Quarterly, 	S4 	(Spring), 	pp. 	) 85-2) 4; 	and 	(1979) 	Hospi/al
Cerrijicate 	of 	Need 	Controls: 	Impact 	on 	Investment, 	Costs. 	and 	Use,
American 	Enterprise 	Institute; 	(2) 	F.J. 	Hellinger 	(1976), 	"The 	Effect 	of
Certificate-of-Need 	Legislation 	on 	Hospital 	Investment," 	Inquiry, 	J3 	(June),
pp. 	187-193; 	(3) 	Frank 	A. 	Sloan 	and 	Bruce 	Steinwald 	(1980a), 	"Effects 	of
Regulation on Hospital Costs and Input Use," Journal of Law and Economics,
23 (April), 	pp. 	81-109;and (1980b) 	Insurance. 	Regulation. and 	Hospital COSlS,
Lexington 	Books; 	(4) 	D.R. 	Cohodes 	(1980), 	Insti/u/ional 	Response 	/0
Regulation: Certificate of Need and Hospitals, Unpublished PhD Dissertation,
Harvard 	University 	School 	of 	Public 	Health, 	January; (S) 	Policy 	Analysis,
Inc .. and Urban Systems Research and Engineering, Inc. (1980), Evaluation of
the Effects of Cerrificate of Need Programs, HRA .231-77-0144, Department
of Health and Human Services; (6) Craig Coelen and Daniel Sullivan (1981),
-An 	Analysis 	of 	the 	Effects 	of 	Prospective 	Reimbursement 	Programs 	on
Hospital 	Expenditures," 	Health 	Care 	Financing 	Review, 	2 (Winter), pp. 	)-40;
(7) 	Paul 	L. Joskow (191:0), "The Effects of Competition and 	Regulation 	on
Hospital Bed Supply and the Reservation Quality of the Hospital," The Bell
( con/inued .. .)
2

examine the effect of CON on the cost of health care, while seven examine
the 	effect 	of 	these 	r:gulations 	on 	hospital 	investment. 7 	Rather 	than
examine the Bureau of Economics' studies alone, I would like to summarize
the conclusions that follow from all of the studies.
In their 1981 review of five studies of CON regulation, Bruce Steinwald
and Frank Sloan, two Vanderbilt University health economists, wrote:
Research on [certificate of need) hospital regulation has produced
remarkably 	consistent 	results. 	The 	empirical 	evidence 	indicates
tha t 	certificate-of -need 	laws 	have 	not 	been 	successful 	in
restraining per diem, per case, or per capita hospital costs.'
There 	has 	been 	little 	evidence 	in 	the 	10 	or 	so 	studies 	that 	have
appeared 	since 	the 	Steinwald 	and 	Sloan 	review 	to 	alter 	the 	conclusions
drawn at that time. 	There is near total agreement among health economists
that Certificate of Need regulation does not achieve the goal of lowering the
costs of 	health 	care: some 	studies 	find 	costs 	are 	the same 	whether or not
CON 	regulations 	are 	present, 	others 	find 	that 	costs 	are 	actually 	higher
where 	there 	is 	CON.~ 	'The 	majority 	of 	these 	studies 	have 	examined 	the
e(... continued)
Journal oj Economics. 	II 	(Autumn), PP. 421-447; (8) Frank A. Sloan (1981),
"Regulation and the Rising Cost of Hospital Care," Review of Economics and
Statistics, 	(November), 	pp. 	479-487; 	(9) 	Paul 	L. 	Joskow 	(1981), 	Controlling
Hospital 	Costs: 	The 	Role 	of Government 	Regulation, 	MIT 	Press; 	(10) 	Steven
Eastaugh (1982), "The Effectiveness of Community-based 	Hospital 	Planning:
Some Recent Evidence," Applied Economics, 14 (October), pp. 475-490; (11) A.
Lee, 	H. 	Birnbaum, 	and 	C. 	Bishop 	(1983), 	"How 	Nursing 	Homes 	Behave: 	A
Multi-Equation 	Model 	of 	Nursing 	Home 	Behavior," 	Social 	Science 	and
Medicine, 	17, 	p. 	1897; (12) 	Dean 	E. 	Farley and 	Joyce 	V. Kelly 	(1985), The
Determino.nJs 	of 	Hospitals' 	Financial 	Positions, 	National 	Center 	for 	Health
Servi.ces 	Research; 	and 	(3). 	John 	W. 	Mayo 	and 	Deborah 	A. 	McFarland
(1989), 	"Regulation, 	Market 	Structure, 	and 	Hospital 	Costs," 	Southern
Economic 	Journal, 	55 	(January), 	pp. 	559-569. 	(In 	some 	cases, essentially 	the
same 	study 	has 	been 	published 	in 	two 	different 	places. 	These 	are 	listed
together and are only counted as one study.)
'1 Two 	of 	the 	studies 	-- 	Sloan 	and 	Steinw<\ld 	(l980a 	and 	1980b) and
Policy 	Analysis, 	Inc., 	and 	Urban 	Systems 	Research 	and 	Engineering, 	Inc.
(1980) 	-- 	examine 	the 	effects 	of 	CON 	regulation 	on 	both 	the 	costs 	of
hospital care and on hospital investment behavior.
• Bruce 	Steinwald 	and 	Frank 	A. 	Sloan, 	"Regulatory 	Approaches 	to
Hospital Cost Containment: A Synthesis of the Empirical Evidence," in A New
Approach 	to 	the 	Economics 	of 	Health 	Care, 	edited 	by 	Mancur 	Olson,
American Enterprise Institute, 198 I, p. 285.
8 Supporters 	of 	CON 	regulation 	sometimes 	point 	to 	the 	number 	or
dollar amount of projects denied, withdrawn or modified as a result of the
(continued ... )
3

effect 	of 	CON 	on 	hospital 	services.l° 	However, similar 	results 	have 	been
o(... continued)
CON 	process 	as 	an 	indication 	of 	the 	value 	of 	CON 	regulation. 	These
amounts, however, are not necessarily an accurate measure of the "excessive"
capital investment deterred because CON regulation may cause the filing of
applications 	for 	more 	projects 	than 	would 	actually 	be 	carried 	out 	in 	an
unregulated market.
CON 	regulation may 	force 	firms 	to compete for 	a limited 	number of
project 	approvals. 	Because 	the 	projects 	are 	approved 	on 	a 	showing 	of
"need", applicants may assume that any CON granted reduces the likelihood
that a approval will be granted to another applicant with a similar project.
This 	generates 	pressure 	to 	file 	preemptively 	or 	defensively 	(particularly
under combined review, where applications for similar projects are required
to file at the same time). 	When several applicants seek approval of similar
projects in 	the same area, that does not mean 	that in the absence of CON
regulation 	all 	proposed 	projects would 	be carried out. 	The applicants may
realize 	that 	demand 	exists 	for 	only one 	project. 	Therefore, 	denial 	of all
applications 	but one would not represent actual savings resulting from 	the
regulations.
Moreover, some applicants may not be committed to carrying out their
proposed 	project 	even 	if 	selected. 	An 	applicant 	may 	be 	protecting 	its
perceived 	long-run 	interests or 	may 	simply 	be 	filing 	to 	delay or 	frustrate
other applications. 	Therefore, the number of applications denied, withdrawn,
or modified may substantially overstate actual deterrence.
The 	anticipation 	t~at projects 	will 	not 	be 	approved 	may 	also 	deter
firms from 	filing 	for projects they would actually construct in the absence
of CON regulation. 	In this case, the number of applications denied would
understate 	the 	effectiveness 	of 	the 	CON 	process. 	Since 	the 	number 	of
applications 	denied, 	withdrawn, 	or 	modified 	may 	either 	overstate 	or
understate the actual number of projects deterred, this is not a reliable way
to mea~ure the effectiveness of the regulations.
10 Of the nine studies examining the effect of CON on hospital costs,
Coelen and Sullivan (1981), Sloan (1981), and Joskow (1981) find 	that CON
has no significant effect on costs. 	Farley and Kelly (1985), Noether (1987),
and 	Sherman 	(1988) all 	found 	that 	hospital 	costs 	were 	higher 	where 	there
was CON regulation. 	Sloan and Steinwald (1980a and 1980b) found that CON
regulation 	has 	no 	significant 	effect 	on 	costs 	where 	CON 	regulation 	is
comprehensive. 	However, 	costs 	were 	higher 	in 	states 	that 	only 	regulated
additions to beds than in states that had no CON regulation.
These 	studies 	use 	cost 	data 	for 	various 	years 	to 	examine 	differences
between states 	with 	CON 	regulation 	and 	states 	without 	it. 	Further, states
have 	enacted 	and 	revi;/.d 	their 	CON 	regulations 	at 	times. 	Two 	of 	the
studies, 	those 	of 	Coele.l; and 	Sullivan 	and 	Sloan 	and 	Steinwald, 	look 	for
differences in the rate of change in costs between states that have CON and
those that do not. 	Sherman (1988) examines differences in costs associated
with different review thresholds -- i.e., with differences in the stringency of
the CON regulation. 	Only one of the studies -- Mayo and McFarland (1989)
- 	finds 	that 	hospital 	costs 	are 	lower 	where 	CON 	regulation 	is 	more
(continued ... )
4

found 	in studies of nursing homes and of firms providing home health care
services.ll
In 	addition 	to 	the 	studies 	that 	empirically 	examine 	the 	effects 	of
Certificates of Need, one recent study has sought to establish the theoretical
relationship between hospital costs and the presence of CON regulation. 	The
author of this study found that, even at a theoretical level, it was not clear
that 	CON 	regulation 	could 	control 	costs. 	Whether a 	perfectly functioning
CON program would reduce hospital costs depends crucially on the way in
which 	hospitals 	are 	reimbursed 	and 	on 	whether 	the 	main 	effect 	of 	CON
regulation is to limit expansion of existing hospitals or to keep new hospitals
from opening. 	According to the study, CON should be expected to lead to
lower 	costs 	per 	admission 	in 	some 	cases; 	in 	others 	it 	will 	lead 	to 	higher
costs. 	Interestingly, 	wh~re hospitals 	are 	reimbursed 	on 	a 	per 	case 	basis--
the method Medicare now uses for hospital operating costs .- CON should be
expected to have no effect on costs. 12
In 	sum, 	the 	economic 	evidence 	concerning 	the 	effectiveness 	of 	CON
regulation 	still 	supports 	the 	1981 	conclusions 	of Steinwald 	and 	Sloan 	that
lO( ••• con tin ued)
stringent. 	While no economic study is methodologically perfect, the problems
with 	this 	study, 	which 	are discussed 	in 	more 	detail 	in 	footnote 	13, seem
particularly great.
Several of the studies have examined the effect of Certificate of Need
regulation on investment in new hospital assets. 	Most of the studies appear
to find 	no significant reduction 	in hospital assets in states that have CON
regulation. (See 	Hellinger (1976), Cohodes (1980), 	Policy 	Analysis, Inc., and
Urban 	Systems 	Engineering, 	Inc. 	(1980), 	and 	Eastaugh 	(1982).) 	However,
some 	of 	the 	studies 	have 	found 	that 	Certificate 	of 	Need 	regulation 	has
limited investment in additional hospital beds. (Salkever and Bice (1976 and
1979) and Joskow (l9S0»· 	Others have found that CON has had no effect or
has 	even 	increased 	the 	number 	of 	beds. (Sloan 	and 	Steinwald 	(1980a 	and
1980b» 	Even 	if 	CON 	has 	reduced 	the 	growth 	in 	new 	hospital 	beds, 	it
appears 	that 	hospitals 	have 	increased 	their 	use 	of 	other 	inputs. 	Salkever
and Bice (1976 and 	1979) found a substitution toward unregulated forms of
capital. 	While 	Sloan 	and 	Steinwald 	(1980a 	and 	1980b) 	did 	not 	find 	a
substitution toward unregulated capital, they found 	that hospitals subject to
CON 	regulation 	tended 	to 	have 	higher employment 	levels than 	comparable
unregulated hospitals.
11 Lee, 	Birnbaum, 	and 	Bishop 	(1983) 	found 	that 	nursing 	home 	costs
were higher in states with CON regulation, while Anderson and Kass (1986)
found 	that 	Certificate 	of 	Need 	regulation 	did 	not 	appear 	to 	increase 	the
attainment of economies of scale in home health care services and also tha t
CON regulation was associated with higher per unit costs.
12 Peter C. Coyte, "Alternative Methods of Reimbursing Hospitals, and
the 	Impact 	of 	Certificate-of-Need 	and 	Rate 	Regulation 	for 	the 	Hospital
Sector," Southern Economic Journal, S3 (April 1987), pp. 853-873.
5

-[Clertificate-of-need controls ... may be regarded as a classic example of
rcgula tory fail ure."lS
Why have Certificate of Need regulations been ineffective? 	To answer
this 	question 	we 	must 	understand 	that 	the 	key 	justification 	for 	CON
lcgislation was the 	belief that 	hospitals did 	not have the proper incentives
to control their costs. 	Patients did not directly pay for most of their health
care - 	particularly hospital care. 	The costs were paid by private insurance
companies or by 	the government. 	In addition, in most cases, payment was
on 	a 	retrospective 	cost 	reimbursement 	basis: 	the 	amount 	the 	hospital
received 	was 	determined 	by 	the 	costs 	incurred 	in 	treating 	the 	particular
patient. 	As a 	result, proponents of planning argued, patients would not be
sufficiently sensitive to high prices; and competition among hospitals would
focus on 	patients' perceptions of quality and convenience. 	Further, it was
argued 	that 	hospitals 	would 	compete 	for 	patients 	by 	competing 	for
physicians, who would be attracted to the hospital that had the fanciest and
most modern faciJities. 14
CON regulation may be ineffective because it does not change any of
these 	incentives. 	Hospitals arc 	no more 	likely 	to compete on 	the 	basis of
price with CON rcgulation than without. 	CON regulation may affect quality
competition 	among 	hosr·itals 	by 	foreclosing 	certain 	avenues 	hospitals could
13 Steinwald and Sloan (1981), p. 296. 	As noted before, only the Mayo
and 	McFarland 	(1989) 	study 	finds 	CON 	regulation 	has 	been 	effective 	in
reducing hospital costs. 	However, there are a number of problems with this
study 	which suggest 	that its 	findings are unreliable. 	First, because of the
way 	the 	authors 	model 	the 	effect 	of CON 	regulation, CON 	is assumed 	to
affect only the number of beds in a hospital. 	Their approach rules ou't the
possibility 	that other 	forms 	of equipment are substituted 	for 	beds or 	that
additional 	staff 	is 	substituted 	for 	capital 	equipment 	where 	there 	is 	CON
regulation. 	As noted 	in footnote 	10, several studies have found 	that these
kinds 	of substitutions occur. 	Second, as 	its 	measure 	of 	the 	stringency 	of
CON regulation the study uses a function of the percentage of applications
approved. 	As discussed in footnote 9, the number of applications approved
or 	denied 	is 	not 	an 	appropriate 	measure 	of 	the 	effectiveness 	of 	CON
regulation. 	Third, 	rather 	than 	compare 	costs 	in 	states 	with 	and 	without
CON 	regulation 	or 	acr-)ss 	states 	with 	different 	cost 	thresholds, 	this 	study
examines costs 	in differc:nt parts of a single state -- Tennessee. 	However,
CON 	decisions 	are 	made 	at 	the 	state 	level. 	Thus, 	there 	are 	no 	real
differences in CON stringency in the data. 	Finally, the authors use data for
the period 	1980 through 	1984. 	During this period, hospital utilization rates
fell 	considerably, 	in 	part 	because 	Medicare 	introduced 	a 	prospective
rcimbursement system (see pp. 12-13) 	The authors do not account for these
changes in an appropriate way in estimating the costs of hospital care.
14 See, e.g., 	Joskow 	(1981), 	pp. 	21-31. 	See 	also 	Health 	Planning and
Resources Development Amendments of 1979, Pub. L. No. 96-99, Section 103
(b), 	93 Stat. 	592 	(1979), 	repealed, 	Publ. 	L. 	No. 	99-660, Section 	701(a), 	100
Stat. 3799 (1986) and the Findings of Fact contained in the North Carolina
Certificate of Need Statute (N.C. Gen. Stat. Section 13IE-175 (I».
6

use to achieve their goals. 	However, it does not eliminate any incentives to
be 	extravagant. 	In 	some 	cases, 	existing 	hospitals 	may 	have 	been 	able,
perhaps with a 	delay, to obtain approval 	for arguably 	uneconomic projects
even with CON regulation. 	Other hospitals were able, the evidence suggests,
to find 	new ways to compete. 	For example, if the CON law only required
approval 	for 	some 	kinds 	of 	capital 	investments 	•• 	e.g., 	new 	beds 	-- 	the
hospitals could add other kinds of fancy equipment. 	If all equipment was
covered, hospitals could compete by providing additional staff.u 	Thus, CON
regulation 	did 	not 	~c wer 	costs 	by 	ending 	supposedly 	unnecessary
expenditures, it merely r:directed such expenditures into other areas.
B. 	CON R~&ulatloD IDterrer~s with Comp~titioD aDd IDDontlon in Health
Car~ Markets.
The 	foregoing 	discussion 	suggests 	that 	CON 	regulation 	has 	been
ineffective. 	But 	more 	than 	that, 	these 	regulations 	have 	probably 	been
harmful. 	The CON regulatory process may increase prices to consumers by
protecting health care providers from competition 	by new entrants. 16 	CON
also reduces the possibility of entry by firms that could provide services of
higher 	Quality 	or 	lower 	cost 	than 	existing 	firms, 	and 	that 	could 	perhaps
replace providers that are not effectively meeting consumer needs. 	This may
explain 	why several of the studies I reviewed 	today 	found 	that costs were
higher in states with CON regulation.
Because Certificate of Need 	regulation delays or 	reduces the 	prospect
of new entry and expansion,17 it increases the likelihood that providers will
16 Salkever 	and 	Bice 	(1976 	and 	1979) 	found 	a 	substitution 	toward
unregulated forms of capital. 	While Sloan and Steinwald (1980a and 	1980b)
did 	not 	find 	a 	substitution 	toward 	unregulated 	capital, 	they 	found 	that
hospitals subject to CON regulation tended to have higher employment levels
than comparable unregulated hospitals.
16 Richard 	A. 	Posner 	(1974), 	"Certificates 	of 	Need 	for 	Health 	Care
Facilities: 	A 	Dissenting 	View," 	in 	Regulating 	Health 	Facilities 	Construction,
edited by Clark Ha vighurst, American En terprise Institute, pp. 113-117.
17 The 	CON 	process 	generally 	places 	the 	burden 	on 	the 	applicant 	to
demonstrate 	that 	a 	need 	is 	not 	being 	served 	by 	those 	currently 	in 	the
market. 	For 	example, 	N.C. 	Gen. 	Stat. 	Section 	131E-183 (a)(3) 	states: 	"The
applicant shall identify the population to be served by the proposed project,
and 	shall 	demonstrate 	the 	~ 	that 	this 	population 	has 	for 	the 	services
proposed 	...." (emphasis added) 	Similarly, N.C. Gen. Stat. Section 	131E-
183 (a)(6) states: "The applicant shall demonstrate that the proposed project
will 	not 	result 	in 	unn~cessary duplication 	of 	existing 	or 	approved 	health
service capabilities or facilities,"
The 	process 	of 	preparing and 	defending 	a 	CON 	application 	is often
costly and time-consuming, particularly if the application is opposed by firms
already in the market. 	For example, an evaluation of the CON program in
(continued ... )
7

exploit 	whatever 	market 	power 	they 	have, 	individually 	or 	collectively, 	to
raise prices above, or reduce quality 	below, the competitive level. 	That is
why 	the Federal Trade Commission 	has cited 	the entry 	barrier created 	by
CON 	regulation 	as 	a 	factor 	significantly contributing 	to 	the 	potential 	for
&nticompetitive effects from hospital mergers. 18
To 	the 	extent 	that 	CON 	regulation 	reduces 	the 	supply 	of 	particular
health 	services 	below 	competitive 	levels, 	prices 	for 	these 	services 	can 	be
expected 	to 	be 	higher 	than 	they 	would 	be 	in 	an 	unregulated 	market.1Q
17( ••• continued)
Michigan found 	that the number and complexity of CON appeals increased
dramatically 	from 	1979 to 	1986, and comparative reviews were particularly
protracted. 	(Michigan Statewide Health Coordinating Council, An E~aluation
of 	the 	Certificate 	of 	Nud 	Program, 	March 	1987, 	pp. 	29-34) 	See 	also,
Hospital 	Corporation 	of 	America 	[Chattanooga 	acquisitions], 	160 	F.T.C. 	at
490-92. 	CON regulation may also create opportunities for existing firms to
abuse 	the 	regulatory 	process 	to 	prevent 	or 	delay 	new 	competition. 	(Terry
Calvani 	and 	Neil 	Averitt 	(J 987), 	"The 	Federal 	Trade 	Commission 	and
Competition in the Delivery of Health Care; Cumberland Law Re~iew, 17, p.
283 (discussing potential (OT health providers to use CON process for "non-
price 	predation"); St. Joseph's Hospital 	v. Hospital Corporation of America,
795 	F.2d 	948, 	959 (lIth 	Cir. 	1986) (defendants' 	misrepresentations 	to state
health 	planning 	body 	concerning 	plaintiff's CON application 	not 	protected
from 	antitrust 	scrutiny); 	Hospital 	Corporation 	of 	America 	[Chattanooga
acquisitions], 106 F.T.C. at 492.)
18 American 	Medical 	International, 	104 	F.T.C. 	at 	200-201 	(J984);
Hospital 	Corporation 	of 	America 	[Chattanooga 	acquisitions], 	106 	F.T.C. 	at
489-496. 	In affirming the Commission's decision in Hospital Corporation of
America, Judge Posner agreed 	that CON 	regulation 	can create a 	barrier to
entry. (Hospital 	Corporation of 	America 	v. 	Federal Trade Commission, 807
F.2d. 1381, 1387 (7th Cir:,lit, 1987»
Some 	shared 	service 	arrangements, 	consolidations, 	and 	other 	joint-
provider 	activities 	could 	significantly 	threaten 	competition, 	unless 	the
prospect 	of 	new 	entry 	would 	keep the 	market 	competitive 	by 	making any
significant, sustained price increases unprofitable. 	CON regulation can thus
conflict 	with 	cost-containment 	by 	limiting 	providers' 	freedom 	to 	pursue
efficiencies 	without 	also 	creating 	unacceptable 	risks 	of 	anti-competitive
effects.
IV Where 	prices are 	regulated, 	the "price increase" may 	take 	the form
of reductions in service quality, so that consumers receive services of lesser
value 	for 	the 	same 	price, 	instead 	of 	paying 	more 	money 	for 	the 	same
services.
Severe shortages of capacity can protect firms that provide substandard
service, 	not 	only 	from 	competitive 	pressures 	to 	upgrade 	performance, 	but
also 	from 	regulatory 	pressures 	to 	adhere 	to 	licensure 	requirements. 	For
example, a state agency may be reluctant to close a nursing home for major
violations 	of 	licensure 	requirements 	if 	the 	patients 	cannot 	be 	placed
(continued ... )
8

Curtailing services or facilities may force some consumers to resort to more
expensive 	or 	less-desirable substitutes, 	thus 	increasing costs for 	third-party
payers or patients. 	For example, if nursing home beds are not available, the
discharge of patients from more expensive hospital beds may be deiayed 20 or
patients may be forced to use nursing homes far from home.
CON 	regulation 	may 	also 	interfere 	with 	competition 	by delaying 	the
introduction 	and 	acceptance 	of 	innovative 	alternatives 	to 	costly 	treatment
methods. 	Regulators may lack the information to determine how many such
facilities are needed, or they 	may not respond 	rapidly enough 	to changing
market 	conditions. 	It 	is 	difficult 	to 	predict 	demand 	for 	innovations 	in
medical practice.21 	Providers have strong financial incentives, which health-
planning agencies 	lack, 	to 	gather 	information 	and 	to adjust 	to 	unexpected
changes 	in 	costs or 	demand. 	Thus, reliance 	on 	market 	forces 	is 	likely 	to
provide more rapid and desirable responses to changing conditions than CON
regula tion.
C. 	Health 	Care 	Markets 	Are 	Now 	More 	Able 	to 	Function 	EHecthely
Without Certificate of Need ReeulatloD.
As I noted, CON regulation was imposed 	in the 	belief that consumers
lacked 	strong 	incentives 	to 	seek 	out 	lower 	cost 	health 	care 	and 	that
providers, 	therefore, did 	not 	have 	sufficient 	incentives 	to 	keep their costs
down. 	But, CON 	regulation 	has not successfully reduced 	health care costs
because 	it 	does 	not 	afr :ct 	those 	incentives. 	However, 	recent 	changes 	in
health 	care 	markets 	th~~t do alter 	incentives shculd 	increase 	the 	likelihood
that unregulated competition will provide health care services efficiently.
Private and public insurance stm pays for the vast majority of health
l~(... continued)
elsewhere. (See J. Feder and W. Scanlon (1980), "Regulating the Bed Supply
in Nursing Homes," Milbank Memorial Fund Quarterly, 58, p. 76.)
20 U.S. 	General 	Accounting 	Office 	(1985), 	Constraining 	Health 	Care
Expenditures. Achieving Quality Care at Affordable Cost, pp. 93-94.
21 	Freestanding 	ambulatory 	surgical 	centers 	(FASCs) 	offer 	an
innovative, 	less 	costly 	alternative 	to 	hospital 	surgical 	facilities. 	However,
evidence suggests that the growth of FASCs generally has been hampered by
the CON process. 	(Dan 'Ermann and Jon Gabel (1985), "The Changing Face
of American Health Ca:-:," Medical Care, p. 407) 	In addition, action on all
CON applications for FASCs in Pennsylvania was delayed by six months while
a 	CON 	task 	force 	reviewed 	the 	need 	for 	these 	facilities. 	(Budget 	and
Finance 	Committee 	of 	the 	Pennsylvania 	Legislature, 	Report on G Study of
Pennsylvania's Certificate of Need Program, February 1987, p. 14.)
9

care. 22 	This 	is particularly true of hospitalization, where only 9.4 percent
of costs were paid directly by consumers in 1986.23 	However, consumers and
employers have become more sensitive to the costs of health 	insurance and
have therefore sought ways to increase incentives for health care providers
to behave efficiently.
One indication of this increased attention to cost control is the growth
in 	health 	maintenance 	organizations 	(HMOs) 	and 	preferred 	provider
organizations (PPOs). 	By 	1988, 	HMO enrollment 	was 	31.4 million.'· 	This
represented more than a five-fold increase from the 6.0 million enrolled in
1976. 26 	Today, an estimated total of 45 to SO million people have the option
of using a 	preferred 	provider. 26 	In December 	1984, only 	1.3 million 	were
enrolled in plans with a PPO option. 27
Another recent change in 	health insurance has been 	the tendency 	for
companies to self-insure. 	By 1987, about 60 percent of employers were self-
insuring 	in 	whole or 	in 	part.28 	According 	to one source, these 	employers
are increasingly interested in competitive alternatives to further reduce their
health costs.2Q
22 Direct 	payments 	by 	consumers accounted 	for 	28.7 	percent 	of 	total
personal 	health 	care 	expenditures 	in 	1986, 	the 	same 	level 	found 	in 	1980.
(Health, United States, /987, table 102, p. 158)
2! Health, 	United 	States, 	/987, 	table 	103, 	p. 	159. 	This 	represents 	a
slight increase from 1977, when only 6 percent of hospital expenditures were
paid 	directly 	by consumers. 	(Paul 	Feldstein (1979), Health Care Economics,
John Wiley and Sons, p. 32.)
2. As 	of 	June 	1~r.8, 	there 	were 	643 	HMOs 	in 	the 	United 	States.
(Telephone interview wi:), David Glazer of InterStudy, February 10, 1989.)
25 Health, United States, /987, table 112, p. 170.
26 There arc currently 637 	PPOs in operation in 	the U.S. 	(Telephone
interview with Ed Pickens, American Medical Care and Review, February 10,
1989.)
27 Jon Ga.bel, Dan Ermann, Thomas Rice, and Gregory de Lissovoy, "The
Emergence and Future of PPOs," Journal of Health Politics, Policy and Law,
11 (Summer), p. 306.
28 Based on data in Steven DiCarlo and Jon Gabel (1988), "Conventional
Health 	Plans: 	A 	Decade 	Later," 	Research 	Bulletin, 	Health 	Insurance
Association of America, p. 	11. 	Before 	1975, less than 	10 percent of health
insurance 	premiums 	went 	into 	self-insured 	plans. 	(Jeff 	Goldsmith (1984),
-Death 	of 	a 	Paradigm: 	The 	Challenge 	of 	Competition, 	Health 	Affairs, 	3
(Fall), p. 9)
10

The percentage of firms requirIng reviews before a patient is admitted
to 	a 	hospital 	and 	requiring 	second 	opinions 	prior 	to 	surgery 	has 	also
increased. 	By 	1987, 	preadmission 	certification 	was 	required 	by 	over one-
third of company-offered health plans,so compared with 2 percent in 1982.:Sl
In 1987, more than 40 percent of employee health plans mandated a second
opinion prior to surgery,S2 while only 28 percent of plans did so in 1984."
Finally, the use of copayments and deductibles in 	private health insurance
plans 	has 	increased.M 	These 	changes increase 	incentives for 	consumers 	to
purchase less expensive health care and forego unnecessary treatments. S6
Government-funded 	programs 	have 	also 	introduced 	changes. 	The
Medicare 	system 	has 	converted 	payment 	of 	hospital 	operating 	costs 	to 	a
prospective payment approach, in which standardized payments are based on
the 	patient's diagnosis, rather than on the costs incurred in each individual
case. S6 	In 	addition, 	many 	states 	have 	been 	experimenting 	with 	cost
so DiCarlo Ga bel (1988), p. II.
SI Goldsmith (1984;" 	p. 14.
S2 DiCarlo and Gabel (1988), p. 11. 	Moreover, an additional 10 percent
required a secoD.d opinion in ,orne cases.
" 	Goldsmith (1984), p. 14.
M 	Between 1982 acd 1984, the percentage of insurance plans paying 100
percent of hospital charges for 	the first days of hospital care fell 	from 67
percent to 42 	percent. 	The percentage paying 	100 percent of surgical fees
fell 	from 	42 	percent 	to 	26 	percent. 	In 	1982, 	only 	4 	percent 	of 	plans
required an annual deductible of $200 or more per family. (Goldsmith (1984),
p. 	12) 	By 	1987, the 	average plan 	had 	an deductible of $330. (Computed
from figures in DiCarlo and Gabel (1988), p. 14.)
35 A 	recent 	Rand 	Corporation 	study 	funded 	by 	the 	Department 	of
Health and Human Services demonstrated that consumers demand less health
care 	when 	they 	have 	to 	pay 	a 	higher 	portion 	of 	the 	cost 	of 	that 	care.
(Willard G. Manning, Joseph P. Newhouse, Naihua Duan, Emmett B. Keeler,
Arleen 	Leibowitz, and 	~_ Susan Marquis (1987), -Health 	Insurance and 	the
Demand 	for 	Medical 	Care: 	Evidence 	from 	a 	Randomized 	Experiment,"
American Economic Review, 77 (June), pp. 251-277)
S6 In 	contrast 	to 	operating 	cost 	reimbursement, 	prospective
reiinbursement for the capital costs of hospjral care has been delayed until
at least 1991. 	(See Omnibus Budget Reconciliation Act of 1987, Pub. L. No.
100-203, Section 4006 (b)(l), 	101 	Stat. 	1330-52 (1987) (amending 42 V.S.C.A.
J39Sww; 	delays 	implementation 	of 	prospective 	reimbursement 	for 	capital-
related costs until 1991.) 	As a result, it is difficult to predict the effect of
Medicare 	prospective 	reimbursement 	on 	hospital 	investment 	activity.
However, 	there 	arc 	several 	reasons 	to 	believe 	tha t 	changes 	in 	Medicare
reimbursement 	already 	made 	and 	those 	proposed 	will 	pressure 	hospitals 	to
(continued ... )
11

containment 	strategies 	for 	their 	Medicaid 	programs. 	Several 	states 	have
begun 	using 	HMOs 	or 	other 	capitation 	and 	case-management 	strategies 	to
provide 	care 	to 	Medicaid 	patients.s7 	In 	addition, 	some 	states 	have 	used
prospective reimbursement plans to pay for nursing home costs. sa
These changes have had substantial effects on 	the demand 	for 	health
care. 	Between 	1975 and 	1986, hospital usage rates declined more than one-
S6(... continued)
reduce 	any 	unnecessary 	investment 	even 	today. 	First, 	the 	costs 	of 	any
major capital investment programs undertaken now will have to be recovered
in hospital charges long past 1991. 	Therefore, current investment plans will
be substantially affecte1 	by 	any anticipated 	changes in 	reimbursement 	for
capital 	costs 	in 	1991. 	Second, 	some 	studies 	have 	found 	that 	increases 	in
hospital capital investment result in increases in hospital operating expenses.
(See Gerard Anderson and Paul B. Ginsberg (1984), "Medicare Payment and
Hospital 	Capital: 	Future 	Policy 	Options," 	Heallh 	Affairs, 	3 	(Fall), 	p. 	37,
citing a 1982 study by Arthur D. Little, Co., "Development of an Evaluation
Methodology for Use in Assessing Data Available to the Certificate of Need
(CON) and Health Planning Problems: Office of the Assistant Secretary for
Health, Department of Health and Human Services, Contract No. 233-79-4003)
Thus, existing 	restraints 	on 	operating 	expense 	reimbursement 	may 	restrain
capital 	spending 	as 	well. 	Finally, 	there 	have 	been 	substantial 	declines 	in
hospital 	admissions and 	average 	length 	of stay for Medicare patients since
the 	prospective 	reimbursl':ment 	system 	was 	put 	in 	place. 	During 	the 	first
three years of the program, total Medicare admissions fell 	by 	11.3 percent,
and admissions per Medicare enrollee fell by 15.9 percent. 	During the same
period, 	the 	average 	length 	of 	stay 	for 	Medicare 	enrollee 	fell 	17 	percent
(Stuart 	Guterman, 	Paul 	W. 	Eggers, 	Gerald 	Riley, 	Timothy 	F. 	Greene, 	and
Sherry 	A. 	Terrell 	(1988). 	"The 	First 	3 	Years 	of 	Medicare 	Prospective
Payment: 	An 	Overview: 	Heallh 	Care 	Financing 	Review, 	9 (Spring), 	pp. 	68-
69). 	Such a significant reduction in hospital utilization is likely to make it
more difficult for hospitals to justify new capital expenditures.
37 Under 	a 	capitation 	plan, 	a 	provider 	receives 	a 	fixed 	payment 	per
month 	which 	depends 	on 	the 	number 	of covered 	patients 	who 	select 	that
firm 	or 	individual 	as 	their 	primary 	care 	provider. 	This 	payment 	is
independent of the costs of any care provided to any particular patient. 	In
some 	cases, 	states 	have 	used 	partial 	capitation 	plans 	where 	the 	provider
receives 	a 	fixed 	fee 	to 	cover 	some 	set 	of 	basic 	services. 	However,
additional 	payments are made 	if 	the 	patient needs care not covered 	under
the capitation agreement. 	(Deborah A. Freund and Edward Neuschler (1986),
·Overview of Medicaid Capitation and Case-Management Initiatives," Health
Car~ Financing Review, (/\ .lnual Supplement), pp. 21-30)
II James 	H. 	Swan, 	Charlene 	Harrington, 	and 	Leslie 	A. 	Grant 	(1988),
-State 	Medicaid 	Reimbursement 	of 	Nursing 	Homes, 	1978-86," 	H~allh Care
Financing Review, 9 (Spring), pp. 33-50.
12

third.3~ 	While some of this decline was probably the result of technological
change 	in 	health 	care, economic 	incentives also 	appear 	to 	have 	played 	an
important 	role. 	Before 	Medicare 	introduced 	its 	prospective 	reimbursement
system, 	hospitalization 	rates 	declined 	more 	slowly 	for 	senior 	citizens 	than
(or 	younger 	patients. 	Since 	1983, 	when 	prospective 	reimbursement 	was
iDtroduced, senior 	citizen 	usage 	has 	declined 	at 	8.3 	percent 	per 	year, the
same rate as for the population at large. 4O
As consumers and health insurers become more cost-conscious, hospitals
with inflated costs and prices will increasingly risk losing business. 	This is
particularly 	true where HMOs and 	PPOs can direct their large numbers of
patients 	to 	hospitals 	that 	charge 	lower 	fees. 	Thus, 	the 	growing 	cost
sensitivity 	of consumers and 	health 	insurers should 	increase 	incentives for
hospitals to provide quality service at the lowest possible price.
D. 	Other 	Arauments 	iD 	Favor of 	Certificate 	of 	Need 	ReaulatioD 	Do Not
Support Such laterrereDce With the Competitive Market.
Several other rationales for Certificate of Need frequently crop up. 	I
would briefly like to comment on three: (I) that CON regulation is necessary
to assure quality health care; (2) that state taxpayers will face higher costs
to provide hospital care for the poor if CON regulation is eliminated; and (3)
that the State's Medicaid costs for nursing home care will increase if CON
regulation is ended.
1. 	CON 	regulation 	is not 	necessary 	to assure Quality in 	the orovision of
health care.
States 	often 	suggest 	that 	assuring 	quality 	is one 	of 	their 	reasons 	for
enacting Certificate of Need legislation. 41 	However, CON does not appear to
be effective in assuring quality. 	A 	Certificate of Need review occurs only
when a hospital or other health care facility 	wishes to make an investment
3~ In 1975, the rate of hospital utilization was 1,254.9 days of care per
1,000 	persons. 	(National 	Center 	for 	Health 	Statistics, 	Health. 	United 	States,
1978, 	DHEW 	Publication 	No. 	(PHS) 	78-1232, 	Public 	Health 	Service, 	U.S.
Government Printing Office, 1978, table 	101, p. 308) 	In 	1986, the rate was
only 833.1 	days of care per 	1,000 persons. (Health. United States, 1987, table
63, p. 109.
40 Between 	1975 and 	1980, hospital 	use by persons aged 65 and above
declined at a rate of 0.33 percent per year. 	For the population as a whole,
the rate of decline was 	1.98 percent. 	Between 	1980 and 	1983, the rate of
decline 	was 	0.77 	percent 	per 	year 	for 	seniors 	and 	2.05 	percent 	(or 	the
population as a 	whole. (Based on 	data 	in 	Heallh, 	United 	Stales, 	1978, table
101, p. 308, and Health, United States, 1987, table 63, p. 109.)
.1 See. 	e.g., 	James 	B. 	Simpson 	(1986), 	-Full 	Circle: 	The 	Return 	of
Certificate of Need Regulation of Health Facilities to State Control: Indiana
Law Review, 19, pp. 1028·1030.
13

that 	requires 	CON 	approval. 	Thus, 	the 	CON 	process 	cannot 	be 	used 	to
discipline 	an 	institution 	that 	is providing an 	unacceptable 	quality 	of care
unless the 	institution 	is regularly applying 	for 	new 	project approvals. 	On
the other hand, periodic accreditation and licensing reviews provide a way to
monitor 	quality 	on 	an 	ongoing 	basis. 	These 	mechanisms 	are 	probably
sufficient to ensure quality. 	CON regulation appears both unnecessary and
largely ineffective as a means of quality control.(2
2. 	Eliminating 	CON 	regulation 	wm 	not 	increase 	the 	costs 	of 	providing
hospital care to the poor.
While 	the 	cost 	of 	indigent 	care 	is 	frequently 	cited 	in 	debates 	on
Certificate 	of 	Need,·! 	the 	relation 	to 	CON 	regulation 	is 	unclear. 	One
possibility 	is that, absent competition, hospitals can 	charge 	paying patients
rates high enough to cover the unreimbursed costs of indigent care.··
If this is occurring, several observations would appear to be in order.
First, such a system may not be able to survive, even with CON regulation,
as 	cost-conscious 	private 	insurers 	and 	governments 	pressure 	providers 	to
operate more efficiently. 	For example, hospitals not subsidizing care for the
poor 	with 	higher 	rates 	to 	paying 	patients 	will 	be 	more 	successful 	in
competing foy patients.
Second, 	hospital 	rates 	that 	cover 	the 	cost of subsidizing care 	for 	the
poor constitute an 	indirect tax on paying users of hospital services and on
all holders of private health insurance. 	While this "tax" is not paid directly
to 	the 	government, 	in 	its 	effect, it 	is 	as 	much 	a 	tax 	as a 	direct 	levy on
.: The CON quality assurance role is sometimes justified by pointing to
epidemiological 	studies 	that 	show 	that 	hospitals 	that 	perform 	certain
operations more frequently have lower mortality rates from those operations.
(See, 	e.g., 	Simpson 	(1986). 	p. 	1030.) 	However, 	accreditation 	and 	licensing
should be able to assure that adequate levels of quality are maintained here
as 	well. 	If 	a 	certain 	volume 	is 	necessary 	to 	assure 	quality 	in 	certain
surgical procedures, a hospital could be granted only a provisional license or
accreditation to perform that operation. 	To be fully-licensed or accredited,
the 	hospital 	would 	have 	to 	demonstrate 	that 	it 	is 	generating 	enough 	of
patients to assure quality. 	If, after some period of time, the hospital is not
meeting this requirement, the license or accreditation to perform that service
would be withdrawn. 	('lie advantage such an approach would have over the
CON 	process 	is 	that 	a 	hospital 	which 	can 	offer 	the 	service 	at 	a 	higher
quality 	level 	or 	a 	lower 	cost 	would 	be 	abk 	to 	enter 	and 	compete 	with
existing providers. 	This is much more difficult, if it is possible at all, with
a CON process.
., See, e.g., 	M 	Lerner, 	et 	al. 	(1987), 	Investigation 	of Certain 	Issues 	in
Connection 	With 	the 	Virginia 	Certificate 	of Need 	Law, 	in 	which 	one of the
questions asked 	of all 	parties expressing opinions on CON 	repeal 	was how
repeal would affect the provision of medical care for the indigent.
.. This argument is suggested in Lerner, et al. (1987), p. VII.22.
14

hospital 	services 	or 	health 	insurance 	premiums.4s 	In 	either 	case, 	the
consumer -- i.e., the taxpayer - 	pays a price that exceeds the value of the
service provided. 	Furthermore, the costs of the current system are probably
greater than would be necessary if an alternative tax was imposed to pay for
indigent care.46
Third, any such indirect tax is highly regressive. 	The costs of health
care 	and 	of 	health 	insurance 	are 	no 	lower 	for 	low 	income 	families 	and
individuals 	than 	for 	the 	well-to-do. 47 	Thus, 	low 	income 	families 	spend 	a
higher percentage of the:r income on health insurance and health care.48
46 Several states, including Florida, Wisconsin, South Carolina, and New
York have developed procedures under which funding for indigent health care
is financed from taxes on hospital revenues. (See M 	Lerner, et aI., (1987), p.
Vll.24, n.·) 	We take no position on whether a tax on hospital revenues or
on 	health 	insurance 	premiums 	is 	superior 	to 	other 	alternatives 	such 	as
funding indigent care from general tax revenues.
4& There 	are 	two 	reasons 	for 	this. 	First, 	as 	noted 	above, 	the 	CON
process probably inhibits competition and 	slows innovation. 	This is apt to
result in higher costs, lower quality, or both. 	This inefficient provision of
services would be eliminated if CON was repealed and an alternative means
found to fund indigent care. 	Second, if CON regulation limits entry of new
health facilities for the purpose of guaranteeing prices high enough to cover
the 	costs 	of 	indigent 	care, 	the 	actual 	payments 	to 	health 	facilities, 	in
aggregate, 	are 	likely 	to 	be 	more 	than 	is 	necessary 	to do so - even 	if the
facilities 	behave competitively in pricing their health services. 	This occurs
when existing facilities differ in their burden of indigent care. 	If prices are
high 	enough 	to 	cover 	the 	costs 	of 	indigent 	care 	at 	a 	facility 	with 	the
highest 	percentage of 	indigent 	patients, the same prices will 	be more than
adequate 	to 	co"Ve~ 	indigent 	care 	costs 	at 	the 	facility 	with 	the 	lowest
percentage 	of 	indigent 	patients. 	Thu~ 	while 	indigent 	care 	costs 	are
·covered" 	at 	all 	facilities, 	the 	aggregate 	of 	payments 	by 	non-indigent
patients wiH 	be in excC'ss of the minimum necessary to cover indigent care
needs. 	In 	this case, 	use 	of 	tax 	money to subsidize indigent care, coupled
with the competitive benefits of eliminating CON limitations on entry, may
result in lower costs of subsidizing indigent care.
47 If anything, families 	with 	lower incomes. tend 	to need 	more health
care. 	In 	1986, families 	with incomes below $10,000 had an average of 6.6
visits 	to 	a 	physician. 	For 	all 	income 	classes 	above 	$10,000, 	the 	average
number of visits was either 5.3 or 5.4 (Health, United States. 1987, table 57,
p. 	J03) 	Hospital 	care 	is even 	more strongly 	related 	to 	income: 	In 	1986,
families with 	incomes less than 	$10,000 had 	J,445.9 days of hospitalization
per 	1,000 population. 	The hospitalization rate fell steadily as income rose.
For families 	with 	incones above $35,000, the hospitalization 	rate was 773.8
days of per 1,000 population. 	(Health. United States. 1987, table 62, p. 108.)
.. Sec, e.g., Alain Enthoven and Richard Kronick (1989), RA ("'onsumer-
Choice 	Health 	Plan 	for 	the 	1990s: Universal 	Health 	Insurance 	ina System
(conI inued ... )
15

3. 	Eliminating CON regulation need not lead to increases in expenditures
(or nursing home beds under the State's Medicaid program.
Turning 	now 	to 	the 	third 	argument, 	states also 	appear 	to 	have 	used
CON to control the number of nursing home beds for which they must pay
under the Medicaid program.4Q 	Thus, it is feared 	that CON repeal will 	be
accompanied 	by 	large, 	uncontrollable 	increases 	in 	the 	Medicaid 	budget.6o
The concern appears to be, not with the cost per bed, but rather with 	the
absolute number of beds.n
4_(... con tin ued)
Designed 	to 	Promote 	Quality 	and 	Economy," 	The 	New 	England 	Journal 	of
Medicine, 320 (January 12), p. 95.
4Q See, e.g., Simpsor. (1986), p. 1032.
50 M 	Lerner, et a1. (1987), VII.24.
51 As 	I 	noted 	above, 	the 	one 	study 	that 	has examined 	the 	effect of
CON regulation on the cost of nursing home care found that per unit costs
were 	higher 	io. 	states 	with 	CON 	regulation. 	(Lee, 	Birnbaum, 	and 	Bishop
(1983»
The experience in two of the states that first ended CON regulation of
new nursing home bed construction would appear to provide some support for
this expectation: in both Arizona and Utah, deregulation was accompanied by
large 	increases in 	the number of nursing home 	beds. 	Arizona repealed its
CON regulation of new nursing home bed construction in July 1982,. 	(Lerner,
et 	a1. 	(1987), 	p. 	VI.12) 	The 	number 	of 	skilled 	nursing 	facility 	and
intermediate care facility beds in Arizona grew from 7,148 in 1982 to 13,734
in 1986 - 	an annual rate of gTowth of 16.3 percent. 	(Charlene Harrington,
James H. Swan, and L~1ie A. Grant (1988), "Nursing Home Bed Capacity in
tJ1.e 	States" 	197!-i986," 	Health 	Care 	Financing 	Rel'iew, 	9 	(Summer), 	p. 	87)
Utah repealed all of its CON requirements in December 1984. (Lerner, et al.
(1987), 	p. 	VI.12) 	In 	1984, 	there 	were 	5,425 	nursing 	home 	beds 	in 	Utah.
(Harrington, 	Swan, 	and 	Grant 	(1988), 	p. 	87) 	By 	1988, 	the 	number 	had
increased 	to 	6,906 	.- 	a 	27.3 	percent 	increase 	over 	four 	years. 	(Telephone
interview 	with 	John 	Williams, 	Office 	of 	Health 	Facilities 	and 	Licensure,
Department of Health, State of Utah, January 27, 1989)
However, 	on 	closer 	inspection, 	it 	is 	not 	clear 	that 	the 	experience 	in
Utah and Arizona would be repeated in other states. 	Prior to deregulation,
bot.h 	states 	had 	a 	ratio 	of 	nursing 	home 	beds 	to 	population 	aged 	65 	and
above that was below the average for the nation as a whole. 	Arizona, with
21 	beds per 1,000 aged citizens, had the lowest bed rate of any state in the
nation. 	By 	1986, 	the 	ratio 	in 	Arizona 	had 'only 	risen 	to 	33.5, 	the 	fifth
lowest ratio among 	the states. 	Utah 	had 42.4 beds per 	1000 aged in 	1984,
compared with a national average of 51.9 beds per 1000 aged. 	(Harrington,
Swan, and Grant (1988), p. 89) 	Thus, it is possible that the construction of
new beds in Arizona and Utah merely represented a catching up with demand
that would not be repeated in other states.
16

States 	may 	well 	be 	interested 	in 	limiting 	their 	Medicaid 	expenditures
and 	may, 	therefore, 	conclude 	that 	they 	will 	limit 	the 	number 	of 	nursing
home 	beds 	for 	which 	state 	Medicaid 	funds 	will 	pay. 	The 	Federal 	Trade
Commission's Bureau of Economics has no expertise here; and therefore we
cannot 	advise 	the 	State 	of 	North 	Carolina 	on 	how 	many 	beds 	should 	be
available 	for 	Medicaid 	patients. 	However, 	if 	the 	state 	chooses 	to 	impose
such a limit, we would point out that Certificate of Need regulation would
Dot appear to be the best way of doing so.
If CON regulation is used, the number of nursing home beds available
to everyone is restricted. 	As a result, citizens paying for their own use of
nursing 	homes, 	or 	those 	having 	private 	nursing 	bome 	insurance, may 	have
more difficulty finding the service and quality levels they desire.n 	Further,
as a 	result of CON entry 	restrictions, such patients will 	likely 	pay 	higher
prices 	than 	they 	would 	if 	the 	market 	were 	allowed 	to 	function
competitively.53
Moreover, if CON regulation 	is used to limit the number of beds for
which Medicaid must pay, it is not clear that the state will pay the lowest
possible cost 	for 	the 	use 	of 	those 	beds. 	Indeed, 	the one study examining
nursing home costs and Certificate of Need found that costs were higher in
regulated states.54
There are alternative mechanisms 	for 	limiting 	the 	number of nursing
home 	beds 	for 	which 	Medicaid 	reimburses. 	One 	approach 	would 	be 	to
control the maximum rate the state pays nursing homes for care of Medicaid
patients. 55 	Another approach would be to directly limit the number of beds
n 	In 	1987, 	51 	pelcent 	of 	nursing 	home 	care 	was 	paid 	for 	by 	those
receiving the care or their families. 	Private health 	insurance paid for 0.8
percent 	of 	care, 	while 	government 	programs, 	including 	Medicare 	and
Medicaid, 	paid 	for 	47.5 	percent 	of care. 	(The 	remainder 	apparently 	came
from 	philanthropy 	or 	care 	(or 	which 	the 	provider 	received 	no
reimbursement.) 	(Health. United States. 1987, table 103, p. 159)
53 Nursing 	home 	costs 	have 	been 	found 	to 	be 	higher 	in 	states 	with
CON 	regulation. 	(Lee, Birnbaum, and Bishop (1983), p. 1903) 	In addition,
the 	rates 	charged 	private-pay 	patients 	may 	well 	be 	higher 	in 	states 	with
CON regulation, either because the regulations reduce competitive supply or
because nursing homes will 	be able to exercise market power because CON
regulation reduces the lik :lihood of entry.
54 Lee, Birnbaum, and Bishop (1983), p. 1903.
55 North 	Carolina 	currently 	uses 	a 	prospective 	reimbursement 	scheme
for Medicaid nursing home beds under which each home is reimbursed at an
individually-determined !ate -- provided that rate is not above an industry-
wide maximum rate. 	(Telephone interview with Jim Barnhill, North Carolina
Division 	of Medical 	Assistance, February 	16, 	1989.) 	One 	way to limit 	the
number of nursing home beds filled by Medicaid patients would be to keep
the industry maximum reimbursement rate low.
17

covered 	by Medicaid. 	A third 	alternative 	would 	be 	to 	rely 	on 	utilization
reviews 	to 	limit 	nursing 	home 	care 	to 	those 	Medicaid 	patients 	who 	truly
need care. 56
In 	short, 	the 	state 	may 	choose 	to 	impose 	some 	form 	of 	regulation
affecting 	Medicaid 	patients. 	However, 	CON 	regulation 	does 	not 	appear
Decessary 	to 	any 	effort 	to 	limit 	Medicaid 	expenditures. 	Further, 	CON
regulation 	is undesirable because it is likely to limit access for self·paying
and privately insured patients, as well as for those on Medicaid.
E. 	CODc:lusloD
In 	conclusion, 	the 	available 	economic 	evidence 	does 	not 	support 	the
view 	that 	Certificate 	of 	Need 	regulation 	reduces 	the 	costs 	of 	providing
health 	care 	services. 	Indeed, 	it 	seems 	more 	likely 	that 	CON 	regulation
increases costs by inhibiting competition and the entry of more efficient or
innovative health care providers. 	In addition, changes in the organization of
health care and in health insurance have improved the ability of health care
markets 	to 	function 	in 	an 	unregulated, 	competitive 	environment. 	Finally,
CON regulation does not appear to be an efficient way to ensure the quality
of 	health 	care 	services, 	to 	assure 	that 	health 	care 	is 	available 	to 	the
indigent, or to control Medicaid expenditures for nursing home beds.
This completes my prepared comments. 	I thank you for your time and
would be happy to attempt to answer any questions you may have.
Ie Under 	federal 	regulations, 	there 	must 	be 	a 	utilization 	review 	of
patients in skilled 	nursing 	facilities at 	the 	first 	scheduled 	meeting after a
patient is admitted.' 	That patient must then be reviewed after 30 days and
after 	an 	additional 	60 days. 	After 	that 	there 	is a 	review every 90 days.
For patients in 	intermediate care facilities, reviews are required every 	180
days. 	While 	a 	utilization 	review 	is 	not 	required 	at 	admission, 	in 	North
Carolina, there is a review of the patient's need for that type of care prior
to admission. 	(Telephone interview with Olivia Hill, North Carolina Division
of Medical Assistance, February IS, 1989)
18

February 1989
VITA
NAME: 	Keith B. Anderson
ADDRESS: 	1648 N. Randolph Street
ArlinlJton, VA 	22207
DATE OF BIRTH: 	December 5, 1945
PLACE OF BIRTH: LaJolla, California
EDUCATION: B.A. (Economics) 1968, Washington State University
MA. (Economics), 1971, Cornell University
A.B.D. (Economics), 1972, Cornell University
Member of American Economic Association
EMPLOYMENT
EXPERIENCE: 	Special Assistant to the Director
Bureau of Economics
Federal Trade Commission
Washington, D.C. 20580
(September 1986 to present)
Resprnsibilities: 	Conduct research into the effects of
government regulations on the functioning of competitive
markets; review studies prepared by others; and perform
other special projects as requested by the Bureau
Director.
Assistant Director for Regulatory Analysis
Bureau of Economics
Federal Trade Commission
(February 1982 to September 1986)
Responsibilities: 	Supervise a staff of between 10 and 15
economists conducting economic analysis on the impact of
government regulation and providing the economic input
for the Commission's Competition and Consumer Advocacy
Program. 	Issues covered include sales of airport landing
rights, restrictions on international trade, and energy
policy.
Assistant Director for Industry Analysis
Bureau of Economics
Federal Trade Commission
(February 1981 to April ]982)
Responsibilities: 	Supervise a staff of ]0 to 15 economists
involved in economic research on microeconomic issues of
interest to the Federal Trade Commission.

VITA - Keith B. Anderson
Assistant to the Director
Burea u of Economics
Federal Trade Commission
(June 1977 to February 1981)
Economist
National Highway Traffic Administration
U.S. Department of Transportation
(May to July 1977 - on detail)
Economist
Division of Industry Analysis
Bureau of Economics
Fede-ral Trade Commission
(September 1972 to June 1977)
Teaching Assistant
Department of Economics
Cornell University
(January 1970 to June 1971)
Research Assistant
Department of Economics
Cornell University
(Summers 1969, 1970)
RESEARCH
PRODUCTS: 	"Do Government-Imposed Ownership Restrictions Inhibit
Efficiency?: The Case of the FCC's Duopoly Rule,"
Working Paper No. J69, Bureau of Economics, Federal
Trade Commission, December 1988 (with John R.
Woodbury)
A Cdtical Evaluation of Petroleum Import Tariffs:
Analytical and Historical Perspectives, An Economic
Policy Analysis, Bureau of Economics Staff Report to the
Federal Trade Commission, April 1987 (with Michael R.
Metzger)
Certificate of Need Regulation of Entry Into Home Health
Care; A Multi-Product Cost Function Analysis, Bureau of
Economics Staff Report to the Federal Trade Commission,
January 1986 (with David I. Kass)
"Utilities and Energy Conservation: 	Competition and
Consumer Protection Issues" in Challenges for Public
Utility Regulation in the 1980's, Harry M Trebing,
editor, Michigan State University, 1981.
2