SOUTH AFRICA: THE ECONOMIC COST OF APARTHEID

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP85T00875R001600030116-0
• Release Decision: RIFPUB
• Original Classification: C
• Document Page Count: 36
• Document Creation Date: December 22, 2016
• Document Release Date: October 31, 2011
• Sequence Number: 116
• Publication Date: September 1, 1970
• Content Type: IM

 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
DIRECTORATE, OF
INTELLIGENCE
Intelligence Memorandum
ER IM 70-116
September,. 1970
Declassified and A roved For Release 2011/10/31 :CIA-RDP85T00875R001600030116-0
pp
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
WARNING
This document contains information affecting the national
defense of the United States, within the meaning of Title
18, sections 793 and 794, of the US Code, as amended.
Its transmission or revelation of its contents to or re-
ceipt by an unauthorized person is prohibited by law.
GROUP I
E,ctud,d Irnm o tomoIk
downpradinp and
ti~daulfimlion
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 CONTENTS
Page
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
Introduction . . . .
Background 1
The Restrictive System . . . . . . . . 5
Prospects for Growth . . . . . . . 7
Economic Cost of Apartheid . . . . . . . 12
Conclusions . . . . . . . . . . . . . 20
Appendix
Methodology
1. Projected Gross Domestic Product,
by Sector . . . . . . . . .
2. Projected Labor Demand, by Sector
3. Projected Factor Availabilities . . 1.0
4. Foreign Exchange Availability
and Requirements . . . . . . . . 12
5. Projected Output and Factor Returns
in Manufacturing . . . . . 14
iii
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
6. Projected Output in Trade . . . .
7. Prospects for Average Annual Rates
of Growth in the South African
Economy, 1970-75
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
Walvis Be
WALVIS RA
(Rap. of S.AfJ
~7U fH-WEST
(;nr.rnetional
LGde-lit Keetmenshoo
SOUTH
ATLANTIC Kara
OCEAN
colonial capital
0 Province capital
-~ ---~-~ Railroad
Road
sumeb
'Groottontein
AFRICA
r: Irbry)
Kimb
roonsta
Beth phem
FMASE
ESOTH
7. ast London
ort Alfred
Port Elizabebl
/Llm ?no
Messina
'drarltzburg
urban
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
SOUTH AFRICA AND SOUTH-WEST AFRICA
ZAMBIA
Z f>tnirr r
LAIR
nkle~
URY
18
SOUT F N
RHO !E, S A
~ t
\BLIPWay Fort Victoria
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
CENTRAL INTELLIGENCE AGENCY
Directorate of Intelligence
September 1970
INTELLIGENCE MEMORANDUM
South Africa: The Economic Cost Of Apartheid*
Introduction
South African racial policies are extremely
restrictive, and the range of economic activity
allowed the non-whites is markedly cir?:umscribed.
As a result, a widely voiced view holds that the
country is faced with a severe manpower shortage
that is restricting economic growth. Economic
pressures, it is argued, may force a relaxation of
at least some of the legislative and institutional
restrictions on the use of non-white labor and
permit increased participation of non-whites in
the economy. This memorandum attempts to evaluate
these economic pressures by assessing South Africa's
growth prospects within the present system and the
economic cost attributable to apartheid.
Background
1. Racial discrimination in employment has
existed in South Africa since the earliest white
settlement. These traditional practices were not
legally sanctioned until 1911 when certain jobs in
The term South Africa as used in this memorandum
includes the Republic of South Africa and the
International Territory of South-West Africa.
Note: This memorandum was produced solely by CIA.
It was prepared by the Office of Economic Receareh
and was coordinated with the Office'of Current
Intelligence.
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
the mines were reserved for workers with "certifi-
cates of competence," which were issued only to
white workers. The increasing unemployment of
unskilled whites during the economic depression
that followed World I, particularly in agriculture,
led to a series of violent strikes in the early
1920s by white workers who wanted the restrictive
system extended. Although the white labor movement
was unable to develop effective economic leverage,
it gained political control of the government in
1924 and, introduced' the so-called "civilized labor
policy."
2. The keystone of the "civilized labor policy"
was continued labor segregation with whites filling
skilled and supervisory positions, Asiatics and
coloreds filling semi-skilled and service positions,
and blacks filling unskilled positions.* The objec-
tive of the government's labor program through the
1930s and World War II was to assure the white
workers a high standard of living. The government
employed a variety of tools to force comp2i.-.nce,
including granting preference in government pur-
chases and tariff protection to those industries
that complied. In addition, a number of occupa-
tions were reserved for white workers in the mine3,
and government jobs were created.for the unskilled
white workers that industry and mining could not
afford.
3. Following World War II, the newly formed
Afrikaans-dominated Nationalist Party, responding
to tensions generated by the large influx of non-
whites to urban arecs during the war, adopted the
concept of greater geographic separation between
the races. Following the party's victory in 1948,
Dr. Henrick Verwoerd, later a Nationalist Prime
Minister, developed tha concept into the more com-
prehensive apartheid ideology of total racial sep-
aration. To implement apartheid, the government,
* The principal ethnic groups in South Africa are:
Asiatice, primarily Indian but including all members
of the Mongolian race except the Japanese who are
considered "honorary whites" (591,000); blacks,
descendants of any aboriginal group, particularly
the Bantu African tribes (13,340,000); whites, all
members of the Caucasian race (3,728,000); coloreds,
primarily descendants of mixed (black-white) par-
entage but including all persons not otherwise
classified (1,959,000).
-2-.
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
since 1948, has enacted legislation restricting the
economic, legal, social, and residential rights of
all people, particularly non-whites.
4. The economic intent of apartheid legislation
during the fifties and early sixties was similar to
that of the "civilized labor policy," The
Nationalists used the administrative and economic
tools at their disposal to maintain the existing
racial composition of the labor force. A number
of occupations were reserved by legislation for
whites in manufacturing, services, and mining.
Labor segregation also was perpetuated by restric-
tions embodied in labor contracts. Certain posi-
tions were reserved for white workers or for skilled
journeymen, who, because of racial restrictions in
apprenticeship programs, were white. Union insist-
ence on the "rate for a job," or equal pay regard-
less of race for a particular task, was another
measure that excluded non-whites from jobs that
traditionally had been filled by whites. Except
in unusual circumstances, employers found it uneco-
nomic "ill a position previously occupied by
whites,,knd for which a white pay scale had been
established, with a generally less productive non-
white.
5. The Nationalists, in the aftermath of the
Sharpeville massacre in 1960, extended their legis-
lative program to include influx controls (restric-
tion of the migration of non-whites to urban areas).
Under the Physical Planning Act, the establishment
or expansion of factories that would employ addi-
tional black workers required government approval
in 37 districts, which included most important
economic entities see the map). Blacks were
denied the right to acquire residential rights in
urban areas, and many workers' dependents were
moved to the reserves, which are considered the
legal home of the black population. Potential
workers on reserves, or so-called homelands or
bantustans, are required to deal with government
bureaus and are limited to one-year labor contracts,
after which they must return to their reserves
for at least one month.
6. Although blacks and other non-whites are
excluded from job opportunities in white South
Africa, they are, according to apartheid theory, to
have equivalent opportunities in their own areas.
- 3 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31: CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
SOUTH AFRICA AND SOUTH-WEST AFRICA
RESERVES AND BORDER AREAS
G O A z
N ZAMBIA ieAnn1n 30
I
O
L
SBURY
~ fPort) ~
Ln/fr1SAL
~
8
ankle
fig
4.4 'v SOUT
RHODESIA,,
0 fontoln auna
Bulawayo.Fort Victoria
fi N RO
Froncatown NlcW
' to Sort
~jjgg
WfNDHOEK
vis
A
BT S W A N A
Bn
ol
!
VVMLVIS BA 1Rop. of S.Af.)
24
4 -2R.
0
GABOkONE
A
U
I>a
R0 L
M U
MA
P
RERO jo
Sbutg
cz~
'
Kootmanshoop u,;;;I MBAR NE;
SOUTH-WES
AFRICA ,~. WA LAN
SOUTH (International
Territory)
? ATLANTIC
? i~CEAN .
Him rle y
Bloomfont MASER
30
LE50Taritzburg
urban
BAS 30-
Bltlarfonloln S O U T F R I C A
INDIAN
OCEAN
Somerset
E t
Ca
orcoster
TO
n art Elizabeth
50 I 0 l 0 MllaA
Irtt6
0 50 100 150 Kllnme
UNDARY KPRK LNTATIONI-
rrc
NOT NKcx$ ARIAUTIIORITATIV^
36 12 1
8 24 30
Transkei Legislative Authority Zulu Territorial Authority
Ovamboland Territorial Authority Matshangana Territorial Authority
Ciskei Territorial Authority Southern Sotho Territorial Authority
Tswana Territorial Authority Other native reserve
Lebowa Territorial Authority Border areas for industrial development
Venda Territorial Authority ~+ ~J Area subject to physical planning
HkFLRO Tribe
79626 7.70
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
The South Africans have, therefore, formulated a
program of homeland development. Adequate job
opportunities are to be created within the home-
lands to employ those who are unable to find jobs
elsewhere in South Africa. The homeland develop-
ment concept has not, however, been realized in
practice. In the Transkei, the most developed
bantustan, only about 10% of those in paid employ-
ment are employed within the homeland, Of this
number, 75% are either employed by the Transkei or
South African government or are domestic servants.
In all the homelands, only about 100 jobs have been
developed per year in manufacturing.
7. The stated objectives of apartheid policy
in the 1970s are extensions of those of the 1960s.
The government hopes to reduce and by 1978 to halt
and reverse the inflow of black workers to urban
areas and to rcnove most black dependents to the
homelands. In addition, the government seeks to
maintains labor segregation and the racial composi-
tion of the labor force.
The Restrirtive?Svstem
8. The econ(:..,nic effect of each of the three
principal restrictive instruments of the "civilized
labor" and apartheid policies -- government influence
to maintain the racial composition of the labor
force, job reservation, and "rate for a job" -- is
similar. All three perpetuate labor segreation
and protect the existing white/non-white labor
ratio. Government influence to maintain the racial
composition of the labor force, supplemented by
legislation requiring the maintenance of existing
white/non-white labor ratios in specific industries
or occupations, achieves this effect directly. Job
reservation and "rate for a job," although permit-
ting some work simplification through which a number
of non-whites may perform tasks previously performed
by white, generally prevent movement of non-whites
into white positions.
9. The E,!conemic effect of influx control varies
among geographic areas. in the 37 controlled dis-
tricts, influx control, when enforced, shifts the
racial composition of the labor force in favor of
whites, or at least non-Africans. This restriction
on labor'supply probably slows the industrial growth
of the area itself. In non-controlled districts,
- 5 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
influx control leaves the racial composition of the
labor force unchanged, and the resulting avail-
ability of labor may, particularly in the border
areas, stimulate industrial growth.
10. Enforcement of the restrictive practice;
and influx control measures has been flexible, and
exemptions generally have been granted in cases
where manpower shortages have been relatively
severe. Government administrative and legislative
power has been used sparingly to limit job simpli-
fication and assumption by non-whites of tasks
formerly perfor.:med by whites. in government enter-
prises, particularly the railroads, certr.in un-
skilled jobs traditionally occupied by whites now
are being filled on a temporary basis by non-whites
with the consent of the white labor unions. Trade
unions generally agree to non-whites assuming cer-
tain tasks previously perf,7rmed by whites when
white job security is guaranteed and the wages of
white workers are increased. Partial exemptions
also have been granted in a majority of cases where
job reservation determinations have been made.
11. Exemptions from influx control regulations
have been numerous. in 1.468, the first year of
the Physical Planning Act, exemptions were granted
to employ about 65% of the blacks for whop, exemp-
tions were requested, Only about 22% of the annual
increase in black employment outside of agriculture
is absorbed in tre border areas; the remainder con-
tinues to be eiployed in traditional locations.
12. In spite of these numerous compromises,
however, labor segregation essentially has been
realized. The white and non-white labor markets
have been separated successfully., especially in
manufacturing, and substitution of the labor of.
one race for another has been reduced substantially.*
The racial composition of the labor force has been
maintained desiite s white labor shortage (as evi-
denced, in part, by 'a white unemployment rate of
less than 0.5%) aye ; a non-white labor surplus '(as
evidenced by the.e3:~nss of non-white labor in agri-
culture) In manuf: -j' turfing, probably also in
mining, and in secondary industry, the sub-
atitution of non-whites for whites has been minimal.
In trade and in ither services, although substitu-
ti_cn has taken place, the ease of substitution has
been reduced substantially.
For empirical verification of this statement,
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
Prospects for Growth
13. Growth of the South African economy has
been impressive despite the apartheid labor restric-
tions. Since 1933, real gross domestic product
(GDP) has grown at an average annual rate of 4.5%.
After World War II, growth accelerated and has
averaged almost 6% annually in the past ten years.
14. The 6% rate of real growth achieved in the
last decade exceeded the planned rate and no doubt
was more than acceptable to South Africa. Under
the. Economic Development Programme for 1968-73,
the South African economy is to grow at a rate of
5.5% annually (see Table 1), with the present
system of labor restriction remaining essentially
unchanged. Manufacturing output is to grow at
a rate of 6.6% and output in trade at 5.4%. The
demand for white and non-white labor in manufac-
turing is expected to increase through 1973 at 3.3%
and 3.9% annually, respectively, and the stock of
fixed capital at about 8% (see Tables 2 and 3).
15. South Africa should be able to achieve at
least the planned growth in output, given the pro-
jected supply of domestic factors of production.
If employment and capital stock in manufacturing
increase as planned and technology and other in-
fluences on productive efficiency improve at the
same rate as in the past, then a growth rate of
7% a year for manufacturing output -- slightly
above the planned 6.6% rate -- seems feasible with-
in the present restrictive system.* In trade,
too, the planned growth rates for white and non-
white labor -- 2.5% and 2.8%, respectively -- appear
adequate under the present restrictive system to
increase output in trade at about the planned 5.4%
rate during 1970-75.**
16. Planned growth rates for other economic
sectors -- mining, other secondary industry, other
services, and agriculture -- also appear to be
realistic, given the input assumptions of the Pro-
gramme and continuation of the existing restrictive
* Based on a production fpnetion derived for the
present restrictive system. See Appendix TabZe Al,
equation 2'
** See Appendix Table A2, equation 4.
7
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
Table 1
Projected Gross Domestic Product, by Sector a/
Million Constant 1969 US $
Average Annual
Rate of Growth
1970-75
1969
1970
1971
1972
1973
1974
1975
(Percent)
Total
14,630
15,420
16,260
17,150
18,095
19,085
20,130
Mining
1,725
1,770
1,820
1,870
- 1,920
1,970
2,020
2=7
Secondary industry
:4,035
4,310
4,600
4,915
-5,250
.5.,605
5,985-
Of which:
Manufacturing
3,045
3,245
3,460
3,690
3,930
4,190
4,465
Services.
.7,390
7,810
8,255
8,725
9,225
9,750
10,305
Of which:
Trade
2,035,
2,145
2,260
2,385
2,510
2,650
2,790
5.4
Agriculture
1,,480
1,530
1,585
1,640
1,700
1,760
1,820
3.5
a. Entries for the period 1969-73 are taken from the South African Development Programme
for,1968-73; entries for-1974 and 1975 are estimates derived on the assumption of a con-
tinuation of Programme trends.
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
Projected Labor Demand, by Sector a/
Labor demand (whites)
Mining
Secondary industry
Of which:
Manufacturing
Z Services
Of which:
Z 1O Trade
~.~ Agriculture
> Labor demgznd (non-whites)
.:Mining
Secondary industry
Of which :
Manufacturing
Services
Of which:
Trade
Agriculture
Thousand Persons
Average Annual
Rate of Growth
1970-75
1969
1970
1971
1972
1973
1974
1975
(Percent)
1,384
1,423
1,460
1,496
1,534
1,572
1,611
59
59
59
58
57
57
57
345
358
371
384
398
413
428
3.7
275
284
294
303
313
323
334
3.3
865
891
916
940
966
990
1,014
2.7
211
216
222
228
233
239
245
2.5
115
115
114
114
113
112
112
--
5,382
5,516
5,654
5,797
5,945
6_,098
6,257
540
537
534
532
529
526
524
--
1,082
1,130
1,180
1,232
1,286
1,343
1,402
4.4
815
847
880
914
950
987
1,025
3.9
1,913
1,976
2,041
2,108
2,178
2,250
2,324
3.3
247
254
261
268
276
284
292
2.8
1,847
1,873
1,899
1,925"
1,952
1,979
2,007
1.4
a. Entries for the. period 1969-73 are taken from the Sauth African Development Programme
for 1968-73;entriea for 1974 and 1975 are estimates derived on the assumption of a con-
tinuation of Programme trends.
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
Table 3
Projected Factor Avaaaabilities a/
Current account foreign
exchange availability b/
Labor availability c/
Whites;
With 20,000 immigrants
annually
With 30,000 immigrants
annually
Non-whites
Capital availability d/
Total
Mining
Secondary industry
Of which:
Manufacturing
Services
Of which:
85 90 95 100 105 110 --
85 95 95 95 100 100 3.8
a. Data for 1969-73 are from the South African Development Programme for 1968-73; data for
1974-75 are estimates derived on the assumption of a continuation of-P.rogramre trends.
b. Foreign exchange available to finance imports of goods and non-factor services. in
order to estimate the total foreign exchange available to finance imports of goods and non-
factor services, net private capital inflows and long-term public capital inflows must be
included. Net private capital inflows have fluctuated widely but averaged $320 million
during each of the last three years, and long-term public capital inflows have averaged
$15 million.
c. Economically active population. V
d. Net investment or additions to capital stock.
t9o9 1970 1971 1972 1973 1974 1975
Million Ccnstant 1969 us $
Average Annual
Rate of Growth
1970-75
-(Percent)
3,360 3,490 3,620 3,755 3,900 4,005 4,200 3.8
Thousand Persons
-- 1,459 1,491 1,524 1,558 1,592 2.2
1,390 1,427 1,463 1,499 1,537 1,575 1,614 2.5
5,580 5,711 5,846 5,985 6,128 6,275 6,426 2.4
Million Constant 1969 US $
2,030 2,125 2,230 2,335 2,450 2,565 2,685 4.8
65 70 70 80 70 70 70
465 485 505 530 560 585 615 4.2
385 400 415 435 460 480 510 --
1,420 1,485 1,560 1,630 1,725 1,810 1,900 5.0
Trade 80
Agriculture 80
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
system. The situation in mining and in other
secondary industry is similar to that in manufac-
turing, while the situation in trade is probably
fairly representative of the entire service sector.
Few restrictions exist on non-white employment in
agriculture. Consequently, achievement of planned
growth for the economy as a whole seems compatible
with the present restrictive system, making the
case that easing apartheid is necessary to achieve
planned growth difficult to support.
17. Indeed, growth may not approach a rate at
which labor will become an effective constraint
because another constraint -- the supply of foreign
exchange -- probably will become operative first.
The South African economy requires increased imports,
especially of investment goods, to support increases
in output. If past relationships between imports,
GDP, and investment continue, imports will have to
increase 4.8% a year to support planned growth of
the economy. But the supply of foreign exchange
to finance imports will grow at an average annual
rate of only 3.8% according to Development Pro-
gramme estimates (see Table 3). Gold output is
expected to remain almost constant for the next
several years before declining, while income from
other exports is expected to rise at about the
same rate as real product, or at about the 5.5%
rate achieved annually during the 1960s.
18. The current account of the balance of pay-
ments probably will be in deficit through most of
the 1969-75 period, and capital inflows, which
averaged $335 million during each of the last
three years, are unlikely to be large enough to
cover the entire deficit (see Table 4). Foreign
exchange earnings and capital inflows almost cer-
tainly will not be large enough to support for long
a growth rate high enough to generate labor shortages
that could possibly be eased by relaxing restrictive
policies. They may even be inadequate to support
the growth rate envisioned in the Programme.
19. E43ing apartheid regulations would probably
have little impact on export earnings in the short
term. The effect on exports of a relaxation of the,
restrictive labor practices normally would be to
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
Table 4
Foreign Exchange Availability and Requirements
Million Constant 1969 US $
Foreign Exchange
Available to Estimated Foreign
Finance Imports Exchange Require-
Year
of Goods and ments for Goods
Services and Services
(Excluding Fa for (Excluding F for
Payments) 1 - Payments)
Balance
1970
3,785
40
1971
3,953
.3,975
-20
1972
4,090
4,175
-85
1973
4,380
-145
1974
4,340
4,615?
-275
1975
4,535
4,860
-325
a.
Official pro eotion through 1973, continues' to
1975, for current account earnings plus assumed
earnings of $335 million a year on capital account.
b. Estimated for the output and investment assump-
tions;of the Economic Development Programme ex-
tended through 1975, using a statistical regression
(see the Appendix).
reduce unit costs of production and make South
African exports slightly more competitive. Because
of the peculiarities of mining technology, however,
a reduction in labor costs and hence in unit extrac-
tive cost would enable the mines to extract profit-
ably a lower average grade of ore than they now do
and thus could actually decrease gold output and
exports in the short run.
Economic Cost of Apartheid
20. The apartheid system exploits non-white
labor. Non-white workers are paid wages far below
what they would receive in a free labor market -- or,
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
in economic terms, far below the value of their
marginal product. While their productivity on the
average is probably about 30% that of whites in
manufacturing, their average wage is only 14% of
the average white wage. This exploitation is
possible because of the considerable surplus of
non-white labor and the restrictions on the sub-
stitution of non-whites for whites. Employment
opportunities in agriculture -- the only sector
where restrictions against non-white labor nre
minimal -- are very limited, and wages there are
far below those paid fui urban occupations. Non-
white labor's share of income originating In manu-
facturing is only 30% while it would be about 37%
if the labor market were free, even if education
and skills remained unchanged (see Table 5). The
beneficiaries of the system are, of course, the
white workers, whose wages are h..gher than they
would be in a free labor market and who atffer
virtually no unemployment.
21. Apartheid creates a misallocation of
resources by barring non-white workers from occupa-
tions where their productivity would be higher.
It holds down the size of the non-white labor
forces in secondary industry and mining, where pro-
ductivity is relatively high, and to a lesser ex-
tent in trade and other services. At the same
time it keeps non-white labor in traditional agri-
culture where labor productivity is very low. In
manufacturing, where the restrictions are most
stringent, it tends to give white and non-white
workers complementary rather than competitive
functions.
22. It is impossible to measure the overall
cost of apartheid, because the system has existed
as an integral part of the economy's development
since the 1920s and even before. The total economy,
including the sectoral distribution of labor and
capital, the technology, the educational system,
the composition of trade, and the pattern of demand
have all been structured by the restrictive system.
We can, however, approximate the cost that the
apartheid system exacts as new resources are allo-
cated.
23. We measure the cost of apartheid by esti-
mating the difference between the potential growth
of output under the existing restrictive system
- 13 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
Table 5
Projected Output and Factor Returns in Manufacturing
Million Constant 1969 US $
Output
1969
Projected in
Economic Development
Programme
Projected for the
Present Restrictive
System
Projected in a Free
Labor Market
Situation
3
045
,
3
045
,
3
045
1970
,
3
245
,
3
260
,
3
275
1971
,
3
460
,
3,490
_3
520
1972
,
3,690
3,736`
3
775
1973
,
3
930
,
3
990
,
4
050
1974
,
4
190
,
4
265
,
4
345
1975
,
4
465
,
4
560
,
4,660
Average annuag
rate of grorsth
(percent)
Year
1970
1975
Percent of
output
Present Restrictive System
Free Labor Market Situation
Return to Return to Return to Return to
White Labor Non-White Labor CzPital White Labor
1,565 980 715 1,345
2,190 1,370 1,000 1,910
Return to Return to
Non-White Labor Capital
1,310 720
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CON FI I)EN';I:'I AL
and the growth that might be achieved if the labor
market in South Africa were free. There are three
stages to the analysis:
-- The effect of apartheid on the
growth of individual economic sectors
using the soctc,ral, tabor projoctions
of the Economic Development Programme;
-- Its affect on overall economic
growth through a redistribution of
total planned employment among sectors;
Its effect if the overall em-
ployment projections of the Programme
are themselves changed.
24. The cost of the misallocation of labor
within sectors was estimated by employing the white
labor, non-white labor, and capital inputs projected
in the Programme in production functions re''lecting,
alternatively, the present restrictive system and a
free labor market.* In manufacturing the present
restrictive system, which allows virtually no suL-
stitution of non-whites for white labor, reduces
the annual rate of growth of output by 0.4% compared
with a free labor market situation (see Table 5).
The reason the effect is so small is that the
Programme cal},s for only a slightly faster growth
of non-white than of white employment -- a pattern
consistent with trends of the past 30 years. In
trade the additional cost of the existing system
shows up as insignificant, partly because the system
allows limited substitutability between the two
types of labor and partly because their growth
rates are very close (see Table 6). No calculation
was attempted for agriculture, because of the
absence there of any effective restrictions f,n the
use of labor.
25. The effects of easing labor restrictions
on intersectoral shifts in employment are difficult
In teohntca1 turno, white and non-whits labor
inputs are ropruoonted in production functions
uoing slaoticitieo of oubatitution of ac ro for
manufacturing, sin' ti for trade, and infi.rz.ty for
a fro* labor market. Soo Appendix Table Al, equa-
tion I and oquart.on 2'; and Appendix Table A2,
equation 4.
- 15 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
Table 6
Projected Output in Trade
1969
1970
1971
1972
1973
1974
1975
Average annual
rate of growth
(percent)
Projected in
Economic Development
Prograi e
2,035
2,145
2,260
2,385
2,510
-2.650
2,790
Million Constant 1969 US $
Projected for the Projected in a Free
Present Restrictive Labor Market
System _ Situation
2,035
2,140
2,250
2,370
2,490
2,620
2,755
2,035
2,140
2,255
2,375
2,495
2,625
2,760
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
to evaluate. Presumably, some increase in the
growth in non-agricultural employment of non-whites
at the expense of agricultural omployment would take
place, but the extent of the shift would depend on
the time allowed for adjustment, the volume of
investment, and many other factors. One possibility
is that the non-white labor force in agriculture
would remain constant instead of increasing as
called for in the Programme. If this additional
labor were shifted to other sectors in proportion
to their projected employment levels, the growth
of manufacturing output could increase about 0.2%
a year and that of trade about 0.1% a year, Thus
the total impact on both intrasectoral and inter-
eectoral shifts in employment would be about 0.6%
a year for manufacturing and about 0.1% for trade
(sea Table 7). Assuming, as we believe is reason-
able, that the effect of apartheid in mining and
other secondary industry is similar to that in
manufacturing and the effect in other services is
similar to that in trade, and assuming that with-
drawal of labor from agriculture has no effect on
agricultural output, the total coat of apartheid
to the economy comes to about 0.2% a year.
26. These estimates accept the overall labor
force projections of the Programme. However, the
white labotc force projections are sensitive to
changes in the number of immigrants. Since 1961
the South African government, pursuing a vigorous
campaign to encourage white immigration, has suc-
ceeded in raising the net number of white immigrants
to more than 30,000 per year, and immigration at
the rate of 30,000 to 40,000 annually is likely to
continue through the 1970s. If the South Africans
were unable t~: continue to attract this number of
immigrants, the cost of maintaining the present
restrictive system in manufacturing would become
slightly greater -- although still not substantial.
27. If net white immigration declined, for
example to 20,000 immigrants tier year, the growth
rate of the economically active white population,
and consequently the growth rate of the white
labor force, would decline by 12%. If the govern-
ment tried to maintain the planned white and non-
white labor ratio by reducing the growth of non-
white employment, the effect on manufacturing
would be to reduce output by about an additional
- 17 V.
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
Table 7
Prospects for Average Annual Rates of Growth
in the South African Economy
1970-75
Percent
Manufacturing Trade Economy
Estimated for Estimated for Estimated for Estimated for Estimated for Cost in Terms
a Restricted a Free a Restricted a Free a Restricted of Forgone
Labor Market Labor Market Labor Market Labor Market Labor Market Growth
Projected in the
Economic Develop-
ment Programme
Estimated on the
basis of the in-
put assumptions
of the Progran~e
Estimated on the
assumption that
underemployed
agricultural labor
is reallocated
Estimated on the
assumption that
net white immi-
gration declines
to 20,000 per
year
Estimated on the
assumption that
influx control
measures are
enforced 6.2
a. Assuming tit underemployed agricultural Labor is reallocated in a free Labor market.
-0.5 a/
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
0.1% a year from what could be achieved in a free
labor market situation (see Table 7). In trade
and in the economy in general, if the government
sought to maintain the present restrictive system
in the face of decreased immigration, the additional
loss in output would also be about 0.1% a year.
The effects on output of various white and non-
white employment assumptions are summarized in
Table 7.
28. It is also of interest to consider the
effect on economic growth of even tighter restric-
tions -- of enforcing influx controls in manufac-
turing and accelerating development of border areas.
About $250 million has been spent in the border
areas by the government since 1960 for such items
as infrastructure, rail rebates, and loans. Most
firms that move to the border areas do so because
of the prohibitions of the Physical Planning Act,
however, rather than in response to the government
incentive program. The real cost of influx control
is, therefore, probably greater than the cost of
government incentives alone.
29. If influx control measures were enforced
strictly, then the growth of the black labor force
would be restricted to jobs created in border
areas and the growth rate in the entire non-white
labor force in manufacturing would be reduced to
less than 2%. The average annual growth in manu-
facturing output consequently would be reduced by
approximately 0.8% from what could be achieved
without influx control and about 1.4% compared
with a free labor market situation (see Table 7).
If, as seems likely, strict influx control enforce-
ment induced increased job formation in border
areas, the cost would, however, be substantially
less.
30. The foregoing discussion suggests that the
amount of economic growth forgone because of
apartheid policies is bound to be small in the
next few years. Moreover, the growth forgone is
in most cases much smaller than the loss of income
white labor would suffer if the market became free.
This situation might not be so in the long run if
the non-white population were able through educa-
tion and experience to raise its productivity. In
the short run, however, a change to a free labor
market would affect the income distribution more
quickly than it would affect the allocation of
V. 19
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
l&.'or and its productivity. The only case dis-
cvssed above that might entail some cut in white
Libor income is the tight application of influx
controls, which no doubt is why this measure has
encountered opposition from a large part of the
white population. Apartheid probably reduces the
return to capital as well as non-white wages
(although, of course, much less), but it is the
white workers who have the votes. Since the supply
of foreign exchange is likely to be more if a
constraint on economic growth in the next i'ew years
than is the supply of labor, the chances seem slim
that economic pressures to weaken apartheid will
become stronger. Indeed, if foreign exchange
shortages should force a marked economic slowdown,
white labor would probably press for even tighter
restrictions on non-white employment to protect its
position.
Conclusions
31. Racial discrimination in employment has
successfully separated the white and non-white
labor markets in the South African economy. The
separation has been accomplished and enforced
through job reservation, restrictive labor con-
tracts, and institutional segregation. Of the
various sectors of the economy, manufacturing is
most affected by the apartheid system.
32. Economic pressures to relax restrictive
labor practices are unlikely to grow in the next
few years. The supply and planned distribution of
labor in South Africa during the first half of the
1970s appear to be adequate to achieve the planned
rate of real growth in GDP of 5.5% per year.
Indeed, the effective constraint on growth may not
be labor but rather foreign exchange, the avail-
ability of which is not very sensitive to the
degree of labor market restriction.
33. Maintaining the present restrictive system
in the South African economy will exact only a
small economic cost. Although the apartheid system
exploits non-white labor, it probably will not over
the next few years seriously inhibit economic
growth. We calculate that maintaining the present
- 20 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
restrictive system retards annual growth in GDP
by only 0.2% from what could be achieved with the
same total employment in a free labor market
situation. This small economic cost, moreover,
is borne by the non-whites, the population group
least able to exert pressure to effect change.
34. The recent extension of apartheid legisla-
tion to include influx control in manufacturing has
not been strictly enforced. Strict enforcement
would increase the difference in growth rates to
1.4% in manufacturing and by about 0.5% in the
economy overall. This cost would probably be borne
by the white population to a greater extent than
has been the case in the existing restrictive
system. The opposition of the potentially affected
whites will probably inhibit strict enforcement in
manufacturing and prevent the extension of influx
control to other sectors.
35. In spite of the absence of economic pres-
sures to force a relaxation in the present re-
strictive system, the sharp differences that exist
between apartheid in theory and in practice probably
will widen although slowly. The government is not
likely to promulgate the job reservation determina-
tions in the service sector that were announced
during the recent election campaign or seek to
remove the numerous job reservation exemptions
that previously had been granted. The government
also probably will not seek to prevent the slow
increase in the non-white/white labor ratio in the
non-agricultural economy.
- 21 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
Methodology
Description of the Production Functions
The statistical analysis of the effects of
apartheid makes use of production functions in
manufacturing and trade of the general form:
Y= ALa Kl a e X t
where Y = output, L = employment,
K = capital stock, and t = time
in this function, labor and capital inputs are com-
bined by means of constant weights, which means that
the elasticity of substitution (a measure of the
ease with. which inputs can be substituted for each
other in production) is equal to one between these
inputs. The coefficients a -and (1-a) represent the
contribution to production of labor and capital,
respectively (and, if labor and capital markets
are competitive, also measure their income shares).
The coefficient A incorporates new technology,
qualitative improvements in labor and capital
inputs, changes in the use of other unspecified
inputs, and all other factors not accounted for
by quantitative changes in labor and capital. The
coefficient A is a scale coefficient. Within this
general, "Cobb-Douglas" function, the labor input
was, for the purpose of analysis, broken down into
its white and non-white components. Various rela-
tionships (elasticities of substitution) between
white and non-white labor were assumed to represent
different labor market situations.
If the South African labor market were free
from racial restriction, production could be ex-
plained through the use of a function of the form:
(1) Y = A (Lw + YLnw) aK (l a) e .Xt
- 23
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
UUNP'!JiJ;N'. IAL
where Lw a white labor employment,
Lnw - non-white labor employment and
y is a weighting factor to combine
labor of different product] ;?i4-.ies
into a component labor pool.
We employ this function to estimate South Africa's
growth possibilities in a free labor market situa-
tion. On the basis of the available data on educa-
tion and relative productivity, non-white labor is
assumed to be about one-third as efficient as white
labor -- that is, the labor weighting factor is
assumed to be equal to 0.3. Labor's share, cap-
ital's share, and the rate of technological prog-
ress are set on the basis of estimates empirically
derived, and thus are affected by the restrictive
system.* This same function also is used to esti-
mate South Africa's growth possibilities in a
situation in which influx control measures are
enforced. In this case, non-white employment is
set equal to the present number of non-whites cur-
rently employed plus the number that might be
employed in border areas and the number of available
Asiatics and coloreds.** Implicit in the free labor
market function is the assumption that the partial
elasticity of substitution among types of labor is
infinite -- that is, that white and non-white labor
can be substituted for one another without affecting
their average productivities.
* By setting the rate of technological progress
at the level estimated for the present restrictive
system, there is some danger that the growth possi-
bilities in a free Labor market situation and conse-
quently vhe cost of maintaining the present restric-
tive system may be overestimated. The residual
estimate for the restrictive system not only may
embody a measure of the rate of technological
progress but also may include some measure of the
increased efficiency arising from changes in the
racial composition of the labor force -- a factor
that is measured directly in the free market function.
** it is recognized that equation (.1) does not
accurately re flect the influx control situation,
but it is a closer approximation than the other
equations, given that non-white labor is the slowest
growing labor category when influx control is en-
forced and that whites can substitute for non-
whites in production.
- 24 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CONFIDENTIAL
if the re .itrictive system were enforced strictly,
no substitution of labor of one race for another
would be possible; white and non-white labor would
enter production in a fixed ratio (the partial
elasticity of substitution would be zero). Production
under an enforced restrictive system could be ex-
plained through the use of a function of the form:
(2) Y = A (min L) (1 K('"a) a At
where (min L) is a measure of the slowest
growing labor category
In practice, the restrictive system has not been
enforced strictly, as is evident from the slowly
rising non-white/white ratio. Function (2), there-
fore, has been reformulated in terms of percentage
rates of growth over time. This new fc.:mulation is
based on the assumption that at any particular time
the use of white and non-white labor reflects the
accepted institutionally determined white/non-white
labor ratio appropriate for that time, and that
over time this ratio is changing.* The respe::ified
function assumes the form:
(2') y = A + a (min 1) + (1- a) k
where y, 1, and k are annual per-
centage rates of growth for the
corresponding outputs and inputs,
and A measures the residual con-
tribution, as does A of equation
(2).**
* The assumption is embodied in the function in
the property that, in any time period ti, the white/
non-white ratio is assumed to be in equilibrium,
and changes only in the slowest growing labor cate-
gory are used to predict output in ti + 1. In
ti + 1, when prediei,ing output in ti + 2, the
ti + 1 white/non-white ratio is then assumed to
be an equilibrium ratio. Although this condition
is obviously artificial, it is a reasonable approxi-
mation to the basic assumption.
** Although (min 1) is a measure of the slowest
growing labor category either white or non-white,
in practice in most years the white category is
the slower growing.
25 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CON1"J )I NTIAL
The free labor market production function of
equation (1) and the restricted labor market pro-
duction function of equation (2) represent two
extreme assumptions about the substitutability
between labor types. Both functions, as well as
all intermediate cases, including the case in
which the partial elasticity of substitution among
types of labor (as well as the elasticity of sub-
stitution between labor and capital) is equal to
one,* are forms of a more general function. This
more general and elaborate function permits the
partial elasticity of substitution between labor
types to be determined from the data. The produc-
tion function is of the form:
(3) Y m A ((1 -Y') Lw P +Y'Lnw P] a/PK(l-a ) e At
where y'is a productivity parameter
equivalent to the y of the free labor
market production function of equa-
tion (1) and p is related to the
partial elasticity of substitution
between 1&bor types by the relation:
elasticity of substitution = 1
p
The distribution of income between labor and
capital and within the labor category can be de-
rived directly from most production functions. If
both the labor and capital markets were competitive,
a would correspond to labor's share of production
and (1 - a) to capital's share. If the South
African labor market were free from racial restric-
tion, non-whites as a group within the labor cate-
gory would receive that proportion of labor's share
determined by the relation:
(Lw + YLnw)
If the South African restrictive system had been
strictly enforced, it would be impossible to
See equation (4), below.
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CON Fl )EN11AL
separate the contribution of the two typed of labor,
The distribution of income between labor types
would, therefore, be a non-economic question.
Description of the Data and Results
Adequate data were available only for manufac-
turing and trade. In manufacturing, a time series
of annual observations of white employment, non-
white employment, and capital stocP in place for
the period 1933-63 was used. In trade, where
adequate data were available only on a cross-
section rather than a time series basin, a series
of observations of white employment, non-white
employment, and fixed capital in each of 77 branches
in 1960 was used. To project output tin trade, it
was assumed that efficiency (the o term in the
production function) increased 2/% a year.
The values obtained for the parameters and
test statistics for each of three of the principal
functions tested in manufacturing are presented in
Table Al. The R2, a measure of the explanatory
ability of the function, is the proportion of the
variation in output explained by the linear in-
fluence of labor, capital, and time. The Durbin-
Watson d statistic is a measure of the degree of
first-order auto-correlation of the residuals. A
d statistic near 2 indicates that one of the
crucial assumptions of the underlying statistical
model, the assumption that residuals are statis-
tically independent of the previous residual value,
is valid. The numbers in parentheses immediately
below the estimates for the parameter value are
the individual confidence limits for each para-
meter. The probability that the true value of the
parameter will be within the confidence interval
is approximately 95%.
? In manufacturing, the restricted labor market
function of equation (21), with its assumption of
zero elasticity of substitution between labor types,
? seems to provide the best estimate of the South
African production relationship. If we use the
free market equation, although the R2 is high,
the calculated contribution of non-white labor is
practically nil -- an unreasonable result if the
labor market were free. This result is compatible,
- 27 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
C t)N1''I I)EN'l'IAL
however, with the aneumptione of equation (2'), an
an increase in the non-white labor input would con-
tribute little to production in a restricted labor
market situation where, an in South Africa, white
labor in generally the slowest growing labor input.
The confidence intervals for the parameters in the
restricted labor market function are much smaller
(the parameters are more si_-nificant), although the
actual parameter estimates themselves are quite
similar. The estimate of labor's share (and of
capital's share) in reasonable in both chess, labor's
share being independently estimated at about 79%
of output.
An attempt to estimate the elasticity of substi-
tution between white and non-white labor by moans
of equation (3) was unsuccessful. The estimating
procedure terminated when p became very large. A
large p corresponds, however, to a small elasticity
of substitution -- another indication of the suita-
bility of the restricted labor market function of
equation (21).
in trade, both the free labor market function
of equation (1) and the general .unction of equa-
tion (3) appear to provide fairly good estimates of
the production relationship. Although of the two
the free market production function has the higher
R2, the estimate of labor's share seems to be
unreasonably low in that function, and the parameter
estimates are less significant. The results for
the function of equation (3) indicate that the
elanticity of substitution between labor types is
equal to about one. If this result is valid, the
number of parameters to be estimated msy be reduced
and the function substantially simplified by trans-
forming the function into the form:
(4) Y d A Lw' Lnws K 11 - (a' + BI )l
where a'and B' equal the elasticity of
output with respect to white and non-
white labor, respectively, and a'
+ B' can be taken to be equal to
labor's share.
- 28 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
(:0N1:11)EN'1'1Al.
The results for the function of equation (4)
are substantially better than the roeulte either
for function (3) or the free labor market function.
'the function has the highest fit, while the eati-
rrate of labor's combined share (and of capital's
eharo) and of the proportion of labor's combined
share going to white labor (and to non-white labor)
are aaoonable, these shares having been inde-
pendently estimated at about 61% and 00%0 reepec-
tively. overall, it seems reasonable to conclude
that production in trade in South Africa can be
dart:ribed adequately by the function of equation (4).
Analysis of the Import ate uirementn
To determine the level of domestic product and
investment compatible with foreign exchange avail-
ability, a multiple regression on imports was used.
Imports of goods and services, excluding factor pay-
mente, were used as a meanure of imports. Gross
domestic capital formation by pilvate enterprise
was used an a measure of investment, and groan
domestic product less investment was used an a
mbasure of other domestic expenditure. Time,
measured in number of years from 1946, was included
an variable in the regression to measure the effect
of the government's import substitution program.
The results are presented in Table A3. The
parameter results indicate that the ratio of imports
to increases in GDP is 0.24 and to increases in
investment in 0.30.. The government's import sub-
stitution program has been quite successful, how-
ever, '.nd in reducing imports for a constant level
of product and investment by more than $21 million
a year.
- 29 -
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875
Prtauction Psusctionj to
4010.4 z Is
di! i2 d Stati.xtic X k os .t' r ~.~~
Pre* labor sarkst function
(1) Y - A 1Lv ? ILmr)4 x (1-*) it 21l .91
1.02 O.99 a.114 .1 a.13000+11 (; .Z34
(1.tk to dG.~}L to (~.43L ta. C..Q' t.
tt.79}( 1).4(I1 -4.t??) Q.ti.
Restricted labor market function
(2+ Y - A* + a(a.in 1) ? (l- s) k 23 .9$ b/ 2.05
(4 . ? S2 to 01121 to
0.449) Q.6)S)
t~tcn
H+E =a+sean
a %,.
a. A. ~uu~ i(w t~ sot o+e.
b. For trasajoritd fwnredon -? :AA faftif(c maa trac~t,ure-~cd (6V
from actual cep s~t ds eu gr. icia y?mr asd Cho tssf a*~ psrc~rs: r at~sxr s :- cu..I7%: Y Ca 'nci.'ls il.
3:x:;~r? sot 'y
comparable to t. la vtktr j%notfa+it.
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
C( NHJ )EN
Tatsle A2
1'roductlt)t) 1'uticti on
Mode 1.
1t r2 c1 f;tatiptic
rrno labor market function
C,enera1 function
? ^ J ~ALttFr d 1. ft f: Ju' ~~4+T t LOO .
74 .01 1.57
(1- )
73 .74 i.80
74 .94
1.56
Intercept
A
0.730
(0 .812-o.6G'.
0.900
(1..00-0.796)
0.932
(1 .04-0.82)
a
- 31 -
CONFIDENTI
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
 Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0
CON'FI1-)I;N'I1, I?
'ct!on )'uttcUotis ft Trade
rlagtieity
of Subgti-
tutioti
Pooling
Ike tote e-n
InLoreelrt Fader Labor Share Labor White Labor ton-White
A it or _ ypL~g_ . 1 ; a r e Labor Shat
0.900 0.367 0.539 1
(1.00-0.796) (0.659-0.075) (0.664-0.414) (--0.327)
0.701 1 2J
0.591 0.110
(0.733-0.449) (0.166-0.05,
CONFIDENTIAL
Declassified and Approved For Release 2011/10/31 : CIA-RDP85T00875R001600030116-0