Jump to ContentJump to Main Navigation
Hunger and Public Action$

Jean Drèze and Amartya Sen

Print publication date: 1991

Print ISBN-13: 9780198283652

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/0198283652.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 08 May 2021

Economic Growth and Public Support

Economic Growth and Public Support

Chapter:
(p.179) 10 Economic Growth and Public Support
Source:
Hunger and Public Action
Author(s):

Jean Drèze (Contributor Webpage)

Amartya Sen (Contributor Webpage)

Publisher:
Oxford University Press
DOI:10.1093/0198283652.003.0010

Abstract and Keywords

Possible dissonance between income and achievements of quality of life is noted. Then two alternatives for removing precarious living conditions are explored: growth‐mediated security and support‐led security, with illustrations from the experiences of various countries. The connection between these alternatives is studied. The strategy of growth‐mediated security is distinguished from that of aimless opulence with examples from the experiences of Kuwait and South Korea. The last part is dedicated to the effectiveness of support‐led security strategies.

Keywords:   achievements, equivalent growth, growth‐mediated security, income, Kuwait, persistent undernutrition, public action, South Korea, support‐led security, unaimed opulence

10.1 Incomes and Achievements

The mistake involved in analysing hunger just in terms of food output and availability has been discussed extensively in earlier chapters of this book. The main issue concerns the power of vulnerable groups to command food and other essentials, rather than just the physical availability of commodities. While focusing on entitlements rather than on what is available has many complex implications (we have tried to analyse some of them and to draw lessons for policy), the basic contrast between command and availability is fairly straightforward.

Sometimes the position that hunger is essentially due to a command failure not necessarily caused by an output failure is summarized in the form of the simple slogan ‘hunger is caused by a lack of income, not of food supply’. Extending this line of straightforward analysis, it is often argued that the real problem lies in the shortage of purchasing power, rather than of anything else.

That conclusion has some obvious sense, but as a causal theory it is rather misleading. This is so not merely because income is itself a derived variable which depends on ownership and exchange and therefore provides a rather poor focal point for the analysis of entitlement failures. But no less importantly, it is also the case that many essential commodities are not bought and sold in the market in the usual way, and conventional estimates of real income may not give us a good idea of the command over a number of inputs which, as we have seen, can play a crucial role in the removal of hunger, such as educational services, health care, clean water, or protection from infectious epidemics. The importance of these non‐food commodities and facilities for nutrition and health was discussed earlier (in Chapters 3 and 9). Since their use may have a major impact on the nutritional status and health of a population, and since they are often provided directly through public delivery, the lacuna involved in concentrating on purchasing power is quite limiting. Income is a rather dubious indicator of the opportunity of being well nourished and having nutrition‐related capabilities.

This is one reason, among others, why the association between Gross National Product per capita, on the one hand, and health, nutrition, morbidity, and mortality, on the other, is far from simple. In Table 10.1, we present some relevant figures for five selected countries to illustrate the point. South Africa, with nearly six times the GNP per head of China and Sri Lanka, has a (p.180) life expectancy of only 55 years, compared with 69 and 70 years respectively for the two poorer countries.1 Similarly Brazil, with many times the income per head of China and Sri Lanka, has nevertheless lower life expectancy than the latter. Oman, with about 20 times the GNP per head of China or Sri Lanka, offers a life expectancy of only 54 years at birth.

Table 10.1 Average Opulence and Survival Achievement: Selected Intercountry Comparisons, 1985

Country

GNP per head (dollars)

Infant mortality rate (per 1,000)

Expectation of life at birth (years)

China

310

35

69

Sri Lanka

380

36

70

Brazil

1,640

67

65

South Africa

2,010

78

55

Oman

6,730

109

54

Source: World Development Report 1987, Tables 1 and 29.

The contrast of infant mortality rates brings out the same dissonance between GNP per head and the capability to survive premature death. Brazil, Oman, and South Africa, despite their much greater opulence, have enormously higher infant mortality rates than China and Sri Lanka.

There are, in fact, two distinct—and in principle separable—causes underlying the dissonance between GNP and achievements of quality of life. First, the GNP gives a measure of the aggregate opulence of the economy, and the translation of this into the pattern of individual prosperity would depend also on the distribution of income over the population.2 Second, as we have seen, the capabilities enjoyed by people depend on many factors other than the command over commodities which can be purchased in the market. Among such factors, public provisions made by the state for health, education, sanitation, etc., are especially important.

While we must recognize this dissonance, there is no reason to dismiss GNP altogether.3 There are, in fact, good grounds for expecting a positive general (p.181) association between GNP and nutrition‐related capabilities. This is partly because the increased incomes associated with greater general affluence do indeed offer the opportunity to buy a number of commodities that are inter alia crucially important for nutrition‐related capabilities, the most notable of which is, of course, food itself. But, in addition, a higher GNP per capita enlarges the material base for public support in areas such as health care and education, and generally facilitates the provision of social security to the more vulnerable sections of the community.

The governments of some of the oil‐rich countries, for example, have been able to use their unusual and relatively recent opulence to make widespread public provisions for their citizens, and this is one of the reasons why life expectancy at birth and similar indicators have, in recent decades, moved to comparatively high figures in countries such as Kuwait and United Arab Emirates (e.g. life expectancies above 70 years). The examples of these countries are admittedly rather exceptional, but the general principle that an expanded basis for public support can be one of the fruits of economic growth applies to other countries too. There are also many cases where average opulence has reached a high level, but public support has been comprehensively neglected, with correspondingly low levels of nutritional and related achievements.

At the risk of oversimplifying the problem, it can be argued that a high level of GNP per head provides an opportunity for improving nutrition and other basic capabilities, but that opportunity may or may not be seized. In the process of transforming this opportunity into a tangible achievement, public support in various forms (and influencing both the distribution of income and the relationship between income and basic capabilities) often plays a crucial role.

Improvements in the quality of life are sometimes seen simply as the result of increases in overall affluence per se, when in fact the expansion of public support may have been the crucial intermediator. Common perceptions of the historical experience of Western countries in enhancing life expectancy often involve this misleading belief in the power of simple opulence per se (e.g. high GNP per head). In fact, the idea that the rich countries have achieved high levels of basic capabilities simply because they are rich is, to say the least, an oversimplification. A good illustration of this point is provided by the history of longevity expansion in Britain during this century.

Table 10.2 presents the increase in life expectancy at birth in England and Wales in each of the first six decades of this century (starting with a life expectancy figure no higher than that of most developing countries today). Note that while the increase in life expectancy has been between one to four years in each decade, there were two decades in which the increase was remarkably greater (around seven years approximately). These were the decades of the two world wars, with dramatic increases in many forms of public support including public employment, food rationing and health care (p.182) provisions.4 The decade of the 1940s, which recorded the highest increase in British life expectancy during the century, witnessed an enormous expansion of public employment, extensive and equitable food rationing, and the birth of the National Health Service (introduced just after the war).

Table 10.2 Longevity Expansion in England and Wales

Decade

Increase in life expectancy per decade (years)

Male

Female

1901–11

4.1

4.0

1911–21

6.6

6.5

1921–31

2.3

2.4

1931–40

1.2

1.5

1940–51

6.5

7.0

1951–60

2.4

3.2

Source: Based on data presented in Preston, Keyfitz, and Schoen (1972: 240–71). See also Winter (1986) and Sen (1987e).

The nature of these experiences illustrates both the importance of the distinction between commodity availability and functioning achievement (discussed earlier), and the role that social intervention and public support played in the expansion of a very basic capability in the history of the first industrial nation.5

(p.183) 10.2 Alternative Strategies: Growth‐Mediated Security and Support‐Led Security

Given the distinct, though interconnected, roles played by overall opulence and public activism in enhancing capabilities, it is possible in principle to distinguish two contrasting approaches to the removal of precarious living conditions. One approach is to promote economic growth and take the best possible advantage of the potentialities released by greater general affluence, including not only an expansion of private incomes but also an improved basis for public support. This may be called the strategy of ‘growth‐mediated security’. Another alternative is to resort directly to wide‐ranging public support in domains such as employment provision, income redistribution, health care, education, and social assistance in order to remove destitution without waiting for a transformation in the level of general affluence. Here success may have to be based on a discriminating use of national resources, the efficiency of public services, a redistributive bias in their delivery. This may be called the strategy of ‘support‐led security’.

The possibility of success through either approach is credible enough in principle. The real question is whether in practice they can be utilized in the way we might expect in theory. There have been, in fact, serious detractors questioning the viability of each of these two strategies. Some have questioned the soundness of a strategy which gives precedence to public support over growth on grounds of the allegedly extravagant nature of generous public provisions for a poor economy. They have argued that deflecting resources to social services from investment reduces economic growth and adversely affects future opportunities.6 State provisioning as such has also been regarded, at times, with considerable suspicion, and it has in fact become a target of relentless criticism in the contemporary intellectual atmosphere of great faith in ‘the market’.

Others have questioned the soundness of a strategy of growth‐mediated security on the grounds that high growth is often accompanied by increased inequality in the distribution of incomes, so that the people in greatest need of capability enhancement may end up benefiting least (if at all) from the general process of economic expansion. It has also been argued that the potential opportunity for expanded public provisioning may not be typically seized by a growth‐oriented government, because of its preoccupation with the expansion of material opulence rather than with the basic quality of human life. It is the slowness or absence of the so‐called ‘trickle down’ (in itself not an electrifying prospect) that makes growth an unreliable means of general advance of a community. (p.184)

What have the actual experiences been in different countries of the world? While intercountry comparisons of performance are often quite unreliable and misleading, they sometimes do provide at least a tentative basis for noting certain elementary relationships and possibilities. We shall examine here in some detail one particular set of intercountry comparisons of performance, based on the observed reduction in infant and child mortality between 1960 and 1985 in different parts of the world. Of course, infant and child mortality can by no means be interpreted as a summary index of the quality of life as a whole, or even of the nutritional status of the population, and the performance of various countries in areas related to nutrition will call for further investigation later on in this book. However, as a starting‐point for our enquiry the incidence of mortality among infants and children in different countries is a useful indicator to examine.

An internationally comparable set of estimates of the ‘under‐5 mortality rate’ (hereafter U5MR) has been constructed recently by UNICEF, and provides a useful basis for a comparative assessment of performance.7 From the information provided about under‐5 mortality rates in 1960 and 1985, it is possible to calculate the percentage reduction of the U5MR in different countries over these 25 years. The ten best performers according to this (p.185) criterion, among the developing countries, are the following (in decreasing order of performance): Hong Kong, Chile, United Arab Emirates, Costa Rica, Kuwait, Cuba, Singapore, China, Jamaica, and South Korea.8 The actual figures are presented in Table 10.39

Table 10.3 Proportionate Reduction in U5MR (1960–1985): The Top Ten Countriesa

Country

Percentage reduction in U5MR

Percentage growth rate of GNP/capita

GNP per head (US dollars)

Level of U5MR

(1960–85)

(1960–82)b

(1965–85)

(1985)

(1985)

Hong Kong

83

7.0

6.1

6,230

11

Chile

82

0.6

−0.2

1,430

26

UAE

82

0.7

n/a

19,270

43

Costa Rica

81

2.8

1.4

1,300

23

Kuwait

80

0.1

−0.3

14,480

25

Cuba

78

n/a

n/a

n/a

19

Singapore

76

7.4

7.6

7,420

12

China

75

5.0

4.8

310

50

Jamaica

72

0.7

−0.7

940

25

South Korea

71

6.6

6.6

2,150

35

(a) Excluded from the comparison are the countries of Eastern and Western Europe, Japan, New Zealand, Australia, USA, USSR, and Canada.

(b) In this column, figures in italics are for a period not exactly corresponding to 1960–82, due to non‐availability of data for the early 1960s (see World Development Report 1984, Table 1).

Source: UNICEF (1987a), Table 1; World Development Report (1984, 1987), Table 1.

On the basis of the information contained in Table 10.3, and of what is known about the experiences of the countries involved, it is possible to divide these ten countries into two distinct groups. Growth‐mediated security has clearly been an important part of the experiences of Hong Kong, Singapore, and South Korea. In fact, it is interesting to note that these three countries were among the five fastest growing economies in the world during the period under consideration, in terms of the growth rate of real GNP per capita.10 The United Arab Emirates and Kuwait can also be put broadly in the same group, even though in this period their growth rates of GNP per capita, as it is standardly measured, are not high. The phenomenal increase in the incomes of these two countries that has in fact taken place as a result of changes in international prices (in this case involving oil) fails to be captured by the growth rate of the real quantity‐index of GNP per capita.11 The fact is that both these countries have become very rich over this period, and their remarkable (p.186) success in reducing under‐5 mortality rates has been much helped by their—relatively new—opulence. Thus, a half of the ten highest performers in terms of percentage reduction of under‐5 mortality rate seem to have resorted to a strategy of growth‐mediated security, of one sort or another.

On the other hand, the other five countries (viz. Chile, Costa Rica, Cuba, China, Jamaica) have had quite different experiences. Their growth rates have been comparatively low. Moreover, as we shall discuss later in this chapter, these countries stand out sharply in having achieved far lower mortality rates than most other countries at a comparable income level. The basis of their success does not seem to rest primarily in rapid income growth, and suggests the possibility of support‐led security.12

Before we move on, two points must be stressed. First, the outstanding record of these five countries can by no means be dismissed as a statistical artefact. The demographic and health records of each of these countries have, in fact, attracted widespread attention on their own, and the statistical evidence establishing their record has been extensively scrutinized in each case.13 As will be discussed later, there is also strong independent evidence (e.g. in the form of anthropometric data and morbidity indicators) of very rapid nutritional improvement in these five countries, in addition to the observed trends in infant and child mortality rates.

Second, the prominent role of public support in bringing about these successes is also well established. All the five countries under consideration have repeatedly attracted attention for their active public involvement in various forms of social support, including the direct provision of vital commodities and social services. We shall return to this at greater length in (p.187) Chapters 11 and 12, but at this point it must already be noted that the evidence in favour of the feasibility of support‐led security is quite substantial.

10.3 Economic Growth and Public Support: Interconnections and Contrasts

The distinction made in the previous section between growth‐mediated security and support‐led security reflects an important strategic aspect of public action, but many interconnections are also involved, which have to be noted to avoid a false, total dichotomy. The precise nature of the contrast should become clearer as we re‐examine the experiences of growth‐mediated security in different countries later on in this chapter, but a few preliminary clarifications and general disclaimers might be helpful here.

First, the distinction involved is definitely not a question of activism versus disengagement on the part of the state. The governments of some of the countries which have pursued growth‐mediated security have, in fact, often been active in widely disseminating the fruits of growth. In these distributive efforts, the constructive role of the state has not been confined only to the domain of public provisioning. This role has also been geared to facilitating wide participation of the population in the process of economic growth. This has been done particularly through widespread promotion of skills and education, and the maintenance of full employment. In addition, state policies have in many cases been crucial in promoting growth itself.

Second, the contrast we have pursued is also not a simple one of market versus state provisioning. The masses can gain a share in general opulence not only through the increase of private incomes, but also through wide‐ranging public provisioning. A striking example is that of Kuwait, where enhanced opulence has provided the material basis for what is clearly one of the most comprehensive welfare states in the world.14 But in fact all the countries which we have identified as pursuing the strategy of growth‐mediated security have taken considerable advantage of the enhanced opportunities for public support provided by rapid economic growth. Other countries with high economic growth but little effort to combine it with social provisioning (e.g. Brazil or Oman) have done much worse in terms of the index of mortality decline, as we have already noted earlier in this chapter.

Third, the distinction made in the last section has little to do with the dilemma that has sometimes been construed between the pursuit of ‘growth’ and the satisfaction of ‘basic needs’. A strategy of ‘growth‐mediated’ security is not at all the same thing as the pursuit of economic growth tout court (an issue further pursued in the next section). The former need not conflict with the satisfaction of basic needs—indeed it is an approach to their satisfaction. Conversely, support‐led security does not imply surrendering the goal of (p.188) economic growth. In fact, often improvements in the quality of human life (e.g. through better health and education) also substantially enhance the productivity of the labour force. And economic growth can be crucial to the sustainability of a strategy based on generous public support. The interconnections and contrasts between the two strategies are both more extensive and more complex than would be portrayed by a simplistic dichotomy between growth and basic needs.

The real source of the contrast lies in the fact that the countries which have been identified as having made substantial use of the strategy of support‐led security have not waited to grow rich before providing large‐scale public support to guarantee certain basic capabilities. The contrast is a real one, but it should not obscure the complementarities that exist between economic growth and public support—and in particular, the prominent role played by public support in the strategies of growth‐mediated as well as support‐led security.

Despite these complementarities, dilemmas can arise in seeking a balance between the two strategies. Both growth‐oriented measures and support‐oriented measures make substantial claims on public resources as well as on public administrative capabilities. There are choices to be made in public policy‐making, and we have to face the conflicts involved.

The strategy of growth‐mediated security and the strategy of support‐led security are basically distinct, but they are not unconnected. It is just as important to recognize that much can be done to improve living standards even when growth has not yet led to a high level of GNP per head, as it is to see that economic growth can be used to provide the basis for raising the quality of life.

10.4 Growth‐Mediated Security and Unaimed Opulence

The strategy of growth‐mediated security is of somewhat deceptive simplicity, and it is important to realize how widely it actually differs from the indiscriminate pursuit of economic expansion or what might be called a strategy of ‘unaimed opulence’. A particularly crude version of the latter approach, which is in fact not uncommon, consists of attempting to maximize economic growth without paying any direct attention to the transformation of greater opulence into better living conditions. Unaimed opulence, in general, is a roundabout, undependable, and wasteful way of improving the living standards of the poor. In countries like Brazil where the poorest quintile of the population have to get by with as little as 2 per cent of national income, exclusively relying on the enhancement of general opulence would amount to accepting the need for generating 50 units of income for one unit that would go to the poor.15 In addition, opportunities for the conversion of private incomes into basic (p.189) capabilities might be expected to be particularly poor in a country where public services are persistently sacrificed at the altar of economic growth.

There are remarkable heterogeneities in what has been achieved by different countries through economic expansion and enhanced opulence. At the simplest level, the effect of increased affluence on the quality of life can be expected to depend strongly on the distribution of income. The twenty‐five developing countries for which income distribution data are available in the World Development Report 1987 include Hong Kong and South Korea from the group of five growth‐mediated successes; both have among the least inegalitarian distributions in the entire list of twenty‐five countries. In contrast, Brazil emerges as the country where the share of the richest quintile is highest, and the share of the poorest quintile second lowest. It is hardly surprising, then, that rapidly increasing general opulence in Brazil seems to have yielded so little in terms of improvements in basic aspects of the quality of life.16 It would not be difficult to find other examples to illustrate how a strategy of unaimed opulence can lead to a tremendous waste of the opportunities provided by rapid economic growth. For example, as was discussed earlier the living‐standard records of countries such as South Africa and Oman (also highly inegalitarian, with the poor left mostly to their own devices) are quite dismal, despite their relative opulence.

In many cases, an important part of the difference between unaimed opulence and growth‐mediated security relates to the expansion of employment opportunities. Each of the five countries with successful pursuit of growth‐mediated security has in fact experienced extraordinarily low rates of unemployment by the standards of developing economies, and several of them have in fact been large importers of labour power.17 The role of the state in promoting full employment in these countries has also been quite conspicuous.

Note that the actual means employed to promote or guarantee full employment have themselves displayed a great variety in different countries. In South Korea, for instance, employment promotion has been based on (1) the encouragement of labour‐intensive export industries, (2) the maintenance of comparative advantage in labour‐intensive manufacturing through the ruthless preservation of highly competitive labour markets, (3) an active policy of education, skills diffusion and training, and (4) supplementary public works (p.190) programmes.18 In Kuwait, on the other hand, the munificence of the welfare state has extended to nothing short of guaranteeing a job in the public sector to every Kuwaiti not employed in the private sector.19 Here too, one finds a plurality of strategic options for public action. But the instrumental role of the expansion of employment opportunities to share the benefits of affluence must generally be seen as a crucial one.

10.5 Opulence and Public Provisioning

We should recall that a strategy of growth‐mediated security does not necessarily make private incomes an exclusive vehicle for spreading the fruits of growth. Direct provisioning by the state can, as we have emphasized, assume an important role even when security is mediated by general economic growth. This fact is clearly illustrated by the experience of several of the countries which we have identified as having pursued the strategy of growth‐mediated security. One obvious case is that of Kuwait.20

The genesis of Kuwait's present affluence goes back to 1946, when the country started exporting oil. Since then, earnings in the oil sector have grown to remarkable heights, and in 1980 they accounted for two‐thirds of Kuwait's Gross Domestic Product.21 Until 1975, when the Kuwait Oil Company (KOC) was fully acquired by the Government of Kuwait, oil earnings accrued mainly in the form of taxes and royalties levied on the foreign‐owned KOC. At the time of nationalization, oil and gas exploitation provided employment only to a tiny fraction of the labour force. The same pattern continued after 1975, though the KOC was from then run by the government, which therefore appropriated its entire profits.

The livelihood of the bulk of Kuwait's population depends directly or indirectly on the use of these huge oil earnings through government activities and transfers. The percolation of oil revenues from the KOC to the masses has of course not assumed a particularly egalitarian character. A sizeable chunk, for instance, goes straightaway to various members of the ruling family in the form of permanent salaries, which allow this privileged élite to live in the material abundance that has often made the headlines in the Western press. Nevertheless, the greater part of oil revenues has been allocated to a massive programme of development activities, public sector employment, social services provision, and direct transfers.

A very large proportion of domestic government expenditure is accounted (p.191) for by wages and salaries, and the magnitude of public sector employment has indeed assumed staggering proportions. As we have already noted, every Kuwaiti citizen not otherwise employed is guaranteed a job in the public sector. The army of government employees has primarily busied itself with implementing the multitudinous activities of the welfare state, and in 1975 as much as 69 per cent of the Kuwaiti labour force was employed in the ‘social services’ sector.22 As Ismael (1982) puts it, Kuwait is a ‘total service society with almost every human need from the cradle to the grave serviced by institutional arrangements’.23

The availability of public services in Kuwait has indeed expanded in record time from one typical of low‐income countries to one typical of rich, industrialized economies; on this see Table 10.4 The stick of compulsory school attendance (first brandished in 1965), combined with the carrot of free education, free books, free meals, free transport, and even free clothes, has lured the younger generation to school in spite of some conservative resistance (particularly regarding girls). In 1977, government subsidies on education exceeded $600 per student.24 The male secondary‐school enrolment ratio of 86 per cent in Kuwait in 1982–4 is bettered only by South Korea and a handful of industrialized market economies (interestingly enough, the Kuwaiti ratio is (p.192) higher than those of France and the United Kingdom). The female secondary‐school enrolment ratio of 79 per cent is a little lower but is still extremely impressive, especially by the standards of developing economies. This ratio of 79 per cent compares, for instance, with 19 per cent in Oman and 36 per cent in Brazil.25

Table 10.4 Kuwait, 1960–1985: Selected Indicators

1960

1965

1970

1975

1980

1985

Population (000s)

322a

467

739

995

1,357

1,697

Government oil revenue (million KD)

159

216

289

2,440

5,187

2,295

Public health expenditure (million KD)

n/a

6

16

39

105

193

Public education expenditure (million KD)

n/a

15

32

81

172

275

Number of teachers in government schools

2,133

4,625

8,652

14,842

22,219

26,463

Number of students in government schools (000)

43

85

134

192

294

361

Number of physicians in government hospitals

n/a

451

540

932

1,921

2,528

Rate of illiteracy (percentage of population aged 10 and above)

n/a

46

39

36

29

16

(a) For the year 1961.

Source: Government of Kuwait, Ministry of Planning, Central Statistical Office, Annual Statistical Abstract (1970, 1974, 1978, 1980, 1982, 1984, 1987).

Sophisticated medical services are provided free of charge to the entire population, Kuwaiti and non‐Kuwaiti. Far‐reaching public provisions are also made in areas such as housing, water supply (a precious commodity in Kuwait), electricity, transport and communications, and the subsidization of basic commodities. In addition, Kuwait has a system of large‐scale direct transfers and financial help to low‐income families which is by any standards one of the most generous in the world.26 If socialism were reduced simply to state ownership of the means of production and generous provision of ‘social wages’, Kuwait would appear to be one of the most obviously socialist countries in the world!

The case of Kuwait is admittedly a special one in many respects, but the general notion that growth can facilitate public support applies more generally. In fact, it is rather striking how extensively the governments of the five ‘growth‐mediated security’ countries have been driven (under the influence of a variety of political pressures and motivations) to use wide‐ranging public provisioning in order to transform the material fruits of growth into secure minimal living standards for the greater part of the population. This applies even to governments which have a fierce reputation of non‐interventionism in the private enterprise economy, of which Hong Kong is a classic example.

It is true that the government of Hong Kong, in sharp contrast with that of Kuwait, commands a relatively small proportion of the national income. However, within total government expenditure social services are the main item, with a share as large as 38 per cent in 1986.27 This has permitted not only substantial improvements in the provision of educational facilities and health care but also the development of a sophisticated system of social assistance, including means‐tested income support along similar lines to the Supplementary Benefits in the UK. It also includes huge housing subsidies which, in terms of generosity and coverage, count among the most substantial income support schemes in the world.28 The results of economic growth as well as (p.193) public support founded on it include a remarkably high life expectancy (viz. 76 years, which is comparable with the expectation of life in the most advanced industrial economies).

The opportunity for enlarged public provisioning provided by rapid growth has also been seized to various degrees by the other three of the five countries identified earlier as cases of ‘growth‐mediated security’. The experience of the United Arab Emirates bears some resemblance to that of Kuwait.29 In Singapore, the government has an impressive record of extensive activism in both economic and social matters, which has been seen as a major factor behind the rapid improvement of living conditions in that country in the last few decades.30 The experience of South Korea is discussed at greater length in the next section.

10.6 Growth‐Mediated Security: The Case of South Korea

A good illustration of the subtleties involved in a strategy of growth‐mediated security, and of the role played at different levels by public action, is provided by South Korea. As we have seen, the growth rate of GNP per capita in that country over the last few decades has by any standards been a highly impressive one. Nor can it be denied that in this case rapid growth has provided the basis for very tangible improvements in basic components of the quality of life. We have already noted South Korea's outstanding reduction in infant and child mortality rates. There are many other, direct and indirect, indications of rapid advances in living standards. Indeed, there is solid evidence that during the period under consideration (viz. 1960 to 1985), the nutritional status of children has markedly improved; morbidity rates for communicable diseases have shrunk very fast; the incidence of absolute poverty (as standardly measured) has rapidly decreased, both in rural and in urban areas; the unemployment rate has stabilized at a remarkably low value (around 4 per cent); general educational standards have reached exceptionally high levels; and real wage rates have consistently increased, both in agriculture and in manufacturing. Some relevant indicators are presented in Table 10.5.31

Table 10.5 South Korea, 1960–1985: Selected Indicators

Year

Infant mortality rate (per 1,000)

Height of children aged 6 (cm.)

Incidence of poverty (%)

Gini index of income inequality

Unemployment rate (%)

Index of real wages in industry (1970 = 100)

Secondary school enrolment (%)

Male

Female

(1)

(2a)

(2b)

(3)

(4)

(5)

(6)

(7)

1960

85

111.1a

110.4a

n/a

0.448

n/a

n/a

n/a

1965

n/a

111.7

110.8

41

0.344

7.4

58

54

1970

53

112.7

111.7

23

0.332

4.5

100

66

1975

41

113.3

112.3

15b

0.391

4.1

127

77

1980

37

115.0

113.7

10

0.389

4.5

219

96

1985

27

116.5

115.4

7

0.363

4.0

286

99

(a) Figure relating to 1962.

(b) Figure relating to 1976.

Sources: (1) Suh (1984: 162) and UNICEF (1987a), Table 1. (2) Government of the Republic of Korea (1963, 1965, 1970, 1987). (3)–(5) Hahn (1989), Figures 5, 7, and 13b. (6) Calculated from Hasan and Rao (1979), Table D. 38, and Hahn (1989), Figure 22. (7) Hahn (1989), Figure 28. We are extremely grateful to Byung Whan Kim (London School of Economics) for guidance to South Korean statistics.

The extent of government involvement in income redistribution and social welfare programmes has been, until recently, rather small by international standards. To take only one example, before 1976, South Korea had no public (p.194) (p.195) health care system worth the name, and no form of broad‐based medical assistance or medical insurance scheme. Health care was predominantly in the hands of private professionals, especially pharmacists.32 Many other aspects of social welfare have also been rather neglected until recently, and the South Korean government has consistently rejected the option of developing into a ‘welfare state’.33 Nor has it expressed much direct concern about income redistribution.34

One could be led by all this to conclude that in South Korea private enterprise has been the driving force not only of economic growth in the narrow sense (as is widely recognized), but also of wider improvements in the quality of life. It is perhaps not without reason, then, that, South Korea has been variously seen as an archetype of the fecundity of capitalism, a ‘free enterprise model’ for other developing countries, an illustration of the redundancy of planning, and, generally, a brilliant product of what some have called the ‘market order’.35

There is clearly an element of truth in these characterizations. However, there are also serious qualifications to be made from several different perspectives.

First, it must be pointed out that the South Korean economy in the late 1940s offered a rather unusual base for equitable growth. A particularly important factor was the relatively equal distribution of assets (including skills, education and land), in the creation of which the government played a major role.36 Equitable growth was reinforced by the labour‐intensive orientation of industry (we have already noted South Korea's remarkable wage and employment figures), and this orientation took place within a structure of incentives and inducements carefully planned by the government. It is impossible to understand South Korea's experience without reference to the major role which the government has played in enabling the population at large to participate fruitfully in the process of economic growth.

Second, as has been widely noted, the hands which signalled South Korea's economic expansion have been much less ‘invisible’ than would appear at first sight. It is true that South Korean economic policy and planning has mainly taken the form of creating an environment conducive to private enterprise of (p.196) the favoured kind. However, the role of government policy in planning the nature of economic growth and in intervening to shape the direction of investment and expansion has been both pervasive and enormously effective.37

Government involvement in this field has taken a wide variety of forms, including extensive credit controls and incentives, import substitution measures, infrastructural investments, the dissemination of information, a sophisticated tax administration system, and the promotion of an active and competitive labour market.38 As far as the alleged ‘redundancy of planning’ is concerned, South Korea has, in fact, had regular Five‐Year Plans since 1962 and also the rare distinction of implementing them successfully.39 The various instruments of state policy for shaping the South Korean economy are systematically put together in this exercise of integrated planning.

Third, a further question concerns the precise involvement of the South Korean government in measures of direct public support. The record in this respect has been highly uneven between different areas of intervention, and is therefore difficult to assess. For instance, while public provisions for health care were rather meagre until the late 1970s (as noted earlier), the state has been extremely active in the area of education for a long time.40 And while it is true that the South Korean government has, at least until recently, eschewed the idea of large‐scale welfare programmes, it has, on the other hand, had a long‐standing and (in some ways) pioneering concern for the prevention of acute destitution.41

Moreover, there have been important policy developments within the period under consideration (viz. 1960 to 1985). Specifically, the commitment of the government to social policies has rapidly increased in the second half of this period. In the area of health care, for instance, ambitious programmes of medical assistance and medical insurance initiated in the late 1970s have marked a reorientation of earlier policies.42 Government expenditure on health, education housing, and social security has grown (p.197) very fast since then—as fast as 33 per cent per year during the period 1978–82.43

The important role played by direct measures of public support in supplementing the normal operation of the economy in South Korea has been particularly clear during the recession of the early 1980s. It has been observed that, in this event, direct measures of public support (such as public works programmes and direct transfers to the needy) emerged as crucial policy instruments for the protection of vulnerable groups.44

The Korean experience is thus far from one of laissez‐faire, and in fact richly illustrates the diverse roles that state planning and action can play in influencing—directly and indirectly—the expansion of basic capabilities within a strategy of growth‐mediated security. The active nature of state policy in South Korea within the structure of promoting security through participatory growth would be hard to deny.

The positive involvement of the state in promoting participatory growth may seem surprising given the repressive nature of the Korean government over this period. The possible contribution of—open or latent—political opposition in influencing the state in the direction of promoting living standards should not, however, be underestimated. The government has had to cope with vocal political dissent (especially from the student community) and frequent outbursts of public protest, which have also been widely reported abroad.45 The precise role of adversarial public action in affecting the direction of state policy is an aspect of the South Korean experience that merits further attention.

10.7 Support‐Led Security and Equivalent Growth

In the preceding sections we have discussed the experiences of some of the countries that have successfully pursued what we have called ‘growth‐mediated security’. In Chapters 11 and 12, we will examine selected cases of ‘support‐led security’. In this last section of the present chapter, we discuss some general issues related to assessing the effectiveness of the strategy of support‐led security.

To start with, it is of some interest to consider briefly how the levels of under‐5 mortality in the countries which we have identified as having pursued a strategy of support‐led security compare today with the levels observed in (p.198) other countries with a similar level of GNP per capita. This can be appreciated from a simple scatter diagram showing the levels of GNP per capita and U5MR (in 1985) for different countries—see Figure 10.1 and also Table 10.6.46

Economic Growth and Public Support

Fig. 10.1 GNP per capita and under‐5 mortality (1985): international comparison

Source: Based on data for 120 countries provided in UNICEF (1987a) (nine observations represent two countries each due to ‘superimposition’). Data on GNP are not available for Cuba.

Table 10.6 Actual Value of U5MR (1985) as Percentage of Value Predicted on the Basis of GNP Per Capita

Ratio (%)

Growth rate of GNP per head (1965–85)

Ten developing countries with lowest ratio

China

32

4.8

Jamaica

33

−0.7

Sri Lanka

35

2.9

Guyana

36

−0.2

Costa Rica

37

1.4

Burma

42

2.4

Chile

45

−0.2

Mauritius

46

2.7

Hong Kong

50

6.1

Madagascar

52

−1.9

Other (selected) countries

Singapore

61

7.6

Zaire

73

−2.1

South Korea

79

6.6

Brazil

172

4.3

Kuwait

216

−0.3

Oman

819

5.7

Note: Calculated from the same sources as Table 10.3. The ‘predicted value’ of U5MR is obtained from OLS regression of U5MR on GNP per capita (both in logarithmic form). The regression involves 120 countries for which data were available on both variables. (For 10 countries, including Cuba, U5MR estimates existed but not GNP estimates.)

The interpretation of the position of different countries requires some caution. For instance, it would not make much sense to presume that any country with a low U5MR relative to other countries at a comparable income level must be particularly exemplary (as is sometimes assumed in analysing ‘basic needs performance’). This is so not only because one simple way to reach that position is to do it, as it were, the Zaire way, to wit, by producing negative growth!47 As income goes down, the U5MR looks better in comparative terms. But more importantly, this way of identifying successful countries would hide the success of countries that have achieved a low U5MR through a growth‐mediated strategy. (p.199)

The goal of development is to improve the quality of life, not to improve the quality of life relative to income. In pursuing the former goal, the growth of income can, as we have already seen, play an important role. The achievement of a low level of under‐5 mortality relative to income (through, say, extensive public provisioning) would give little grounds for congratulation if the expenses involved had the effect of slowing down the growth of incomes to the extent that the absolute value of U5MR ended up being higher than it would have been with less public support and more growth.

In spite of these reservations, the information provided by Figure 10.1 and Table 10.6 is instructive in giving us a part—an important part—of the picture. It can, in fact, assist our understanding of the gains that can be obtained through public support for a given level of opulence. The figures indicate the extent to which the success of the five countries which we have (p.200) identified as having made effective use of a support‐led strategy (viz. Chile, China, Costa Rica, Cuba, Jamaica) has indeed relied on their ability to deviate from the ‘standard’ relationship between GNP and mortality. Their record in breaking the shackles of low income and poverty is indeed impressive, with (for instance) China achieving a level of under‐5 mortality rate (U5MR) less than a third of that predicted on the basis of its income level alone. Similarly striking advantages are enjoyed by the other countries in this group.48

This exercise is also helpful in identifying other countries which have achieved great success in reducing mortality despite low incomes, but did not appear in our initial group of ‘top ten’ performers because the period of rapid improvement in their case failed to overlap substantially with the time period used in this classification (viz. 1960–85). On the basis of Figure 10.1 and Table 10.6, the most obvious candidate for this diagnosis is Sri Lanka. It can be seen that Sri Lanka, like China, has a level of U5MR about a third of what one would expect (on the basis of international comparisons) given its level of GNP per capita. But the period of spectacular advance in Sri Lanka did not lie primarily in the last two or three decades, and for the 1960–85 period specifically Sri Lanka has not been an outstanding performer. Accelerated breakthrough in mortality reduction occurred earlier, during the 1940s and 1950s. These were, in fact, also the decades of rapid growth of public support in the form of free or subsidized distribution of rice (introduced in 1942) and intensive expansion of public health services (beginning in the forties, partly related to a campaign to conquer malaria). Though it has a different time pattern from the other countries discussed in this chapter, Sri Lanka's pioneering experience is of great general interest, and will be further investigated in Chapter 12.49

A careful examination of the levels of per‐capita GNP and under‐5 mortality can also help us to assess the relative gains that different countries might expect to obtain from alternative strategies. We could, for instance, examine questions of the following sort. If a country like, say, China had resorted to a ‘standard’ amount of public support instead of an exceptional one, how much faster would it have had to grow over a specified period in order to reach the level of U5MR it is observed to have in 1985? Clearly, only rather speculative—and at best approximate—answers to this question can be arrived at. Simple examinations of the current relationship between U5MR and per‐capita GNP (p.201) in different countries can, however, assist our intuition about the plausibility of different answers.

An elementary but nevertheless useful answer to the question posed can, for instance, be obtained in the following manner. First, we estimate the level of per‐capita GNP which an ‘average’ country appears to need in order to experience China's current level of infant and child mortality.50 Second, we calculate the extra growth of income that China would have been required to achieve over a specified period to reach that level of GNP per capita.51 The results of such a calculation, taking 1960–85 as the reference period, are presented in the first column of Table 10.7. They suggest that, to reach the observed level of U5MR in 1985 in the absence of outstanding public support measures, China would have had to raise its annual growth rate over the whole period by 7 to 10 per cent of the GNP per capita (e.g., if the actual growth rate experienced was 4 per cent, it would have had to raise it to somewhere between 11 and 14 per cent per year).

Table 10.7 Extra Annual Growth of Per‐Capita GNP Required by China Between 1960 and 1985 in Order to Reach Its Observed Level of U5MR in 1985 in the Absence of Outstanding Public Support Measures (%)

Additional growth requirement for alternative assumed values of ‘country advantage’ (CA)

CA = 0

CA = 0.10

CA = 0.20

CA = 0.30

7.1

6.4

5.7

5.0

(9.9)

(8.9)

(7.9)

(6.9)

Source: See text. The numbers without brackets are based on a regression involving all countries for which the relevant data are available in UNICEF (1987a, 1988). The numbers within brackets are based on a regression involving only developing countries.

It may be argued that this calculation can be somewhat misleading in that it attributes the whole of China's deviation from the average relationship between GNP and U5MR to its outstanding measures of public support. What if China's outstanding record is due, say, to favourable ecological circumstances, or some other advantage applying to China as a country? There is no obvious reason to believe that such advantages—ecological or otherwise—are enjoyed by China. Nevertheless, Table 10.7 also presents (in the last three columns) alternative estimates assuming different levels of ‘country advantage’, where the latter is defined as the proportion of China's deviation from the average relationship between GNP and U5MR that can be attributed to favourable circumstances unrelated to public support.52 For reasonable values of country advantage, additional growth requirements over the 1960–85 period in the absence of outstanding public support measures remain very high—5 per cent of GNP per capita each year under the set of assumptions most favourable to the growth scenario.53 (p.202)

In spite of their illustrative nature, the calculations we have presented bring out that the ‘economic growth equivalent’ of well‐planned public support is very large indeed.54 That is, the results that countries such as China have achieved through direct public support could only have been obtained through extremely fast economic growth if they had followed a ‘path’ similar to that typical of other developing countries.55

Whether immediate and extensive measures of public support in a poor country lead to slower economic growth is, of course, an extremely complex question, and we shall not pursue it here. As was mentioned earlier, the interactions between public support and economic growth include not only dilemmas and trade‐offs (e.g. the allocation of resources between immediate consumption and investment), but also many positive links (e.g. the effects of improved health and nutrition on productivity). What is worth noting here is that, given the very large ‘growth equivalent’ of public support, only the existence of some remarkably powerful (and negative) trade‐off between public support and economic growth would seriously undermine the case for extensive involvement in public support at an early stage of development.

In the next two chapters, we shall discuss in some detail the experiences of (p.203) selected countries in the pursuit of support‐led security. Some of the general issues and dilemmas arising from a programme of public action for social security will be reconsidered in the concluding part of this book.

Notes:

(1) The figure of 69 years for China appearing in the World Development Reports may be a bit of an overestimate. On this see Chapter 11. There have been abrupt and unexplained upward revisions of the reported life expectancy for South Africa in the later World Development Reports.

(2) It is worth noting that just as we may be concerned with the distribution of GNP in addition to its mean value per capita, similarly also in the case of such indicators as life expectancy, there is the need to distinguish between their mean values and their patterns of distribution over the population.

(3) Cf. Robert Kennedy's remark: GNP ‘measures everything, in short, except that which makes life worthwhile.’ (Statement made by Robert Kennedy at the University of Kansas during his presidential campaign in 1968; cited in Dowd 1987.)

(4) See Winter (1986) for an illuminating analysis of the effects of the First World War on public distribution and public involvement, and their impact on living conditions in Britain. The experience of the Second World War is discussed in great detail by Titmuss (1950: Chapter 25), who examines the evidence indicating a strong relationship between the surprisingly good health conditions of the British population during the war (including a rapid improvement of the health status of children) and the extensive reach of public support measures in that period. As Titmuss put it, ‘by the end of the Second World War the Government had, through the agency of newly established or existing services, assumed and developed a measure of direct concern for the health and well‐being of the population which, by contrast with the role of Government in the nineteen‐thirties, was little short of remarkable’ (p. 506). According to Titmuss, the most influential part of social policy during the war related to employment provision and food rationing. This conclusion is strongly corroborated by Hammond's detailed study of the ‘revolution in the attitude of the British State towards the feeding of its citizens’ which took place after 1941 (Hammond 1951). On these issues, see also Marrack (1947), McKeown and Lowe (1966: 131–4), McNeill (1976: 286–7), Szreter (1988).

(5) It is of some interest to note that in the case of Japan, too, the rate of expansion of longevity was substantially higher during the period covering the Second World War and immediately after (see the figures presented in Preston et al.] 1972, as well as the further discussion in Drèze and Sen 1988). This was also a period of rapid expansion of public support, and it is plausible that, as in the case of Britain, this expansion was a crucial factor in the reduction of mortality rates especially after the war. We are grateful to Akiko Hashimoto for helpful discussions on the empirical evidence related to this observation. See also Taeuber (1958), Shigematsu and Yanagawa (1985) and Morio and Takahashi (1986).

(6) This line of reasoning has also been influential in bringing about drastic reductions in social services as part of ‘adjustment programmes’ in developing countries faced with mounting debts and trade imbalances. On the adverse social impact of such cut‐backs, see Jolly and Cornia (1984), Cornia et al.] (1987) Bell and Reich (1988), and the WIDER studies by Nora Lustig and others in Jayawardena (forthcoming).

(7) The nature of the U5MR index is explained in UNICEF (1987a: 126). The information on U5MR for 130 countries, on which our analysis is based, appears in Table 1 of the same publication. In the remainder of this chapter, the term ‘developing countries’ will be used to refer to the hundred countries left in that table after excluding the countries of Western and Eastern Europe, North America, Japan, New Zealand, Australia and the USSR.

(8) In fact, North Korea was also in our preliminary list of countries selected for highest reduction in U5MR. However, it was removed from the list because it turns out that the figures of U5MR for North Korea published by UNICEF were not obtained independently but were simply assumed to be the same as those applying to South Korea. Note also that the recorded reduction in U5MR for the period 1960–85 in China would be completely misleading if the base‐level mortality rate actually corresponded to 1960, when famine was raging and mortality had shot up sharply (see Chapter 11). However, the 1960 figure given in UNICEF (1987a) is clearly not one based on the famine years, and rather appears to be based on an extrapolation from pre‐famine figures (see the infant mortality data in Piazza 1986, and Jamison and Piazza 1987).

(9) It could be argued that a given percentage reduction in U5MR between 1960 and 1985 provides different indications about a country's ‘performance’ in mortality reduction depending on the initial level of U5MR in 1960. However, when percentage reductions in U5MR over this period are examined after ‘controlling’ for these initial levels (this is done by looking at the residuals of a regression of the 1985 U5MR levels on the 1960 levels, both in logarithmic form), the identification of the best performers is not substantially affected. In fact, by this alternative criterion all the ‘top ten’ countries appearing in Table 10.3 still outperform all other developing countries except three (viz. Jordan, Saudi Arabia, and Syria).

(10) See World Development Report 1987, Table 1. This statement refers to the period 1965–85 (rather than 1960–85), the only one for which the figures are presented in that report. For the period 1960–82, however, these three countries are also among the five fastest growing economies in the world (see World Development Report 1984, Table 1).

(11) This problem is explicitly mentioned in the Technical Notes accompanying the World Development Reports, with special reference to oil‐producing countries (see e.g. World Development Report 1984, p. 275). In the case of these countries, the use of, say, a Consumer Price Index to deflate nominal GDP figures (instead of the GDP deflator applied in the World Development Report figures) gives a much fairer idea of the massive increase in purchasing power experienced by their populations in recent decades. For instance, using the Consumer Price Index published in the Annual Statistical Abstract of Kuwait as a deflator, the annual growth rate of real GDP in Kuwait after 1973 (the starting year of this price index series) is of the order of 11% until the early 1980s (after which it decelerates due to a sharp reduction in oil prices). The annual growth rate of real consumption expenditure (public and private) over the 1973–81 period is even higher, of the order of 14% (all figures have been calculated from the 1970, 1982, and 1987 issues of the Annual Statistical Abstract of Kuwait).

(12) The Chinese growth rate appearing in Table 10.3 is quite impressive, and might be seen as suggesting that the basis of China's success may well lie as much in economic growth as in direct public support. This question is further discussed in the next chapter, where it is argued that (1) China's growth rate during the period of interest has been much exaggerated, and (2) economic growth has followed rather than preceded the wide‐ranging measures of public support which must be seen as the main source of China's success.

(13) There tend to be many gaps in international statistics of life expectancy and related indicators (on this see Murray 1987). However, the reliability of such statistics is much greater for these five countries. The abundance and quality of demographic and health statistics in Costa Rica and Chile since the 1950s are rather exceptional, and the impressive records of these two countries in the areas of health and nutrition have attracted very wide attention—the experiences of these two countries are further discussed in Chapter 12. Cuban statistics have to be used with great care, but health and demographic statistics are among the more reliable ones (see the thorough discussions in Mesa‐Lago 1969, 1979, Diaz‐Briquets 1983, and Santana 1987: Appendix), and Cuba's achievements in the area of health and nutrition are well established—see Brundenius (1981, 1984), Handelman (1982), Diaz‐Briquets (1983), Valdes‐Brito and Henriquez (1983), Werner (1983), Muniz et al.] (1984), Meegama (1985), Eckstein (1986), and Santana (1987). The literature on China's experience of rapid health and nutritional improvement (further discussed in the next chapter) is enormous. Some useful references, which also carefully discuss the statistical evidence, include Jamison and Trowbridge (1984), World Bank (1983a, 1984a, 1985), Jamison et al.] (1984), Jamison (1985), Xu Su‐en (1985), Piazza (1986), Riskin (1986), Banister (1987), Jamison and Piazza (1987), Hussain (1987), and Hussain and Feuchtwang (1988). On the Jamaican experience, which too will be commented on (Chapter 12), see Jameson (1981), Cumper (1983), Gunatilleke (1984), Boyd (1987), Samuels (1987), Moran et al.] (1988), and Mesa‐Lago (1988a).

(14) See section 10.5 below. It is important to note that welfare provisions in Kuwait, while generally extensive, also discriminate sharply in favour of Kuwaiti citizens as opposed to other residents.

(15) Even this is conditional on the assumption that the distribution of income remains unchanged; in fact, inequalities may well increase with the single‐minded pursuit of economic growth. The income distribution figures are from World Development Report 1987, Table 26.

(16) In 1985, Brazil had exactly the same U5MR as Burma, even though the latter had just about one‐tenth of Brazil's GNP per capita. In fact, Burma had started off with a 43% higher U5MR in 1960, and grew at about only a quarter of the rate of Brazil, and still caught up with Brazil in terms of U5MR by 1985. Among all developing countries, Brazil had the tenth fastest growing economy between 1960 and 1982 (with an estimated annual growth rate of GNP per capita of 4.8%), but only occupies the 56th position in terms of percentage reduction in U5MR (1960–85). See UNICEF (1987a), Table 1, and World Development Report 1984, Table 1. For an informative analysis of Brazil's experience, see Sachs (1986).

(17) For further evidence and discussion of the outstanding employment records of these five countries, see al‐Sabah (1980), Ismael (1982), Koo (1984), Sherbiny (1984), Hajjar (1985), Nijim (1985), Government of Hong Kong (1986), Krause (1988), Richardson and Kim (1986), and Hahn (1989).

(18) See section 10.6 of this chapter.

(19) See e.g. Ismael (1982). We should stress that the guarantee of employment applies only to Kuwaiti nationals. With respect to non‐Kuwaiti residents (a large part of the population), full employment seems to be ensured through the no‐nonsense method of stipulating that ‘a non‐Kuwaiti must leave the country once unemployed’ (Ismael 1982: 119).

(20) The factual basis of the following account is derived from al‐Sabah (1980), Ismael (1982), Harrison (1985), Nijim (1985), Public Institution for Social Security (1985), Hammoud (1986, 1987), and Nagi (1986), aside from official statistical sources.

(21) Annual Statistical Abstract of Kuwait 1987, pp. 266–7. As noted earlier, the contribution of oil revenues to Kuwait's economy declined after (1980 due to the fall in oil prices.

(22) Annual Statistical Abstract of Kuwait 1987, p. 136.

(23) Ismael (1982: 105).

(24) al‐Sabah (1980: 58).

(25) UNICEF (1987a), Table 4.

(26) According to al‐Sabah, the percentage of Kuwaiti families benefiting from direct financial aid from the government is the highest in the world (about 25 per cent), and the government of Kuwait is strongly committed to the ‘lower‐income Kuwaitis’, the latter being defined as those ‘whose monthly income does not exceed $550’ (al‐Sabah 1980: 57, 158; emphasis added).

(27) Government of Hong Kong (1987), Appendix 8a.

(28) The strides that have been made in Hong Kong during the last few decades in the areas of health care, education, social assistance, and housing are discussed in Heppell (1973, 1974), Chow (1981), Drakakis‐Smith (1981). King and Lee (1981), Lee (1983), Yeh (1984), and Government of Hong Kong (1987). The role played by welfare provisions in the development experience of the ‘four little tigers’ (Hong Kong, South Korea, Singapore, and Taiwan) is also investigated in Midgley (1986).

(29) See e.g. Taryam (1987), especially Chapter 7.

(30) On this see the in‐depth analysis of Ng Shui Meng (1986b). The pervasive influence of the government on economic and social life in Singapore (which ranges from housing 81% of the population in 1986 to ‘discouraging long hair and corrupting music and dance’) is also discussed in detail by Krause (1988).

(31) See also Suh (1984) on nutrition and morbidity; Richardson and Kim (1986) on employment; and McGinn et al. (1980) on education. According to UNICEF (1987a), Table 4, in 1985 South Korea had the fourth highest male literacy among all developing countries. Its secondary‐school enrolment rate was the highest among developing countries for males (92%), and second highest for females (86%). In the latter respect South Korea surpassed a large number of developed, industrialized countries.

(32) South Korea's health care system is discussed in detail in Park and Yeon (1981) and Yeon (1982, 1986). See also Golladay and King (1979), Suh (1984), Yoon and Park (1985a), and Bahl et al.] (1986). The year 1976 marked a turning‐point in the development of health care policy in Korea, with the introduction of the medical assistance and medical insurance programmes. In recent years increasing efforts have also been made to develop the public health care system.

(33) See the references cited above, and also Republic of Korea (1979). It is worth noting that important changes have taken place in South Korea in the area of social welfare policy since the mid‐1970s—see e.g. Suh (1984), Government of Korea (1986), Midgley (1986), and Suh and Williamson (1987).

(34) See e.g. Kim and Yun (1988).

(35) On these various characterizations and important analyses related to these perspectives, see Bauer (1972, 1981, 1984), Little (1982), Lal (1983) and Balassa (1988).

(36) See e.g. Suh (1984), Bahl et al. (1986), Michell (1988).

(37) Numerous writers have stressed and documented this point. See, for instance, Datta Chaudhuri (1979), Hasan and Rao (1979), Sen (1981b), Wade (1983), Evans and Alizadeh (1984), Koo (1984), Bahl et al. (1986), Hamilton (1986), Midgley (1986), Richardson and Kim (1986), Bagchi (1987), Toye (1987), Amsden (1989), Kim and Yun (1988), Kuznets (1988), Michell (1988), Qi (1988), White (1988), and Alam (1989).

(38) On the last point, see e.g. Richardson and Kim (1986). It must be noted in particular that the South Korean government's anxiety to maintain competitive labour market conditions has had a counterpart in very energetic and effective policies aimed at upgrading skills through education and training.

(39) See Suh (1984) and Kuznets (1988).

(40) See McGinn et al. (1980) and Amsden (1989).

(41) This concern has been visible, for instance, in the promotion of a ‘Work‐conditioned Assistance Programme’ (initiated in 1964) and in the provisions associated with the Livelihood Protection Act of 1961. For further discussion of these and related social security measures, see Yoon and Park (1985a, 1985b).

(42) On this, see e.g. Park and Yeon (1981) and Yeon (1986).

(43) Suh (1984), Table XII.3. As Yeon observes: ‘In recent years government policy in the Republic of Korea has recognised the fact that rapid economic growth is a necessary but not sufficient condition for improving the income and standard of living of the population’ (Yeon 1986: 153). This concern is indeed evident in recent planning documents (e.g. Government of Korea 1986).

(44) For a detailed analysis of this issue, see Suh and Williamson (1987), as well as Cornia et al. (1987), vol. i.

(45) See Steinberg (1988) on adversarial politics in South Korea and their relation to economic policies. See also McGinn et al. (1980).

(46) Note that Figure 10.1 is drawn on the basis of logarithmic scales. This means that, each time we move one unit up the vertical axis, U5MR is multiplied by a little less than three (and correspondingly for GNP per capita along the horizontal axis). Table 10.6 is based on a log‐linear, ordinary‐least‐squares regression of U5MR (1985) on per‐capita GNP (1985) for 120 countries. The over‐all results reported in this section in this section are quite robust to alternative specifications of the set of countries considered in the analysis (e.g. including or excluding developed economies, or oil‐exporting countries).

(47) Zaire had an annual growth rate of –2.1% in 1965–85 (World Development Report 1987, Table 1). It has also experienced a steady deterioration of its living standards in recent years.

(48) For Cuba, however, the relevant data (particularly GNP estimates) needed for inclusion in Table 10.6 are not available.

(49) The case of Mauritius (also appearing in Table 10.6) bears some resemblance to that of Sri Lanka. Indeed, the links between public policy and social achievements in Mauritius in the 1940s and 1950s have attracted considerable attention. On this and related aspects of Mauritius's development experience, see Meade et al.] (1968), Titmuss and Abel‐Smith (1968), Tabutin (1975), Mehta (1981), Minogue (1983), Tabutin and Sombo (1983), Selwyn (1983), Joynathsingh (1987).

(50) This estimated level of income, say y *, is inferred from the regression relating U5MR (1985) to per‐capita GNP (1985), mentioned earlier.

(51) It is easy to show that, if y is China's actual per‐capita GNP in 1985, and t the length of the specified period, then this ‘extra growth requirement’, say g, is simply (In y * – ln y)/t. Note that this formula does not require us to know China's actual growth rate at any time. This is fortunate since, as we shall see in the next chapter, China's past growth rates are extremely hard to ascertain.

(52) Formally, the ‘country advantage’ is the proportion of China's ‘residual’ in the earlier regression that is considered attributable to favourable circumstances rather than to public support. It can be shown that, if the assumed value of the country advantage is α‎, then the ‘extra growth requirement’ is simply (1 – α‎).g, where g is calculated as before.

(53) The highest value of ‘country advantage’ for China used in Table 10.7 is 30 per cent. This upper bound is obtained by treating country advantage analogously to a ‘fixed effect’, and attributing all of China's favourable deviation from the international regression line around 1960 to its country advantage. Note that this is likely to be a substantial overestimate of China's ‘country advantage’ since by that time China already had an outstanding history of public support. Of course, whether there is any ‘country advantage’ at all for China is far from clear, and our use of this ‘scaling down’ of the achievements of support‐led strategy is motivated by making as conservative an estimate as possible.

(54) An alternative approach to the analysis of the ‘growth equivalent’ of public support, leading to a similar general conclusion, was presented in Sen (1981b). For a critical examination of the possible interconnections, see Birdsall (1988, 1989).

(55) It must be emphasized that this ‘standard’ path as it exists today is not the same as what we have called ‘growth‐mediated security’. In fact, it is interesting that all the countries identified earlier as having followed that strategy have large negative residuals in regressions of U5MR (1985) on GNP (1985) and U5MR (1960). This confirms that their experiences have involved a great deal more than a vague reliance on economic growth, as with ‘unaimed opulence’.