Global constraints and a new vision for development—II

TFCHNOLOGICAL

FC.,RECASTIN(;

AND SOCIAL CHANGE 6.37 I-388

f 1974)

371

onstrain and a New Visio for Developm IO

YOICHI KAYA and YUTAKA XIZUKI

ABSTRACT in considering the limits to growth of our earth, we deal here (PII t II) with the poverty and inadequate growth rate of Ihe developing nations. Development is i:npeded by rapid populatton growth. chrorlic malnutrition, and often b) a state of unstable pohricd order. A further impasse is foreseen when the manufactured goods of developing nations begin to compete with those of the advanced nations in the world market. We conclude that in the interests of progress, the advanced nations must take positive steps to reform the structure of their own indusery, and we have investigated lvhat kind of reform in the industry pattern of the world would be most desirable. Thij is the sequel tr Part I, in the previous issue, which dealt with depletion of the world’s mineral resourc-es.

A New Plau for Development: Worldwide Redistribution of Industq. to Help the Developing Nations 2.1. THE S7‘ARTING POINT t:OR 0l;R

PLAN

It is hardly an exaggeration to say th2.t alleviating the poverty of the vast number!; of people living in the ljevelopiilg nations is the greatest world problem we CV.C’ today. While the I&A, Japan, and the countries of Western Europ; have advanced to a civilizsiic: enjoq ing :1 GNP of several thousand dollars per person, the developing cations have a per c:lpita GNP of less than $X0 a year; chronic malnutrition and hidden unemployment are rampant. To promote development and reduce this great gap between the northern and southern nations, various efforts, usually in the form of economic aid, havd been made. Even prescinding from the significance of the aid given so far, we must admit that it has not effectively achieved its purpose. The economic growth rate of the developing nations shows only a weak iitatistical correlation with the quantities of aid which have been received [Ref. 2, Part 1, pre Iious issue 1.

Historically, most of the advanced nations developed tllernselvcs without relyink. on any special 3iti from o;ltsic:c. In the li&t of this cxperiencc, it is natural to look for the developing nations to follow this same path without being overly depenclW on econcmic aid. Precisely in order to discover concrete ways in which this might be done have WC cntcred on this study. --Y0tCHI KAYA (Univ. of ‘rdcy~)j md YUTAKA SUZUKI (Osaka University) are nicmbera oi the Japan Wclrk Team of the Club ,jf Rome. This work was sponsored by The Club of Rome JaPznese C’ommitke and the Japan Techne-Economics Societv The first pdrt of this paper was published the preceding issue, TF&SC, Vol. 6, No. 3 (1974). 0 American

Ilsevirr

Publi ,hing Company,

Inc.. 1974

372

YOICHlKAYAANDYUTAKASUZUKl

One way to approach this problem of se!f-development would be to encourage the poorer nations to expand their agricultural and mining industries. In the past, actually, their exports have been principally mining and agricultural products; furthermore, most of the developing nations lie in the tropics where they are blessed with a warm, moist climate conducive to agriculture. ~‘his approach has two weak points. The first comes from a mistakenly exaggerated idea of the richness of the natural r+:‘sourcesin the developing nations. In Table 1 we have listed the figures for arable land, petroleum and uranium reserves for the va,rious regions of the world. .4s is obvious from the table, Asia, with over half of the world’s people, is singularly poor in land and resources. Africa and South America seem to have a considerable amount of potentially arable land, but much of this soil is of the ratsol type found in the tropics. This type of soil lacks the ability to store up organic nutrients, and so, with present agricultural technology, cannot be made nearly as productive as the soil in the temperate zones of the world.’ In the developeC regions such as North America, Oceania, and the Soviet Union, both rich soil and mineral resources exist in such abundance that they should be called )untries,.” The only reason that it is now possible for the developing countries “resou, to expo: nining and agricultural products is because their domestic consumption is extremel, small. Even if this export pattern of the developing countries were to be accepted now whiie their living standard is still low, it cannot be continued as they develop, because cJ201mestic consumption will begin to take up the entire product and make export more difficult. If development should be based on this export, the development itself ,will be stopped. The second point is that agricultural products will not maintain their income elasticity throughout the period of development. Since agricultural products are so basically necessary for human life, their income elasticity is now very large in the developing nations. As income increases, however, the demand for agricultural products, which are mostly raw foodstuff, does not expa,nd very much, and so the economic elasticity of these products decreases. If we compare, for example, the demand for agricultural products in the USA and in India, we find that th6 demand is only SIX times as great in the USA although the per capita GNP is 40 itimes rs great. Without a great increase in demand for agricultural,products, an ~onomy based on agriculture simply cannot grow very much. Munufactunhg hdustty: Its Need cmd Some Problems In view of what we pointed out above, one can see that the development of’ manuiacturicg industry is indispensable. Fundamentally speaking, the most desirable way 108pur-*le development would be to import raw mineral resources, manufacture products which can be produced anywhere, but which have a high ir‘come elasticity that does not fall off as the per capita income increases. Actually, in the past almost all of the present advanced nations have reached what Rostow [3] has called the economic “take-off point” by rapidly expanding their manufacturing industry. The countries of Asia, although they recognize agriculture as their basic industry, still have, not unexpectedly, a great desire to expand their manufacturing industry. There are a number of obstacles to the development of manufacturing, such as the lack -’ In the “burnt-field” type of farming, the field must be left fdlow for sis to twelve years after being used for only three or four. Otherwise, the organic nutrients in thz soil will be e.xhausted.

_-

40 ~--

Middle and Near East - -- _______~

income

370

1770

160

200

1500

I 700

430

3800

0-J. S. $)

0.8

0.46

0.28

0.93

0.33

0.32

1.1

Arwble land

7.6

2. I

26.6

2.1

6.8

0.2

3.8

1.46 0.34

1.5

8.9

i 5.6

Absolute existing am+unt of petroleum (100 tons of coal equivalent)

0.37

2.3

2.1

Poentially arable land fnot yet under cultivation)

2.0

0.2

0.6

0.4

3.2

or

Absolute existing amount of uranium IU,G,) (kg)

Energy resources per capita

Notes. (1) Based on Hopper’s assumption that only 30% can be used. (2) Quality of ore valved at $lO/ton -ofXl-P;n l977 <‘) J&cd pri::;ipdly m &ta from Zre WorkibiiudProhkm, G’.S. Cow. Printing Office. i97G.

20

Australia and New Zealand

340

240

Soviet Union

Africa

460

Elurope

1810

310

Cenrraland South America

Asia

220

North America

--___

Region

(1969) Population (millions)

(i966) Rr Capita

Natural Resources of the World >;J Regions -. _. -Land area per capita (hectares)

TABLE 1

374

YOtCiHI KAYA

ANDYUTAKASUZUKI

of domestic capital and the lack of technical education. In the long run, however, an even greater ob,stacie will be the difficulty of securing adequate markets for the manufactured products. Up till now, ail the world markets for manufactured goods have been dominated by the aavanced nations, a situation to be expected to continue as long as manufacturing is their central economic activity. If, under these conditions, the developing nalions should advance a little in manufaciuring, they must invade the markets of the advanced nations and enter into head-on competition with them. It is hard to imagine, however, that the developing nations would be able to expand their industry to such an extent that they wollld be able: to compete successfully with the advanced nation!: No matter how one looks at it, from the standpoint of supply and demand, unless the advanced nations change the pattern of their being the exporters of manufactured goods, the developing nations will always remain importers of such goods and never be able to develop their own manufacturing industry to any great extent. The Need for a Plan to Redistrih tc the W&d’s Manujbcturing Industries Since, over the long run, the greatpzt obstacle to the economic growth of the developing nations will be their inability to secure markets for their products, we must attack this problem. One way would be to accelerate the economic growth rate of the advanced nations in order to increase the demand for the mining and agricultural products exported from the developing nations. This policy would be only a stopgap, though, for the developing nations are not so rich in natural resources, nor would the demand for imported agricultural products increase very much in the advanced nations. An even greater drawback to such a policy would be that the ear!h’s resources could be used up at an ever faster pace, and more pollution would be thrown off by ever-accelerating manufacturing in the advanced nations. From a global point of view, the most desirable policy is one that will help the developing nations to grow while restraining the global economic growth rate. How, then, can we accelerate the development of the underdeveloped, while not accelerating the economic growth rate of the devalLpf:d nations? More abundant economic aiC might be given, but this will hardly golve the problem PSlong as the developing nations cannot obtain markets within the advanced nations. By giving economic aid, the advanced nations would have to reduce their demands for investment 2oods and thereby reduce the markets available to developing countries. Also since manufacturing requires an able labor force and technical abiiities, large inputs of economic aid cannot produce results Lrla country where these requisites are not yet found. The most effective solution would be for the advanced nations to modify their own industrial structure so as to create markets throughout the world for the products of the developing ncltions. The advanced nations could give up the manufacture of some products which could then be made in the developing nations. In the advanced nations, knowledge-intensive and agricultural 5ndustries could be further developed to make up for the industries given up. This iype 0,” policy would certainly help the developing nations effectively. To foresee in detail, however, just exactly how the industrial structure of each nation should be changed to achieve the above effect, is not a very easy problem. In each country for exampls, the natural resources, the land, labor force, amount of capital, and market demand are all influenced by physical and econom:c factors; furthermore, these factors are constantly changing. Also, looked at from a global viewpoint, if the supply and Idemand for a certaiii type oT goods gets out of balance, political and economic

GLOBALCONSTRAINTS

AND A

NEWVISEON FORDEVELOPMENT-II

375

instability may result. Consequently, production must be carried out in the various countries of the world iti such a way that year by year the balance of supply and demand is maintained. We have taken up this problem using a regionally-sectored dynamic world model. By optimizing the behavior of the model a desirable pattern for the redistribution of the world’s industries was worked out and v;irious aspects of the solution were investigated. The plan developed here-we will call it the “World Industry Redistribution Plan” -is a kind of “international division of labor.” Ever since Ricardo first proposed it, such an international division of labor has always been viewed primarily as a means of increasing productivity. Our plan, however, is quite different in that it aims at the development and increased productivity of the developing nations.’ Of course, the plan is an ideal which cannot be realized without overcoming many international political obstacles. This direction is. nevertheless, the most effective way to help the developing nations, and we hope that the advanced cations will realize this and begin to make efforts to re-structure their industy to this purpose. The model WChave built is not yet perfected; also the data that we have used so far must be checked over and improved. The scope of the problem the model is intended to solve is too large and complex for just one group of private researchers. Therefore we hope that the work will be made a project of some international organization such as the UN or the OECD. We believe that the methodology and the results we present here could be made the basis for effectively pursuing this as a large project. 2.2.THEMETHODUSEDTOTESTOURPLAN

To study this plan in greater detail and from many aspects we have built a world industry model which has individual suQ*stems for each geographical region and for each sector of industry in each region. Here we will give only a simple outline of our method. The details of the mathematical formulation and method can be found in an appendix. The Parts of the Model The entire world was subdivided into nine regions, and in each region six industrial sectors were considered. The regions are: North America, Oceania (Australia and New Zealand), Western Europe, Japan, the Soviet Union and Eastern Europe, Central and South America, China. and Asia (excluding Japan and China). The six industrial sectors are: agriculture, mining, light manufacturing, heavy manufacturing, assembly (knowledge-intensive) manufacturing, and service industry. For eac9 region there are models for industri:d production, potential demand, and population. The hchavior of these interacting models is controlled by investment according to region and industrial scclor. These investment inputs are devised to minimize a cost function which will bc explained hclow. The basic structure of the model is shown in Fig. 1

77~ Factors Included in the Cost Function There are four factors each of which influences the value of the cost function.

2 Although other cLonomic mod& have al\o used mathematical propramming, as far as the authors know, theirs is the first to attempt a dynamic slmula!ion of a distribution by regions and economic se.-tars.

YOICHA KAYA AND YUTAKA SUZUKI

376

portion which

flnol

!

t-

i

of production final demand meets

demond to meet

I

XI

,

x:

r-

D

/

: & itlmedelay of j one period

Per coplra Income

I investment by 1 regions ond 1 lndustrlol , sectlon I

L

I

I -_

r

_imtting

1

I’

)

I I

porometers lund

areo

reserves oblllty to

resources of miner01 of environment obsorb pollutants

_-_I

I

i

the

arable

-----

-!-

-7

I-th region

! ; I

the

yet;;;

t

FIG. 1. Basic structure of the model.

(1) The production-gap factor: This factor is ?esigled to become smaller for a particular region as the total per capita production increases. Therefore, using investment to decrease the cost function with respect to this factor tends especially to raise the industrial output of the regions with a low gross regional product (GRP). (2) 7%e supply-demand imbalance factor: This factor is a sum of the squares of the global differcfnces in supply and demand within each industrial sector. When this factor is small it means that, supposing an ideal form of international trade, the supply and demand in exh industrial sector is quite well balanced. By inserting this factor, we assure that the redistribution of indust?/ obtained by our model wil! not fail to meet this most basic requirement for a realistic solution. These are the two principal factors in the cost function, but we also add other factors as required for studying particular aspects of the problem. The two most important of these are: (3) 7%e ecouonk self-sufficiency factor: The degree in which a region supplies its own demands in a particular industrial sector is the degree of its economic selfsufficiency By incorporating such a factor into the co$t function, we are, in effect, recognizing that for the security of a region and also to reduce fIei$ht transportation fiom region to region, each region requires a certain degree of such self-sufficiency.

GLOBAL CONSTRAINTS

AND A NEW VISION FOR DEVELOPMENT-II

377

(4) E’nerfl pollution factor: This is a factor related to the energy consumption per unit of surface area in each region. We use this factor m the cost function when we want to secure that the model will produce a solution in which the regional environments will not be unequally burdened by pollutants from energy consumption. Capital Coefficients for Each Region and kiustrial Sector The industry model for each region takes into account the relations between the six industrial sectors. For simplicitlr;. we have adopted the simple relation defined by: (investment) (increment in value added) = -(II arginal capi.tal coefficient)

(2.1)

The marginal capital coefficients for each sector were based on data for past years. The coefficients used for the three manufacturing sectors are shown in Fig. 2. In a particular region, when the income level rises, then, in general, the level of knowledge and manufacturing ability also rise. This fact is reflected in the way the coefficients in Fig. 2 slope down to the right. A more complete explanation of our way of setting the marginal capital coefficients can be found in an appendix. At this point, it is also peritinent to note that th.:: capital coefficient for light manufacturing is the smallest of the three sectors, and the highest IS for assembly manufacturing. For the a&ultural sector we made separate models for the agricultural use of land ;rhPady under cultivation and for the land just being brought under cultivation. In each region, these agricultural models take into account the differences in arab!e land area as well as the fertility of the soil. In the mining sector we considered petroleum as the representative product upon which we based the structure of the rnodel. Compared to other sectors, the mining sector exerts so little economic influence ?hat it has very little effect on the overall behavior of the model.

Knowledge-lntenslve

;I

5

tieovy

monufocturtng

Light

monufocturlng

Monufcccurlng

~ i.._.

I00

I

200

Producttv?

400

1600

BCO

tiu1pu1

FIG. 2. Assumed ma&d

r$

U 5 per

3200

$

coplto)

coefficients for

the three

ITUWIUfaCtUring

sectors.

YOICHAKAYAANDYUTAKASl~ZUKI

378

In each regiaa we divided the potentir;l demand into that for consumption goorls and that for investment capital. The demand for consumption goods was set up as a fr nction of gross regional product (GM’); data for the past years in Japan was used to determine the exact functionti relation. The demand for investment capital, taken mercl) as an over& investment ratio. was also set lip 3s a function of per capita GRP. The population model we are using for the time being does not take into acco..mt the age structure of the population. but works according to the simple formula: (increment in populati.orl)= [(clvde birth rate)
Annual investment by region nnd industrial ssctor constitute the cor.trol variables in our model. One cxceplio.1 to thii: is the service sector investment which we connected to the per c:api,LatiW, since there is very little export of service to meet aemands in other regions. in the mining sector also, we made a special provision so that the investment in this sector swould always be enough to supply the other sectors with th(l necessary raw materials for production. On all of the investment variables we imposer; the following restraints: (amount invested) 2 0,

(2.3)

(overall amount invested)=(overall investment rate) X (GRP)

(2.4)

7&e Future Period Simulated Since new develo?mcnts in technology ar,d th”Ir sffccts arc den’ (Difficult to predict. we do not think that our model can give meaningful results for ;uty more th:.~nforty years into the future. and 50 we rcstrictcll our studies to the period frt)m 1’170 ta 20 IO A.D.

if one seeks to minimize the cost function a? each one-year interval, the problem can be set up to apply quadratic programming to the sear41 for lhc best investment pattern. In this report. all of the results presented were obtained in this way. Another approach to the problem would be to look for an irvestmcnt :Jattern which minimizes. the cumulative value. that is the integral, of the cost f:mction over a period 01 years. Then. instead of optimization at one-yc;\r intervals. the problem bccomcs one ot optimizing the dynamic performance of a large. nonlitlcar syslcm. For this, fhc ncccssary computation is immense. We did, however. try this approach tbl ;I 1’~ cases: and ti~nnd. foitunate!y, that the results were not much Jit’fcrcnl tiom thosr olbtained by optimiL;ltion at one-year intervals. WC think that this dose a::reeinent is probably due to the scarcity of‘ time-delay elements in our motlel, but WC i::tcnd to do further work un this dynamic optimization. 2.3. USINGTHE MODELTO tSVALUATl~Ttll~lNUUSTRY RE-DISTRIHUTION PLAN

Using the tcchniqucs outlined in the previous section. WC made simulation runs to examine the effect of the industry redistribution plan. It is, of i:ourse. very difficult, for a

GLOBALCONSTRAINTS

AND A

NEWVISIONI-OR DEVELOFMENT-II

379

40.year period in the future, to determine the correct values of the various parameters which affect the behavior of the model. Consequently, it would be rash to give too much credence to the absolute nurnericJ results the model produces. The model can be used, however. very effectively to compare the results of different approaches to the problem. We have compared, for example, the results obtained when increased productivity is made the sole criterion for the optimum redistribution of industry. Such a run with the model we have called a “productivity-oriented run”; in the model, it represents a policy similar to that proposed in Ricardo’s “international division of labor” theory. In terms of the cost function explained in the PI~vious section. this run is carried out by making only one change: the production-gap factor is omitted from the cost function. In contrast to this result. the plan we propose is snnulated by the model in what we call the “deveiopmentoriented run.” Indusrty Retiistrihtctio~~: The Resulting Pattern and its Ej&-t

First let us see what effect redistribution of industry has on the developing regions. Of course, the relative numerical importance we give to the production-gap factor in the cost function will affect the result, but in Fig. 3 we show one typical result. The figure displays, for comparison, the result obtained by the production-oriented run (solid line) together with that obtained for the development-oriented run (dotted line). As one can see from Fig. 3, the results of the two policies are about the same until after 1980 when a clear difference appears. The deveioprrent-oriented run clearly decelerates the growth rate of the adva,lced regions and accelerates that of the developing _

Production-

- - - - -

Development-

orlented

run

oriented

run

USA Sovret Union Western Europe Ocean10

Central

iL___l___1910 1980 I990

ond

South

Amerlco

_----L--_-‘L_-___

FIG. 3. Comparison capita GRP.

of production-oriented

2cw

2010

jeer

and development-oriented

simulation

results for per

YOICHAKAYAANDYUTAKASUZUKY

380 regions. Thus, even though

the growth rate of the developing regions is much higher than in the productivity-oriented run, still the global economic growth rate is a little less than ir; the productivity-oriented run. This conforms to the requirement of Sec. 2.1: “From a g!oLal point of view, the most desirable policy is one that will help the developing nations to grow while restraining the global economic growth rate.” For these: same runs we can examine the resulting industry re-distribution patterns. Figure 4(l) to (9) shows these patterns region by region. The graphs show only the relative contribution of each of the six sectors to the gross regional product (GRP). One should be careful to note that, even though the relative contribution of a sector drops in the future. this does not by any means indicate that the gross product of that sector has derreased; it only indicates a drop relative to other sectors in that same region. If we look at -North America in Fig. 4( 1) we see that the development-oriented run differs little from the productivity-oriented run in the beginning, but later the share of the GRP due to

a)

Oevelooment

-0rlented

b)

run

mrnfng

-

run

mlnlng Industry

too

Industry

service

servxe

-----l

Producllvlty-Ortented

Industry

Industry

know le4ge - intri

----heavy

mofg -heavy

monufacturlng

monufoclurlng tight mofp

1970

1990

year

zoo0

199C’

,pgrlculture

i’

-_

0

1970

1980

I990

2000

yeor

FIG.4(I).Structure of the GRY fc t North America.

a)

Development-Or‘le?ted

---

run

b 1

loo

mlnlng Industry

Productrvrty-Ortented

-

run

mlnlng Industry

t service

Industry servxe

industry

1990

1990

a

._l OL----

ogrlculture

b----L

1970

19BO

1990

2000

,&G

1970

FIG. 4(2). ,Stxucture oftheGRP for Oceania.

2000

yeoi

GLOBAL

CONSTRUNTS

AN@ A NEW VISION

Development-

0)

OrIented

FOR DEVELOPMENT-II

run

b 1

381

Productlvlty-OrIented

servtce

-_..

.nduslry

ogtlcul?YJre

\ 3

. .._-.-A._.

__.

19.90

19.70

FIG. 4(3). 01

_..-

year

2000

1990

.

F

1.

A...__.._

run

~ogr1culture

.._

2oor,

yeor

Structure of the GRP for Western Europe.

Development-Oriented

r.rn

bl

_ Ortented

PrOduCfiv\ty

mtnlnq ndujtry

m,ns”Q

100;

-- Industry

‘-

rur

A

8

:

-

i

service

a

Industry

service

knorled?e-

lntensve mofg heavy mofg * 3grlculture

~I I.

1 1970

1980

1990

---i

light

year

2000

monufocturlng oorlculture

0

FIG. 4(4). II)

Industry

L_ 1970

1980

1990

2000

yeof

Structure of the GRP for Japan.

Development-Oriented

b)

run

PrOduc+!vltf-Orlenled

run

m,nlng I00

Industry

3 service

.A

industry

B w .I-

OS0 : ; L

8

0

__L

_____I

1970

980

c-1990

FIG. 4(S).

__I-

_

2000

_~

---

_.

yeor

---i-___ 1970 1960 Lo__

_J------L--.

'990

Structure of the GRP for U!;SR and Eastern Europe.

2000

^~

--

"=Ol

YOICHA KAYA AND YUTAKA

382 of

Development-Oriented

bl

run

100,

__-_.___~-.__-____ -2_~

knowledgemtenslve

monufocturlng

ogrlcullurc

u’ogrlculture ,

I

0 LTQ.70--

\ 990

,,~.

-2000

FIG. 4(6). 0)

Industry

E

hnowledge-

tnduslry

-

seivl’;e

Industry

lfght

run

mtnlng

m~nlng

-=industry

service

Productlvlty-OrIented

?UZUKI

_.I__~>

-._ -.___

~_

1970

yeor

- ~~

___

1990

1980

ye01

2000

Sltructure of the GRP for Central and South America.

Development-C

lented

b)

run

Ploducllvlty

-0rlented

run

mlnlng mlnlng Industry

service

=

2

mdustrv

service

Indust!

y

Industry

knowledqe-

light

monufocturlng

ogrlcu

Ilure

cgrlculture

i_-.--.__ 197c

0

1990

IS.30

I

yeor

2000

i

t

I’?70

l9RO

,990

yeor

2000

FIG. 4(T). Structure of the GRP for Asia. 0)

_

1

9

/

ICI0

t)eveloprnent-OrIented

b)

run

---

,

loo mtnlq

~nrlustry

Productivity-Ortented

c

--Y--c--cI

rvfnlng

Industry

------_ service

,

Industry

G

e

knowledge‘.intenslve mofg heavy light

mofq

ogrlrulture

L _>I____~

-L_I__-~.

1980

IQSO

2000

PIG. 4(8).

___.___~~ yeor

Industry

knowledge%

S = a”

monufoclurlng

S~‘rvlCe

lntenslve mofg

50, 1

heavy

I / I

llqhl

ogrlcu

1970

run

-L-_-L 1970

lture

..~___

...L-_L__ 19RO

Structure of the GRP ior Africa.

1990

mofq mofq

2000

year

GLOBAL(‘CINSTRAINTSAND A NEWVISION FOR DEVELOPMENT-11 o I

Development-

Orlenred

run

b)

383

Producttwty-OrIented

run

mlnlnq

industry

servtce

llqht

lnduslry

1970 . ~-~- ~-1980 _--

service

mint ng InduStTy

Industry

muwfoctwng

agrfculture

“-l-

ogrlculture 0.

-

1990 *

2000

-

FIG. 4(9).

I

ye,,.

0

1470 Lo._

1980 ._ i

1935 I

2ObO

yeor

Structure of the GRP for China.

agriculture incrcsscs while that due to light manufacturing decreases. By contrast, in Asia, as can be seen in Fig. 4(7), agriculture increases its importance very rapidly in the beginning of the development-oriented run due to the initially high income elasticity from rapidly increasing demand and to the high capital efficiency attained by laborintensive agriculture. But as developme,lt advances, the portion of the GRP due to agriculture drops considerably and that due to light manufacturing grows very large. The reason for these changes in the agricultural sectors of North America and Asia is not hard to find. North America has, on a Fer capita basis, very abundant farmland having a high productivity. In Asia, on the other hand, agriculture develops initially, but then the capital efficiency drops because of tl e shortage of land, and so the economically more promising light-manufacturing sector expands very much. In North America also, the capital efficiency of agriculture is lesi than that of light manufacturing, and so the expansion of agriculture there accompanieIj by a decreasing share of the GRP from light manufacturing means that the economic gr#.)wth rate is decreased. One could raise against this result the Ilbjection that progress in agricultural technology will undoubtedly raise the producti,ity and the capital efficiency of the scarce farmland of Asia, With such progress, the o:)jectiun would run, would not the structure of industry in Asia be quite different from that obtaint:d here? To investigate this point with the mode we made a run in which we now incorporated a 2% annual increase in agricultural produ$:tivity. This figure of 2% is, we believe, on the optimistic side, since the FAO goal of a global 3% increase in total agricultural production, which also includes the yield from land newly brought under cultivation, has never yet been att;iined. The results of this run ale. t’or comparison, plotted together with those of the previous run in Fig, S. The results in the agricultural sector of North America and of Asia show the tenticncy one might expect. The relative contributji>n of a:;riculrure in Asia is increased; that in North America is decreased. The model, hl)wever, does tell us that this assumed improvement in agricultural productivity c\oes not :.ffect the optimum pattern for the distribution of industry very much. Not only ilk this experiment, but in a number of other cases we examined, the model :Ilways gives I result sin&u to the above: jn North America and Oceania. agriculture

YOICHAKAYAANDYU'I'AKASUZUKI

394

“I

ICO

Knowledge

- ~nlens~vc

monufortur~ng Heavy

monufocturlng

Light

manufacturlnp

Agriculture 1970

IOU0

I390

2oca

yeor

FIG. 5. Influence of improved agricultural technology on the structure of the GRP (left) in the USA; (right) in Asia.

GR’Vcopt

1o

iu s $1

Japan North Amerlco Soviet bloc oceonlo Western Europe

$104 Central South

---

and Amer~co

Afrlco

$10'

810"

-----

ASIO

--

Lose

I

case

Influence of the degree of self-suffkicncy

HG.

6.

2010

A.D.).

Chino

2

case

on the

3

GRP of each region (CRT given is that for

GLOBAL CONSTRAINTS AND A NEW VlSlON FOR DEVELGPMENT-II

385

1:xpands its weight in t+e GW; in Asia. this sector decreases in impo:tance :nanufacturing expands. Bud Ejfects

of a Policy

while light

of Regional Self-sufficiency

Obviously, if a region can more or less supply its own demands, the nations there mjoy greater security. But if economic self-sufficiency should be overly emphasized, then only the regions with adequate natural resources would actually attain this goal. Such a policy is clearly opposed to the principle of our plan, namely that of seeking arr international division of industry to promote the economic productivity of the developing nations. Using our model, w investigated the bad effect that would result from a certay;l degree of emphasis on zgional self-sufficiency. Since, in the mining sector, there are some regions for which sell-sufficiency is impossible, we did not inc!ude this sector in the experiment. Nor did we consider the service sector, since service is of its very nature supplied in a self-sufficient manner in each region. In Fig. 6 we show the results obtained for the year 2010 when three different degrees of self-sufficiency are incorporated into 11.‘: model. Case 1 gives no emphasis to self-sufficiency; Case 2 supposes a degree of se!i-sufficiency about equal to -that operative in the world economy today; Case 3 emphasizes self-suffic+ncy the most, considerably more than irl the actual world economy today. To evaluate the results we use an index which indicates rhe degree in which selfT.\BLE 2

Aver3gcIndex of Dependence for Each

Region in 2010 A.D. (for Development-Directed Form of Model) --. -_ Case 3 Case 2 Case 1 (strong emphasis on (some emphasis on (no emphasis on self-sufficiency) self-sufficiency) Value for 1970 self-sufficiency)

North Americn

0.194

0.482

0.036

0.0?7

Oceania

0.082

1.114

0.628

0.’ 85

Western Europe

0.185

0.283

0.497

0.1?4

Japan

0.182

0.579

0.0511

0.01;7

Soviet 1~lOC

n.120

0.247

0.073

0.0,;7

0 233

0.696

0.309

0.151

Asia

11HO4

0.84 3

0.688

0.490

Africa

0 277

o.ti12

0.378

0.130

China

0.400

0.806

0.803

0. j52

0.276

0.629 ___I

Central and Soutl: America

Average

for all regions -c__--

0.385 .I_

--

0.190 _--

YOICHA KAYA AND YUTA’:A

386

SUZUKI

sufficiency has not been attained. This index could be called the “index of dependence.“3 The “average index 0; dependence” of each region for each of the three cases is given in Table 2. By comparing Cases 1 and 2 in Fig. 6 we notil:e that the degree of emphasis supposed in Case 2 leads to some changes within the advanced nations, but does not adversely affect the development of the poorer regions to any significant degree. When selfsufficiency is more strongly emphasized, however, as in Case 3, then the GRP of North America and Oceania with their abundant farmland and natural resources increases, while the GRP drops for the heavily populated regions such as Japan, Western Europe, and the developing n$\tions in general. At present, we see in the policies of the USA and Australia a kind of Monroeism. We should strongly demand that these nations recognize the way this works against develop ment and change their policies. Foreign Aid and the Industry Redistribution Pla;v Our industq- re-distribution plan is actually iI form of indirect foreign aid whereby the advanced countries cooperate to facilitate the development of the less wealthy nations. Now let us take a look at the effect that the traditional form of economic aid would have within the framework OTthe world industry redistribution plan. To get some concrete results we used our model to compute the effects that would appear after 40 years ucder a policy of economic aid to the developing nations. For comparison we again made the “production-oriented” and the “development-oriented” runs, but this time incorporating a policy of aid from the advanced to the developing nations. Many forms of aid are possible, but we assumed one of its ideal forms: 1% of the total GNP of the advanced nations i.sgiven annually as untied non-project aid in the form of investment capital. The overall gross world product for the various cases is shown in Table 3. TABLE 3 The Effect of Foreign Aid on the Gross World Product (Kestdts for 2010 A.D.; Unit is IO” U.S. $) ~-.--_ -_--Without foreign aid ~--~--____

.-.

With foreign aid Difference ~.___ ..-. _..____-. ~_

Production-orknted form

44.66

45.49

-+0.X.7

Devel’.)pment-oriented form

42.48

44.9 I

+2.4I

3The index of ileperdence for each industrial sector in a given region is defined as:

I

total demand for product -- 1. 1 total production of product The “average index of dependence” i:)r a region is the avcrape of tile indices of dependence for four industry sectors of that region, namely, agriculture, and light, heavy, and assembly manufacturing. Consquerltly, a sm,dl value of this index indicates that the region has a high degree of self-sufficiency. (index of dependence)

=

GLOHALCONiTRAlNTSANDA

NEWVISIOUFORDEVELOPMENT-II

387

Areas for Further Investigation The intfustfy re-distribution plan has. in the limited investigation we have made so far, shown irseif to be a very effective way to help the developing nations witfin the framework of realistic world nstraints. But our investigations of this plan are far from being completed, and so we would like here, at the conclusion, to list the points which still require investigation. (1) hindin@ t3 be expected from sensitivity analysis of the model. As we mentioned in the beginning of this section, the parameters we have used in the model might change their values witii time, and 50 each of these values must be checked carefully-nor onl~r their numerical values, but aiso the sensitivity of the result to variations in these values. The results presented here have all been checked to assure that they are not unreasonable or unrealistic, we do feel, however, that further useful findings will come from pcrsuing this analysis more thoroughly. There are som e effects in our model which arz ver}. sensitive to aarametcr chang,es, and so no zoncludons can be based on them. In buth the production-oriented and the development-oriented runs, the policy of economic aid slows down the growth rat? of tha advanced regions while accelerating that. of the developing regions. In both cases, thoug!:, the policy of economic aid produces a larger gross world product forty years hence than does the no-aid policy. The increase of gross world product caused by aid in the develtipment-oriented run is three times larger than that crused by aid in the production-oriented run.4 This difference indicates that the economic aid is more effective within the framework of the development-orientec’ world economy. The reason is that the economic aid received is invested in industries thal contribute more to development because of ‘heir higher capital efficiency. Also, the redistribution of industry assures that the deve’oping nations have adequate markets for their developing production. We found, fcr example, that the pattern of industry distribution in the advanced nations-except for the increase of the apricultrrre sector in North America and Oceaniachanged considerably when the service-industr; portion of the GRP was changed only slightly. For this rzason, we realize that, ,t the present stage of our work it is not possible to gain any detailed knowledge about the desirable industry pattern in the advanced nations. We intenLl to pursue this line of investigation much more thoroughly. (2) Imprcwiu~ the data. There are m: ny im$rfections in the data we have made use of, and there are many points about w’dch we-must still gather and evaluate data. For example, we espczially would like to gat better dat3 about the productivity of various soils, about the regional differences in the consusnption pattern expressed as a function of income, and about the regional input-output matrix and its variation with time and as a function of income. (3) Injlucrw o.f far’tors omitted frolrl the iwustn+ model. Our industry model is an extremely simple one which omits a number of factors. Among the more important of these are the labor supply and the changes in price. In the future, we must see how these factors, when introduced, will affect the -esults. (4) The ejfect of freight trarlsportation. By allowing an economic policy of selfsufficiency to play some role in the behavior of the model we implicitly ack.nowledged that transportalion from one region to a:lc$her should be held down. However, the factor of transportation of manufactured goocis from one region to another does nut appear 4 Later irn,xovemellts in the data we US,: in our model may change these tigurec. and so they should not be given any absolute significance.

388

YQICHI KAYA AND YUTAKA SUZUKI

explicitly in the model, and as such is not considered. Actually, though, shipping manufactured goods back and forth from one region to another would use up energy resources and also cause prices to increase. This would work against the type of redistribution of industry we propose, and so we intend to investigate this point more thoroughly. (5) Effict ofr egionul conditions. The industry re-distribution plan we presented has taken up the problem in a dobal way and from the physical and material aspects of each region. To further refine this plan to fit the real world, we must also look deeply into the political, economic, and cultural conditions of each region. To this end we will very likely have to adopt some qualitative and static analytic methods to be used in conjunction with our dynamic optimization method. We believe that only through a synthetic combination of these two methods of analysis can we show the way each region should develop in the future. Note. Space limitations unfortunately dic’>!teomission of the Appendix, which describes the details of the world model used in this project. Interested readers should contact the authors directly. References 1. Report of Commission on International Development, Partners in Development, 1969. 2. A Report of the President’s Science Advisory Committee, The World Food Problem, The White ‘House, 1967. 3. W. W. Rostow, The Stages o.f Economic Growth, Cambridge Univ. Press, 1960. 4. J. Kuaynski, Die Geschichte der Liege der Arbeiter unter dem Kapitalismus, Akademic-Verlag, 1967. Received February I, I974