FTC Staff Comment Before the Department of Labor Concerning Proposed Rulemaking to Replace Ban on Homework in Six Industries with Certificate System

~ 	PBt-'tG:>l 1-
COi~~~ ~fi~ i551 ON'
APPROVED
COMMENTS OF THE BUREA US OF ECONOMICS,
COMPETITION, AND CONSUMER PROTECTION OF
THE FEDERAL TRADE COM1vfISSION
Washington, D.C. 	20580
December 4, 1986
Introduction
The staff of the 	Federal Trade Commission ("FTC) is pleased to offer
comments 	on 	the 	Notice 	of 	Proposed 	Rulemaking 	("NPRM") 	concerning
homeworkers 	published 	by 	the 	Department 	of 	Labor 	("DOL").l 	DOL
requested 	comments 	on 	the 	possibility 	of 	replacing 	the 	outright 	ban 	on
homework 	in 	six 	industries 	with 	a 	certificate 	system 	designed 	to 	insure
compliance with the minimum wage law.
In 	brief, 	we 	conclude 	that 	relaxing 	the 	homework 	ban 	in 	the 	six
currently 	restricted 	industries 	will 	likely 	reduce 	total 	production 	costs 	in
those 	industries. 	As 	a 	consequence, 	prices 	to 	consumers 	will 	likely 	be
reduced. 	Using 	the sparse available data, 	we 	roughly estimate 	the 	gains 	to
consumers and producers from 	removing the ban. 	Based on the productivity
experience of one apparel firm and applying that experience to four apparel-
related 	restricted 	industries, 	the 	estimated 	gains 	are 	on 	the 	order 	of 	$28-
S 150 million 	per year. 	Against such 	potential 	benefits must be weighed 	the
potential 	costs 	of 	relaxing 	the 	ban, 	principally 	the 	possible 	evasion 	of 	the
Employmel" t 	of 	Homeworkers 	in 	Certain 	Industries, 	51 	FR 	30036
(A ugust 	21, 	1986). 	These 	comments 	represen t 	the 	views 	of 	the 	Burea us 	of
Economics, 	Competition 	and 	Consumer 	Protection 	and 	do 	not 	necess:Hily
reflect 	the 	views 	of 	the 	Federal 	Trade 	Commission 	or 	any 	individual
Commissioner. 	The Commission has, hcwever, authorized the 	filing of these
comments. 	Inquiries 	regarding 	these 	comments 	should 	be 	directed 	to 	John
Woodbury, Bureau of Economics.

minimum 	wage 	la w. 2 	Because 	employers 	who 	wish 	to 	use 	homeworkers
legally 	would 	have 	to 	comply 	with 	DOL's 	proposed 	certification 	standards,
DOL will be in a position to monitor these employers to insure that they do
not 	violate 	the 	minimum 	wage 	law. 	Indeed, as employers 	bring 	themselves
within 	the 	monitoring 	system 	and 	the 	new 	option 	of 	legal 	homework
employment, minimum wage violations may actually decline.
In section I of these comments, we consider the benefits from 	relaxing
the 	ban, while 	in 	section 	II 	we 	address 	the 	possible costs 	of 	lifting 	it. 	In
section III we summarize our conclusions.
1. 	BENEFITS FROM RELAXING THE BAN
Originally promulgated by DOL in the 1940's, the current homework ban
applies to six industries employing 550,000 production workers and producing
goods 	valued 	at 	S29 	billion 	in 	1982. 3 	While 	employers 	in 	these 	industries
have 	been 	prohibited 	from 	hiring 	homeworkers, 	the 	total 	number 	oi
homeworkers 	in 	other 	industries 	has 	grown 	dramatically. 	In 	1959, 	DOL
estimated 	that 	a 	total 	of 	22,580 	homeworkers 	were 	employed 	by 	C"S.
Z 	For 	the 	purpose 	of 	this 	commen t, 	we 	do 	not 	address 	whether 	the
minimum wage law advances the interests of consumers or efficient resource
alloca tion.
3 	The 	six. 	industries 	are 	women's 	apparel, 	gloves 	and 	mittens,
embroideries, 	handkerchiefs, 	jewelry, and 	buckle and 	button 	manufacturing.
Approximately 91.2% of employees and 85.6% of the value of shipments are
derived 	from 	the 	four 	apparel-related 	industries. 	U.S. 	Department 	of
Commerce, 1982 Cen~us of Manufacturers, General Summar v , Table 3, pp. 1·6,
1-8, 	1-10, 	1-18. 	The 	ban 	was 	issued 	following 	passage 	of 	the 	Fair 	Labor
Standards 	Act 	("FLSA") 	of 	1938, 	which 	first 	promulgated 	(among 	other
provisions) 	a 	national 	minimum 	wage. 	The 	original 	homework 	ban 	also
encompassed a seventh industry, knitted outerwear. 	In December 1984, DOL
replaced the ban in knitted outerwear with a detailed certification system.
2

manufacturers.· 	By 	1980, that number had grown to 731,660. 5 	By 	1985, the
number of homeworkers had further grown to three million, 57,000 of whom
were 	in 	apparel 	manufacturing. 6 	With 	recent 	advances 	in 	computer
technology permitting an increasing use of remote work stations, the number
of homeworkers 	in all industries is forecast to grow to ten 	million by 	1990 7
and twenty million by the year 2000. 8
This 	increasing 	prevalence 	of 	homework 	suggests 	that 	employers
consider 	homework 	an 	effective 	means of 	reducing 	production costs. 	Thus.
the 	primary 	benefit 	from 	relaxing 	the 	homework 	ban 	in 	the 	six 	restricted
industries 	is 	likely 	to 	be 	the 	ability 	of firms 	in 	these 	industries 	to employ
less 	costly 	techniques 	of 	production. 	For 	example, 	while 	the 	competitive
wage for homeworkers may be above the minimum wage, it may be less than
that for factory workers. 	If so, then to comply with the existing ban, firms
in these industries have likely utilized more capital and plant space and more
expensive 	labor 	inputs to produce any given level of output. 	Alternatively.
because of differences in the working envircmment or because of the form in
which wages are paid (for example, a piece rate rather than an hourly rate).
horneworkers 	rna y 	be 	more 	productive 	than 	factory 	workers. 	For 	reasons
such 	as 	these, 	the 	prices consumers 	must 	pay 	for 	the output of 	these 	firms
:
• 	DOL, 	Emolovmen t of 	Workers 	Under 	FLSA 	(1959), Table 	4, 	p. 	31,
and DOL, Emolovment and Earnings (April 1960), Table SB-2.
5 	Bureau 	of Census, 	Work 	at 	Horne 	Tabulation: 	Reference 	Week 	in
J980 (1985), Table 2, p. 19.
6 	Bureau of Labor Statistics (BLS), "Work Patterns and Prefe~ences of
American 	Workers: 	News 	(A ugust 	J 986), 	p. 	I 	and 	Table 	2, 	and 	from 	BLS
personnel.
7
8
"Telecommuting," Cleveland Plain Dea!er (July 14, 1985), pp. 61, 66.
"Homework," Fortune (August 25, 1984), pp. 10-11.
3

are 	probably 	higher 	than 	they 	would 	be 	absent 	the 	ban. 	By 	permitting 	a
more 	efficient 	use 	of 	resources, 	::! 	relaxation 	of 	the 	homeworker 	ban 	will
tend to reduce both costs and prices in the restricted industries. 9
Not 	only 	will 	consumers 	in 	general 	benefit 	from 	a 	relaxation 	of 	the
ban, 	but 	current 	and 	prospective 	home workers 	will 	benefit 	as 	well.
Relaxation 	of 	the 	ban 	will 	increase 	the 	employment 	options 	available 	to
individuals 	who 	prefer 	to work at home. 	The ban on 	homework 	in 	the SlX
industries 	may 	mean 	that 	some 	of 	these 	individuals 	now 	work 	at 	home 	in
other 	unrestricted 	industries, 	work 	in 	less 	preferred 	(even 	if 	more 	highly
compensated) factory 	positions, or do not work at all. 	Further, because 	the
utilization 	of 	homeworkers 	is 	reduced 	by 	the 	ban, 	the 	compensation 	of
homeworkers is now less than it otherwise would be.
In 	the 	appendix, 	we 	have 	estimated 	both 	the gains to consumers from
the 	lower 	apparel 	product 	prices 	that 	would 	prevail 	in 	the 	absence 	of 	the
ban 	and 	the 	gains 	to 	homeworkers 	(and 	other input suppliers) 	from 	higher
compensation 	levels. 	Our 	calculations 	are 	based 	on 	the 	productivity
experience 	of 	a 	single 	apparel 	firm 	in 	a 	restricted 	industry 	that 	employed
9 	A large number of economic studies have concluded that, in gene;:l1.
restrictions 	on 	production 	techniques 	increase 	the 	cost 	of 	production 	and
increase 	prices 	paid 	by consumed. 	For example, see J. Pertof, "The Imo3ct
of Licensing Laws on Wage Changes in the Construction Industry," Journal of
Law and Economics (October 1980), pp. 409-28, D. Caves, L. Christensen, and
J. 	Swanson, 	"Economic 	Performance 	in 	Regulated 	and 	Unregulated
Environments: 	A 	Comparison 	of 	U.S. 	and 	Canadian 	Railroads," 	OU:lrterlv
Journal 	of 	Economics 	(November 	1981), 	pp. 	559-81, 	and 	A. 	DeVany, 	W.
Gramm, T. Saving, and C. Smithson, "The Impact of 	Input Regulgtion: 	The
Case of 	the 	U.S. 	Dental 	Industry," Journal 	of 	Law 	and 	Economics (October
1982), pp. 367-81.
The 	ban 	may 	also 	have 	skewed 	the 	outcome 	of 	normal 	competitive
processes in other ways. 	The :"-iPR~ notes a number of anomalies created by
the 	current 	ban. 	For 	example, 	homeworkers 	on 	produce 	bathrobes 	and
metal 	buttons 	but 	not 	housecoats or wooden 	buttons, so that production and
consumption are likely to have been diverted from the latter to the ::>rmer.
4

homeworkers 	until 	DOL 	required 	their 	termination. 	These 	calculations
assume that the productivity experience of this single firm would be typical
of 	that 	in 	four 	restricted 	apparel 	related 	industries. 	We 	estimate 	that
consumers 	and 	production 	inputs 	would 	experience 	a 	gain 	of 	between 	$28
million and $150 million per year if the ban were removed. IO
Finally, 	relaxation 	of 	the 	ban 	may 	improve 	the 	international
competitiveness of the restricted industries. 	For example, import penetration
in women's apparel has grown from 	1.8% in 1967 to 17.0% in 1983. 11 	These
increasing imports, particularly in the apparel-related industries, have led to
the 	imposition 	of 	quantitative 	restrictions 	on 	imports 	that 	deny 	consumers
the 	opportunity 	to 	choose 	imported 	instead 	of 	domestic 	apparel. 12 	By
ena bling 	the 	firms 	in 	the 	restricted 	industries 	to 	employ 	the 	most 	cost-
effective production 	techniques thr.ough a relaxation of the ban, the rate of
10 	We provide a range of estimated
and a 	number of simplifying assumptions.
be 	quite 	imprecise. 	A 	complete discussion
the appendix. 	Because of data 	limitations
of the six restricted industries.
gains because of data difficulties
Thus, our estimates are likely to
of 	these 	issues can 	be 	found 	in
the estimated 	gains 	are 	for 	four
11 	These figures cover only SIC 233, which represented approximately
75% 	of 	value 	of 	shipments 	in 	women's 	apparel 	in 	1982 	(1982 	Census 	of
Manufactures). 	Time series data on import penetration are not available for
the 	rest 	of 	the 	restricted 	industries 	or 	for 	the 	remainder 	of 	the 	women's
apparel 	industry. 	The 	data 	for 	SIC 	233 	are 	from 	U.S. 	Department 	of
Commerce, 	J984 	U.S. 	Industrial 	Outlook (1984), 	pp. 41·1 	to 41·11, and 	J978
U.S. Industrial Outlook (1978), pp. 245·50.
12 	The 	U.S. 	imposes 	textile 	import 	quotas 	on 	22 	countries 	under 	the
umbrella 	of 	the 	Multifiber 	Arrangement (MFA). 	M. Morkre, 	Import Ouotas
on 	Textiles: 	The 	Welfare 	Effects 	of 	U.S. 	restrictions 	on 	Hong 	Kong 	(FTC,
1984), 	p. 	1. 	Tariffs 	on 	textiles and 	clothing 	in 	industrial 	countries 	are
between two and three times higher than tariffs on manufactured goods as a
whole. 	The 	textile 	and 	clothing 	industries are 	the 	only ones given 	formal
exemption from 	the normal world trading rules administered by GATT. 	Th~
Economist (July ~8, 1984), p. 57.
5

increase 	in 	imports 	and 	industry 	demand 	for 	trade 	protection 	may 	be
reduced.
In 	sum, 	replacing 	the 	ban 	in 	the 	six 	restricted 	industries 	with 	a
certification 	system 	may 	reduce 	prices 	to 	consumers, 	increase 	the 	eco,,::;rr.ic
well-being 	of 	current 	and 	prospective 	home workers, 	and 	improve 	the
international competitiveness of the restricted industries.
II. 	POTE~TIAL COSTS OF LIFTI01G THE BAN
The primary objection raised by those who oppose relaxation of the b3.n
on 	homework 	is 	the 	possible 	inability 	of 	DOL 	to 	detect 	violations 	of 	the
minimum 	wage 	provisions 	of 	FLSA. 13 	However, 	the 	NPRM 	describes 	an
intensive 	monitoring 	and 	enforcement 	program 	adopted 	by 	DOL 	when 	it
dropped 	the 	ban 	in 	knitted 	outerwear 	and 	replaced 	it 	with 	a 	certification
program. 	By all appearances, this program 	has achieved 	its goals. 	Between
December 	of 	1984 	and 	April 	of 	1986, 	DOL 	reports 	that 	it 	;:ompleted
investigations of 35 of 50 certified employers in 	this industry. 	Of these 35,
14% 	(1.&., 	5 	em ployers) 	were 	found 	to 	ha ve 	u nderpa id 	some 	of 	the i r
home workers, 	and 	by 	an 	average 	of SIOO 	per 	underpaid 	worker 	during 	the
period 	of 	investigation. 	This 	contrasts 	sh3rply 	with 	the 	results 	of 	DOL
complaint-based 	investigations 	of: employers 	in 	all 	other 	industries, 	66% 	of
which 	have 	disclosed 	the 	existence 	of 	minimum 	wage 	violations. 	If 	DOL
applies the same enforcement program to the currently restricted industries,
13 	International 	Ladies' 	Garment 	Workers' 	t...:'nion, 	"Fact 	Sheet 	on
Homework," 	before 	the 	Subcommittee on 	Labor Standards of 	the 	Committee
on 	Education 	and 	Labor 	of 	C.S. 	House 	of 	Representatives (Sept. 	16, 	1986).
pp. 2-13; 	Federationist. "Homework: 	A 	Return 	to 	the 	Industrial Sweatshops
and No Protection" (August 2, 1986), p. 3.
6

the certification 	program should 	have comparable results 	in deterring FLSA
viola tions by certified employers. 14
Because 	certified 	employers 	of 	homeworkers 	will 	have 	to 	identify
themselves 	to 	DOL, 	one 	significant 	administrative 	cost--that 	of 	locating
employers 	of 	homeworkers--will 	be 	reduced. 	Therefore, 	the 	certification
15
program may result in a lower administrative cost to maintain the same level
of FLSA compliance in the six restricted industries as in other industries. IS
Another 	argument 	that 	opponents 	of 	the 	proposed 	relaxation 	have
advanced is that 	while 	homeworkers may be paid at or above the minimum
wage, 	the 	rate 	of 	compensation 	for 	homeworkers 	will 	nonetheless 	be 	lower
14 	It 	is 	our 	understanding 	that 	employers 	of 	homeworkers 	in
industries other 	than 	the original seven 	in 	which 	homework 	was prohibited
must comply with DOL's detailed record-keeping requirements. 	DOL in turn
conducts investigations of a sample of such firms to insure compliance with
the provisions of the FLSA. 	However, none of these unrestricted homework
employers 	must 	first 	be 	certified 	by DOL in order to employ 	homeworkers.
Given 	the dramatic economy-wide growth in homework 	noted above, we are
aware of no policy reason why homework employers in these seven industries
should be certified to hire homeworkers. 	We are also unaware of any policy
reason why DOL would adopt a higher FLSA compliance level for homework
employment in the six restricted industries and knitted outerwear than in all
other 	industries. 	Because of 	recent court decisions, 	we 	recognize 	that 	DOL
may 	be 	required 	to 	institute 	a 	certification 	system 	if 	the 	ban 	is 	relaxed.
NPRM, p. 30036. 	We suggest 	thai DOL carefully 	weigh 	its legal options in
this regard. 	If DOL does adopt a more stringent compliance level, DOL may
wish 	to 	consider 	alternative 	schemes 	for 	enforcing 	FLSA 	that 	could 	result
through deterrence in lower administrative costs.
Another 	reason 	why 	the 	proposed 	relaxation 	may 	result 	in 	fewer
violations 	is 	that 	under 	a 	certification 	program a 	homeworker 	may 	be 	less
reluctant to report FLSA 	violations to DOL. 	Under the proposed relaxation.
the provisions of the FLSA would prevent such an employee from being fired
(provided 	that 	the 	employer 	corrects 	the 	FLSA 	violations 	and 	thus 	retains
his 	or 	her 	homework 	certificate). 	Under 	the 	current 	ban, 	a 	homeworker
complaint to DOL compels DOL to order the employer to terminate homework
which puts its homework employees out of work.
7

than 	that 	of 	their 	factory 	counterparts. 16 	Even 	if 	that 	is 	true, 	it 	is 	not
necessarily 	undesirable. 	Differences 	In 	competitive 	wages 	be:ween
home workers 	and 	factory 	workers 	can 	be 	explained 	by 	a 	whole 	array 	of
factors. 	On the demand side, for example, there may be a tendency for the
homeworker 	wage 	to be higher 	than 	that for 	factory 	workers 	if 	the 	use of
homeworkers results in 	reduced 	need 	for 	plant space and, therefore, savings
to 	the 	employer. 	But 	there 	may 	also 	be 	additional 	costs 	to 	the 	employer
utilizing 	homeworkers 	that 	are 	not 	incurred 	with 	factory 	workers. 	For
ex.ample, 	because 	all 	workers 	in 	a 	factory 	are 	physically 	prox.ima re,
assignments 	can 	be 	made 	and 	changed 	more 	easily· than 	for 	homeworkers.
Monitoring 	of 	work 	quality 	of 	factory 	workers 	can 	likewise 	occur 	as 	the
work 	is 	in 	progress 	rather 	than 	after 	the 	work 	is 	completed. 	These 	other
factors would tend legitimately to reduce the compensation of homeworkers.
On 	the 	supply 	side, 	there 	maybe 	differences 	in 	skill 	levels 	between
homeworkers 	and 	factory 	workers 	that 	affect 	the 	competitive 	wage 	of
homeworkers. 	In 	addition, 	homeworkers 	will 	not 	require 	a 	higher 	wage 	to
compensate them for travel time to and from work. 	Homework also provides
direct 	benefits 	to 	employees, 	including 	satisfying 	the 	desire 	to 	rem2.in 	at
home 	with 	children 	or 	other 	family 	members, 	the 	ability 	to 	set one's 	own
work 	schedule, and 	the ability to' pursue other interests while working 	full-
or part-time at home.
In 	order 	to 	compare 	the 	level 	of 	competitive 	compensation 	of
hcmeworkers to that -of factory 	workers, all of these factors and others must
be considered fully. 	Since competitive labor markets are unlikely to sustain
16 	Thirty-Ninth 	Report 	by 	the Committee on Government Operations.
Homebased 	Clerical 	Workers: 	Are Thev 	Victims of 	E~Dloitation'} 	(July 	16.
1986).
8

unwarranted 	wage 	discriminations, 	there 	is 	probably 	little or 	no 	difference
between the ultimate 	compensation 	paid 	homeworkers and that paid factory
worke:-s 	after 	accounting 	for 	these 	factors. 	Further, 	if 	compensation 	for
homework 	in 	the 	now-restricted 	industries 	is 	less 	than 	that 	for 	homework
elsewhere, 	individuals 	are 	unlikely 	to 	accept 	a 	less 	desirable 	homeworker
position in the currently restricted industries.
Two other 	possible objections to repeal 	of 	the current 	ban 	should also
be 	discussed. 	First, 	some 	have 	argued 	that 	homework 	will 	exacerbate 	the
discomfort 	of 	the 	already 	crowded 	living 	conditions 	with 	which 	some
homeworkers 	must 	contend. l1 	Because 	of 	the 	equipment 	and 	supplies
required by the homeworker, homework could aggravate living conditions that
are 	already 	crowded. 	This 	argument 	is 	not 	persuasive. 	The 	selection 	of
homework over factory 	work 	by an individual in these circumstances would
indicate that the gains from 	homework outweigh the costs, including that of
additional 	crowding. 	Further, 	by expanding earning opportunities 	for such
individuals, 	a 	relaxation 	of 	the 	homework 	ban 	should 	enable 	them 	to
ameliorate their living conditions.
A 	second 	possible 	objection 	to 	repeal 	is 	that 	employment 	of 	factory
workers may 	be 	reduced 	in 	the 	now 	restricted 	industries. 1s 	However, there
is 	some 	evidence 	to 	suggest 	that 	when 	an 	employer 	has 	more 	options 	in
choosing and 	paying 	the 	workers hired, total employment possibly 	including
17 	Federationist (August 2, 1986), p. 2.
18 	International 	Ladies' 	Garment 	Workers' 	t.:nion, 	"Questions 	and
Answers on Industrial Homework" (198:n, p. .,
9

that 	of 	factory 	workers 	actually 	increases. 19 	Further, 	even 	a 	small
reduction in 	the costs of the 	restricted 	industries may dramatically 	improve
their 	international 	competitiveness. 	Consequently, 	consumers 	will 	tend 	to
increase 	their 	purchases 	of 	domestic 	goods 	and 	reduce 	their 	purchases 	of
foreign 	goods, 	leading 	to 	an 	even 	larger 	increase 	in 	domestic 	output 	and
employment.
In sum, there appears to be no reason to believe that homeworkers will
be 	paid 	a 	lower-than-competitive 	wage 	by 	employers or 	that 	the 	relaxation
of 	the 	homeworking 	ban 	would 	aggravate 	the 	crowded 	living 	conditions
experienced 	by some individuals. 	Further, there is some evidence to suggest
that 	total 	employment 	in 	the 	affected 	industries 	could 	increase 	following
relaxation of the ban.
III. 	CONCLUSION
The 	existing 	homeworker 	ban 	In 	six 	industries 	has 	likely 	resulted 	In
inefficient 	production 	techniques, 	higher 	costs, 	and 	higher 	prices 	to
consumers than 	would exist otherwise. 	Using the limited available data, we
provide DOL with an estimate of the possible magnitude of the benefits that
could accrue 	if 	the 	homeworker 	ban 	were 	relaxed. 	Our calculations, while
valid 	only 	within 	the 	confines 	of 	the 	assumptions 	used, 	indicate 	that 	the
gains could be on the order of $28-$150 million annually. 	Included in these
gains, 	but 	worthy 	of 	separate 	mention, 	is 	the 	social 	benefit 	of 	giving
19 	A 	recent 	study 	shows 	that 	total 	employment 	may 	expand 	if 	a
wage-differential 	is 	permitted 	for 	different 	worker 	categories. 	C. 	Brown.
"Estimating 	the 	Effects of 	A 	Youth 	Differential on 	Teenagers and 	Adults,"
Minimum Wage Stud v Commission, Vol. V (June 1981), p. 389. 	As a result of
a cost decline, industry output will expand, generating a higher level of tOD.!
employment. 	If the output expansion is sufficiently large. total employment
can rise beyond the level experienced prior to the cost decrease.
10

workers 	a 	wide 	range 	of 	employment 	options 	from 	which 	to 	choose. 	The
alleged potential costs of removing the ban include increased violation of the
minimum wage 	laws and 	therefore reduced 	income 	for 	workers. 	Such costs
do 	not seem 	likely 	to occur 	to 	any 	substantial 	degree 	in 	competitive 	labor
markets 	given 	the 	certification 	program 	aDd 	the 	current 	level 	of 	DOL
enforcement. 	Assuming 	that 	the 	administrative 	costs 	of 	the 	certification
program 	do 	not 	outweigh 	the 	estimated 	gains 	from 	relaxing 	the 	ban,20 	we
recommend that the homeworker ban be removed.
20 	But see note 14 above.
II

APPENDIX
Estimating 	the 	price 	and 	Output 	Effects
of 	Lifting 	the 	Homework 	Ban
by
Susan 	Foster
Bureau 	of 	Economics
In 	this 	appendix, 	we 	describe 	the 	method 	used 	to 	estimate
the 	price 	and 	output 	effects 	of 	lifting 	the 	homework 	ban 	in 	four
1
apparel 	industries, 	predominantly 	women's 	apparel. 	We 	first
describe 	the 	model, 	and 	the 	supply 	and 	demand 	conditions 	assumed
in 	these 	industries, 	and 	then 	we 	detail 	the 	sources 	for 	the
parameters 	used. 	Next, 	we 	derive 	the 	new 	prices 	and 	quantities
resulting 	from 	use 	of 	more 	efficient 	production 	techniques
following 	removal 	of 	the 	ban. 	Finally, 	we 	describe 	the 	method
used 	to 	calculate 	the 	social 	benefits 	from 	lifting 	the 	ban.
Al. 	~ 	Model
We 	assume 	that 	the 	four 	apparel 	industries 	can 	be 	integratec
into 	a 	single 	industry 	with 	one 	homogeneous 	output 	that 	we 	will
ca 11 	I g 1 ov e s I 	mea sur e din 	do zen s 	0 f 	pa irs. 	~e 	assume 	perfect
competition, 	and 	we 	further 	assume 	that 	linear 	relationships
2
exist 	between 	price 	and 	quantity 	for 	both 	demand 	and 	supply.
1 These 	industries 	are 	women's 	apparel, 	gloves 	and 	mittens,
embroideries, 	and 	handkerchiefs. 	We 	excluded 	jewelry 	and
button/buckle 	manufacturing 	because 	the 	available 	data 	upon 	which
we 	base 	our 	calculations 	appears 	more 	relevant 	for 	apparel-
related 	industries.
2 This 	convenient 	assu~ption is 	consistent 	with 	other 	analvsis
of 	the 	apparel 	industry. 	See, 	for 	example 	M. 	~orkre, 	lmoort
Quotas 	on 	Textiles: 	The 	Welfare 	Effects 	of 	United 	States
Restrictions 	on 	Hone 	Kone 	(FTC, 	August 	1984), 	p. 	14 	(hereafter,
'Morkre').

Demand is the same 	before and after lifting of 	the ban:
(1) 	pd 	= a 	+ 	b 	Qd
where: 	Q 	d 	= the 	quantity 	of 	glove 	units 	demanded 	by 	D.S.
cons urner s
p 	d 	= the 	price 	of 	gloves 	which 	U.s. 	consumers 	are
willing 	to 	pay 	at 	any 	given 	quantity
a = the highest price 	consumers will 	pay 	to move from
zero 	to 	positive 	consumption
b = the 	reciprocal 	of 	the number 	of 	glove 	ur.its a
consumer w ill 	give up in response 	to a 	one
unit 	increase 	in 	price.]
In 	contrast 	to 	the 	stable 	demand, 	the 	supply 	relationship 	is
assumed to 	shift down after lifting of 	the ban so that supply
price 	is 	less 	by 	the 	same 	dollar 	amount 	at 	every 	previously
. 	4
supplied 	quantlty. 	Supply 	before 	the 	ban 	is 	lifted 	is:
(2a) 	P s 	= c 	+ 	d 	Q s
while 	supply 	after 	lifting 	the 	ban 	is:
( 2b ) 	P 	s' 	= 	C I 	+ 	d 	Q s'
where: 	Q s 	= the 	quantity 	of 	gloves 	t~at ~anufacturers
will 	supply 	at 	ar.y 	siven 	price, 	with 	ban
p s 	= 	the 	supply price 	which 	will 	call 	forth
a 	given 	quantity 	of 	gloves, 	with 	ban
:
3 	Because 	of 	the 	pervasiveness 	of 	import quotas, 	we 	assume 	t::at
the quantity 	of 	imports 	in 	the 	four 	restricted 	industries 	remains
constant, 	Le., 	even 	at 	a 	lower 	price, 	the 	quantity 	of 	irr,ports
supplied 	would 	not 	be 	reduced. 	In 	effect, 	we 	assume 	a 	pen':-up
demand 	for 	imports. 	Thus 	our 	representation 	of 	consumer 	demand
should 	be 	considered 	as 	domestic 	demand 	net 	of 	this 	cons':ant
quantity 	of 	imports. 	Because 	only 	incomplete 	data 	on 	imports 	in
the 	restricted 	industries 	is 	available, 	we 	have 	not 	calculated
the 	consumer 	gains 	derived 	from 	purchasing 	the 	fixed 	quantity 	of
imports 	at 	a 	lower 	price 	following 	the 	ban's 	relaxation.
Therefore, 	the 	consumer 	gains 	presented 	in 	the 	text 	will
understate 	the 	true 	extent 	of 	gains.
4 See 	footnote 	12 	for 	details 	of 	this 	shift.
2

c 	= the 	price 	necessary 	to 	begin 	glove 	production,
with 	ban
d 	= the 	reciprocal 	of 	the number of additional
glove 	units 	supplied 	in 	response 	to 	a 	one 	unit
increase 	in 	supply 	price
p 	s' 	= the 	supply 	price, 	without 	ban
Q Sf 	= the 	qua nti ty 	of 	gl ove 	uni ts 	suppl ied 	f or 	any
given 	supply 	price, 	without 	ban
c' 	= the 	equivalent 	of 	c 	but 	without 	ban.
Finally, 	market 	equilibrium 	before 	and 	after 	the 	ban 	is
relaxed 	requires 	that 	at 	the 	market 	price, 	quantity 	demanded
equals 	quantity 	supplied.
A2. 	Parameters 	~ 	in 	~ 	Model
The 	starting 	values 	for 	our 	model 	of 	the 	restricted 	portion
of 	the 	apparel 	industry 	were 	those 	for 	the 	calendar 	year 	1982.
To 	arrive 	at 	quan~ity we 	multiply 	an 	apparel 	output/labor, ratio
without 	homework5 by 	the 	total 	number 	of 	production workers. 6
in 	the 	four 	restricted 	apparel 	industries. 	To 	arrive 	at 	price,
5 The 	output/labor 	ratio 	at 	one 	glove 	factory 	which 	employed 	85
homeworkers 	and 	165 	factory 	workers 	was 	60.0 	dozen 	pairs 	a 	week;
this 	ratio 	fell 	to 	48.5 	dozen "pairs 	per 	week 	when 	the 	homeworkers
were 	laid 	off. 	These 	data 	were 	provided 	by 	plant 	personnel 	at
Tom 	Thumb Glove, 	Wilkesboro, 	N.C. 	Tom 	Thumb 	is 	a 	firm 	which
illegally 	employed homeworkers and 	was 	compelled 	by 	DOL 	to 	termi-
nate 	its 	employment 	of 	homeworkers. 	Tom 	Thumb's 	homeworkers
averaged 	$5 	to 	$6 	per 	hour. 	This 	information 	was 	provided 	by 	The
Center 	on 	National 	Labor 	policy.
6 	U.S. 	Department 	of 	Commerce, 	~ 	Census 	of 	Manufactures.
3

f 	h . 	7
we 	divide 	the 	value 	0 	s 	lpments
by 	the 	qua nti ty 	just 	cal c ul a te d.
price 	figures 	are:
for 	these 	restricted 	industries
These 	initial 	quantity 	and
8
Q = 1,262.8 	million 	dozen 	pairs 	of 	gloves
P = $19.79 	per 	dozen 	pairs 	of 	gloves.
In 	addition 	to 	initial 	price 	and 	quantity, 	we 	require
parameters 	for 	the 	demand 	and 	supply 	models. 	To 	calculate 	the
slope 	of 	the 	demand 	equation 	(Ib' 	in 	equation 	(1) 	), 	we 	utilize
an 	estimate 	of 	-.282 	for 	the 	price 	elasticity 	of 	demand 	for 	the
entire 	apparel 	industry 	at 	the 	original 	price 	and 	quantity 	of 	the
four 	restricted 	industries. 9 with 	this 	estimate 	and 	our 	initial
price 	and 	quantity 	estimates 	we 	calculate 	a 	in 	equation 	(1) 	as
7 Value 	of 	shipments 	in 	1982 	was 	estimated 	to 	be 	$24,992.0
million. 	U.S. 	Department 	of 	Commerce, 	1982 	Census 	of
Manufactures. 	This 	estimate 	was 	made 	in 	the 	following 	manner.
The 	Census 	of 	Manufactures 	provided 	complete 	data 	for 	one 	of 	the
four 	industries, 	women's 	apparel 	(SIC 	233,234, 	and 	236). 	value
of 	shipments 	in 	womenls 	apparel 	totaled 	$24,278.6 	million. 	SIC
2381, 	3151, 	and 	3199 	encompass 	gloves 	but 	SIC 	3199 	also 	encompasses
other 	leather 	goods. 	Both 	SIC 	2395 	and 	2397 	include 	embroideries
as 	well 	as 	pleating, 	buttonhole 	making, 	crochet 	ware 	and 	tucki~g.
Handkerchiefs 	are 	part 	of 	SIC 	2389 	as 	are 	armbands, 	footlets,
garters 	and 	suspenders. 	For 	these 	four 	mixed 	SIC 	groups, 	we
attributed 	only 18 	percent 	of "the 	combined 	value 	of 	shipments 	to
the 	four 	restricted 	industries. 	Because 	employment 	data 	for
gloves, 	embroideries 	and 	handkerchiefs 	were 	also 	included 	in 	the
same 	mixed 	SIC 	groups, 	we 	only 	attributed 	18 	percent 	of 	the
employment 	in 	those 	SIC 	groups 	to 	the 	four 	restricted 	industries.
8 Q = 500,700 	workers 	x 52 weeks 	x the output/labor 	ratio
(48.5).
9 H. 	Houthakker, 	"New 	Evidence 	on 	Demand 	Slasticities,"
Econ 0 met rica 	(1 96 5) , 	p. 	2 80 .
4

589.97 	and 	b 	as 	-5.56 	X 	10 	_8. 10 Because 	there 	are 	no 	available
estimates 	of 	the 	supply 	elasticity, 	we 	utilize 	two 	different
a ssum pti ons. 	In 	one 	case 	we 	assume 	that 	the 	supply 	elasticity 	is
infinite, 	Le., 	that 	the 	supply 	curve 	is 	horizontal 	at 	the 	market
price. 	In 	the 	second 	scenario, 	we 	assume 	that 	the 	supply
elasticity 	is 	20, 	imparting 	a 	modest 	upward 	slope 	to 	the 	supply
curve. 	With 	the 	horizontal 	supply 	curve, 	c 	in 	equation 	(2a)
equals 	S19.79 	and 	d 	equals 	zero. 	When 	the 	supply 	curve 	is 	upward
sloping, 	c 	equals 	S18.80 	and 	d 	equals 	7.84 	X 10 	-10. 11
The 	last 	parameter 	necessary 	to 	proceed 	with 	the 	calcula-
tions 	is 	c t , 	the 	intercept 	of 	the 	post-ban 	supply 	curve 	in 	equa-
tion 	(2b). 	The 	lower 	intercept 	represents 	a 	downward 	shift 	in
t~e 	supply 	curve. 	The 	downward 	shift 	results 	from 	the 	reduction
in 	production 	costs 	following 	relaxation 	of 	the 	homework 	ban. 	We
calculate 	this 	downward 	shift 	by 	applying 	the 	percentage 	increase
in 	productivity 	(Le., 	the 	percentage 	change 	in 	the 	output/labor
ratio) 	resulting 	from 	t~e 	use 	of 	homeworkers 	to 	the 	::>ercent 	of
total 	variable 	costs 	accounted 	for 	by 	production 	workers 	in 	t~e
:
10 	With 	equation 	(1), 	the 	demand 	curve, 	and 	the 	initial
values 	of 	price 	and 	quantity, 	we 	can 	solve 	for 	a, 	the 	initial
intercept 	value 	for 	the 	demand 	equation:
a 	= P D - 	b Q D
= 19.79 	+ 	(.0000000556) (1,262,765,400) 	= 	S89.97
11 The 	intercept 	for 	the 	upward 	sloping 	supply 	curve 	is 	found 	in
the 	following 	manner:
c 	= 	P - 	d 	Q
= S19.79 	- 	(.000000000784) (1,262,765,400) 	= S18.80
5

· 	d" 	. 	12four 	restrlcte 	lnaustr:es. 	The 	new 	intercept 	(c ' ) 	is 	S19.55
for 	the 	horizontal 	supply 	curve, 	while 	that 	for 	the 	upward
sloping 	supply 	curve 	is 	S18.58. 	Essentially, 	removal 	of 	the
homeworker 	ban 	reduces 	costs 	by 	1.2 	pe=-cent.
12 Based 	on 	the 	experience 	of 	Tom 	Thumb, 	the 	effect 	of 	the 	ban's
relaxation 	is 	to 	increase 	the 	output/labor 	ratio 	from 	48.5 	to
60.0, 	a 	23.7 	percent 	increase 	in 	productivity. 	This 	higher 	level
of 	productivity 	is 	associated 	(in 	Tom 	Thumb) 	with 	a 	homeworker 	to
total 	worker 	ratio 	of 	.34 	(85 	homeworkers 	out 	of 	250 	total
workers). 	To 	calculate 	the 	likely 	percentage 	of 	homeworkers 	in
the 	four 	restricted 	industries 	following 	the 	ban's 	relaxation, 	we
first 	assume 	that 	all 	57,000 	homeworkers 	in 	the 	apparel 	industry
(DOL, 	Bureau 	of 	Labor 	Statistics 	(BLS),~, August 	7,1986, 	and
from 	BLS 	personnel) 	are 	employed 	by 	the 	unrestricted 	industries.
Thus, 	homeworkers 	compose 	12.4 	percent 	of 	production 	employment
in 	states 	that 	do 	not 	ban 	homework 	in 	these 	industries 	(produc-
tion 	employ~ent in 	restricted 	and 	unrestricted 	apparel 	industries
from 	U.S. 	Department 	of 	Commerce, 	lia2 	Census 	of 	~anufactures).
We 	then 	assume 	that 	the 	percentage 	of 	homeworkers 	in 	the 	restric-
ted 	industries 	will 	also 	be 	12.4 	percent 	following 	the 	ban's
relaxation. 	Assuming 	no 	change 	in 	total 	employment 	in 	the 	re-
stricted 	industries, 	there 	will 	be 	49,900 	honeworkers 	(based 	upon
402,400 	production 	workers 	in 	restricted 	industries 	who 	are 	not
covered by 	state 	bans 	on 	homeworkers). 	Rather 	than 	estimate 	a
continuum 	of 	production 	functions 	from 	zero 	homeworkers/all 	fac-
tory 	workers 	to 	all 	homeworkers/zero 	factory 	workers 	-- 	both
the 	ends 	of 	t~;e spectrum 	are 	not 	uncommon 	-- 	we 	have 	assumec 	a
dichotomous 	production 	pattern. 	That 	is, 	we 	have 	assumed 	that
all 	horneworkers 	combine 	with 	all 	workers 	at 	the 	To~ Thu:,:",b
ratio, 	.34 	(Le., 	49,900 	homeworkers 	out 	of 	a 	total 	of
146,800 	homeworker/factory 	workers) 	and 	the 	re!7\aining 	production
workers 	produce 	in 	factories 	"Yithout 	any 	homeworkers. 	Thus, 	cut
of 	500,700 	production 	workers, 	146,800 	(29.3 	percent) 	produce 	at
the 	output/labor 	ratio 	of 	60.0, 	and 	the 	other 	353,900 	(70.7
percent) 	produce 	at 	the 	output/labor 	ratio 	of 	48.5.
An 	increase 	in 	worker 	productivity 	indicates 	that 	the 	costs
of 	producing any given level 	of output falls: 	the sa:7:e amount of
labor 	can 	produce 	a 	greater 	output 	or 	equivalently 	any 	given
level 	0: 	output 	can 	be 	proc.uced 	with 	fewer 	labor 	inputs. 	Thus, 	a
23.7 	percent 	increase 	in 	productivit::' 	results 	in 	Co 	19.2 	percent
decline 	in 	labor 	costs. 	To 	see 	why 	this 	is 	so, 	consider 	tr.e
following 	example. 	A.ssume 	that 	the 	current 	daily 	output 	level
(Footnote 	continues)
6

Thus, 	we 	now 	have 	all 	the 	information 	needed 	to 	calculate
the 	new 	prices 	and 	quantities 	following 	the 	ban's 	relaxation. 	By
way 	of 	summary, 	the 	parameters 	for 	the 	demand 	curve 	are:
a 	= 	$89.97
b 	'- 	-5.56 	X 10 	-8
For 	the 	horizontal 	supply 	cu r...re :
C 	= 	S19.79
c' 	= 	S19.55
d 	= o .0
For 	the 	upw ar d- sl opi ng 	supply 	curve:
c 	= 	S18 .80
c' 	= 	S18.58
d 	= 7.84 	X 10 -10
( Footnote 	continued)
(without 	homeworkers) 	is 	100 	units, 	the 	daily 	wage 	of 	workers 	is
S50 	and 	that 	there 	are 	a 	total 	of 	100 	workers. 	Then 	the 	current
labor 	cost 	per 	unit 	of 	output 	is 	S50. 	After 	the 	homework 	ban 	is
r e1 a xed, 	d ailyout put 	ri s est 0 	1 2 3 . 7 	un its 	and 	the 	a'.' era gel abo r
cost 	per 	unit 	of 	output 	falls 	to 	540.42, 	a 	decline 	of 	19.2
per ce nt.
In 	the 	four 	restricted 	industries, 	the 	percentage 	of 	total
variable 	(labor 	plus 	materials) 	cost 	accounted 	for 	by 	productic:i
workers 	is 	22.2 	percent 	(from: l..i.a2. 	Census 	of 	Manufactures).
Therefore 	the 	reduction 	in 	total 	variable 	costs 	by 	firms
employing 	homeworkers 	in 	the 	restricted 	industries 	is 	19.2
percent 	times 	22.2 	percent, 	or 	a 	4.26 	percent 	cost 	reduction.
Because 	the 	cost 	reduction 	is 	only 	attributable 	to 	29.3 	percent
of 	employment 	in 	the 	four 	restricted 	industries, 	the 	average 	cost
reduction 	experienced 	by 	the 	four 	industries 	in 	total 	will 	be
29.3 	percent 	times 	4.26 	percent, 	or 	1.2 	percent..
For 	the 	horizontal 	supply 	curve 	scenario, 	the 	vertical
downward 	shift 	in 	the 	supply 	curve 	is 	.012 	tir:les 	519.79, 	the
original 	supply 	price. 	Thus, 	in 	this 	case, 	the 	price 	follow':'ng
the 	ban 	will 	be 	S19.55. 	In 	the 	case 	of 	the 	upward 	sloping 	supply
curve, 	we 	calcclated 	the 	absolute 	cost 	decline 	15.22) 	at 	the
original 	intercept. 	Had 	we 	used 	the 	original 	price, 	the 	absoll.::e
cost 	reduction 	""ould 	have 	been 	somewhat 	larger 	and 	the 	gains 	frc:71
the 	ban's 	relaxation 	would 	ha',/e 	been 	somewhat 	greater.
7

AJ. 	Quantities 	and 	Prices 	Resulting 	from 	Lifting 	the 	Ban
The 	original 	and 	post-ban 	market 	positions 	are 	depicted 	in
figure 	1 	(horizontal 	supply 	curve 	case) 	and 	in 	figure 	2
(upward 	sloping 	supply 	curve 	case). 	In 	both 	figures, 	the 	original
quantity 	supplied and 	demanded 	is Qs 	while 	the 	market 	clearing-
price 	is 	Pe. 	Following 	the 	ban's 	relaxation, 	the 	supply 	curves
in both figures 	shift down as a result 	of 	the 	induced 	reduction
in cost. 	The new 	equilibrium quantity 	is QC and the new 	market
clearing-price 	is 	PC' 	Equating 	price 	in 	equation 	(1) 	(demand)
with 	price 	in 	equation 	(2b) 	(new 	supply), 	we 	can 	solve 	for 	the
new 	quantity 	resulting 	from 	lifting 	the 	ban, 	Q C:
a 	+ 	b 	Q C 	= 	c ' 	+ 	d QCi
QC 	= (a 	- 	c')/ 	(d 	- 	b).
In 	the 	case 	of 	horizontal 	supply, 	the 	ne w 	ir.te r ce pt 	(c' ) 	is 	the
new 	price 	(PC) 	and 	d 	(slope) 	= O. 	New 	output 	(QC) 	is:
Q C = (a 	- 	P 	C)/(-b) 	= 70.42/ .0000000556
= 1,267.0 	million.
In 	the 	of 	upw-ard- sloping: supply, 	output 	is:	case 	new
Q C = (a 	- 	c ') / (d 	- 	b) 	= 71.39/ .0000000564
= 1,266.8 	million.
In 	the 	case 	of 	horizontal 	supply, 	the 	new 	Darket 	price 	(PC) 	is 	t~e
same 	as 	the 	new 	intercept 	(c ' ), 	or 	S19.55. 	In 	the 	case 	of 	the
up war d 	s lop i ng 	~ up ply 	cur v e , 	mar ke t 	p ric ere mair: s 	t 0 	be
determined 	using 	tr.e 	new 	output 	(QC):
8

Pc = 	a 	+ 	b 	Q C = 	589.97 	- 	(.0000000556) (1,266.8 	mil)
= $19.57
A4. 	Estimated 	Gains 	from 	Relaxing 	the 	Ban
Here 	we 	detail 	a 	method 	of 	measuring 	the 	gains 	to 	the 	economy
from 	relaxing 	the 	ban 	on 	homework. 	Lifting 	the 	ban 	would 	induce
an increase 	in the output and lower 	the price 	of 	goods in 	the
restricted 	apparel 	industries. 	Gains 	from 	lifting 	the 	ban 	would
consist 	of 	the 	changes 	in 	consumer 	and 	producer 	surplus 	that
result 	from 	the 	improved 	use 	of 	resources. 	To 	compute 	these
changes 	we 	first 	look 	at 	what 	consumers 	would 	gain. 	At 	the
initial 	point B in figures 	1 and 2 the consumer 	surplus is 	that
area 	above 	price, 	but 	below 	the 	demand 	curve. 	This 	is
rep res e n ted 	by 	the 	are a 	0 f 	the 	t ria n g1 e 	a- PB- B. 	At t e r 	the 	ba n 	i s
lifted 	consumers 	and 	producers 	will 	b€ 	at 	point 	c. 	The 	consumer
surplus 	is 	represented 	by 	the 	area 	of 	triangle 	a-pc-C.
Changes 	in 	consumer 	surplus 	as 	a 	result 	of 	lifting 	the 	ban
are 	the 	differences 	between 	the 	second 	and 	first 	triangles.
For 	the 	horizontal 	supply 	curve 	the 	gain 	is 	5300,411,438.
For 	the 	upward 	sloping 	supply, curve 	the 	gain 	is 	5281,392,757.
We 	now 	compute 	the 	changes 	in 	producer 	surplus 	resulting
from 	lifting 	the 	ban, 	represented 	in 	figures 	1 	and 	2 	as 	a 	move 	fror.:
point 	B 	to 	point 	C. 	In 	the 	case 	of 	a 	horizontal 	supply 	curve
this 	area 	is 	zero. 	For 	the 	upward 	sloping 	supply 	curve 	the
methodology 	is 	identical 	to 	that 	in 	the 	consumer 	surplus 	case.
Producer 	surplus 	before 	the 	ban 	is 	lifted 	is 	represented 	!:y
the 	triangle 	PS-C-B. 	This 	area 	can 	be 	calculated 	as:
9

Area (PB-c-B) 	= 1/2 	(PB 	- 	c)QB
= $624,800,000.
Producer 	surplus 	after 	the 	ban 	is 	lifted, 	where 	point 	C
represents 	the 	new 	price 	and 	quantity, 	is 	represented 	by:
Area(Pc-c'-C) 	= 1/2 	(PC 	-c')QC
= $628,767,638
The 	gain 	in 	producer 	surrlus 	is 	thus 	$3,967,638.
Finally 	the 	total 	gain 	to 	society 	as 	a 	result 	of 	lifting 	the
ban 	is 	simply 	the 	sum 	of 	the 	gains 	in 	consumer 	and 	producer
surplus.
For 	the 	case 	of 	the 	horizontal 	supply 	curve, 	the 	change 	in
producers 	surplus 	is 	zero. 	Therefore 	the 	total 	gain 	is
$300,411,438.
For 	the 	upward 	sloping 	supply 	curve 	case, 	the 	total 	gain 	is
$285,360,395.
We 	note 	that 	the 	assumptions 	used 	to 	generate 	the 	percentage
cost 	reduction 	may 	result 	in 	the 	producer 	and 	consumer 	gains
being 	overesti~ated. 	Approximately 	97 	percent 	of 	the 	value 	of
shipments 	in 	these 	four 	industries 	(an 	estimated 	$25 	billion) 	:.s	,
accounted 	for 	by 	women's 	apparel. 	It 	may 	be 	that 	the
productivity 	gains 	using 	homeworkers 	in 	w6men's 	apparel 	would 	not
be 	as 	substantial 	as 	those 	experienced 	by 	Tom 	Thumb 	in 	glove
production. 	Further, 	our 	baseline 	calculations 	assume 	that
absent 	the 	ban, 	the 	percentage 	of 	homeworkers 	in 	the 	four
industries 	would 	be 	identical 	to 	that 	in 	the 	unrestricted 	apparel
i:;dustries 	(Le., 	12 	percent 	of 	all 	production 	employees 	in
states 	that 	do 	not 	ban 	homework). 	But 	the 	actual 	percentage
1'1

could 	be 	lower 	if 	cost-efficient 	production 	techniques 	in women's
apparel 	in 	particular 	do 	not 	include 	an 	extensive 	use 	of
homeworkers. 	For 	example, 	in 	knitted 	outerwear, 	certified
homework 	employees 	accounted 	for 	only 	2.7 	percent 	of 	all 	knitted
outerwear 	employees 	(data 	provided 	by 	OOL 	officials). 	Finally,
we 	do 	not 	know 	the 	extent 	to 	which 	illegal 	homework 	occurs 	in 	the
four 	restriced 	industries 	even 	while 	the 	ban 	is 	in 	effect.
For 	these 	reasons, 	the 	estimate 	of 	the 	gains 	are 	deflated 	by
90 	percent 	and 	by 	50 	percent 	to 	reflect 	our 	uncertainty. 	For
example, 	the 	10 	percent 	estimates 	would 	be 	roughly 	equivalent 	to
assuming 	that 	only 	1.2 	percent 	of 	the 	employees 	in 	the 	four
restricted 	industries 	would 	be 	horneworkers 	absent 	the 	ban. 	We
present 	these 	deflated 	estimates 	in 	Table 	1, 	the 	basis 	for 	the
gains 	reported 	in 	the 	text.
l:!.

Table 1
Effects of 	t.~e Ban's Relaxation*
Horizontal SUpply CUrve
Change in' Consumer SUrplus
~pward Sloping SUpplv curle
10 perCEnt of 	basel ire
50 perCEnt of 	baseli;,e
Change in Producer SUrplus
10 perCEnt of 	basel ire
50 perCEnt of 	taseline
Total G1.ange in surn1us
10 p:rCEnt of basel ine
50 perCEnt of baseline
S 30.0
$150.2
$0
SO
S30.0
S150.2
S28.1
S140.7
S.4
S2 .0
S28.5
$142.7
* 	The 	baseline 	calculations 	assume 	(1) 	that 	Tom 	Thumb's
productivity 	experience 	would 	be 	typical 	for 	the 	four 	restricted
industries 	in 	the 	absence 	of 	the 	homeworker 	ban 	and 	(2) 	that
homeworkers 	in 	the 	four 	restricted 	industries 	after 	the 	ban's
relaxation 	will 	account 	for 	the 	same 	percentage 	of 	total
production 	employment 	as 	in 	the 	unrestricted 	apparel 	industries.
:
12

PRICE
QUANTITY
a
P 	= C I--------~-------- SUP PLY
B
Pc = c I ...-------~------"r_--- SU P PLY I
0=$89.97
c 	= P s = 	$19.79
I
c 	= P 	= 19.55	c
:
Q s = 	1,262.8 million dz. pairs gloves
Q C = 1,267.0 million dz. pairs gloves
FIGURE 	1
HORIZONTAL SUPPLY
13

.!
PRICE
a
PB
c
I
C
a 	= $89.97
c 	= 18.80
I
c 	=18.58
QUANTITY
.. PB = $ 1 9.79
19.57
1.262.8 million dz. pairs gloves
1.266.8 million dz. pairs gloves
FIGURE 2
UPWARD SLOPING 	SUPPLY
14