• - guaranteed employment
• - the distribution of productive assets
• - food security
• - economic security

Structural reforms and safety nets

The discussion in section 2 emphasized the importance for human development of the set of incentives, i.e., of ensuring that price signals accurately reflect costs and benefits, of guaranteeing access to markets to everyone on equal terms and of lowering barriers to entry where there is discrimination. Section 3 underlined the role of government in achieving human development objectives and emphasized the large potential gains that could be realized by a reallocation of public expenditure. In a few countries improvements in incentives and in the composition of government spending will suffice, but in most developing countries these are necessary steps that will have to be reinforced by structural reforms and institutional changes.

In this section a small number of structural reforms are discussed: guaranteed employment, an equitable distribution of productive assets, food security and nutrition, and economic security and welfare. These structural reforms are not substitutes for reforms of incentives and government spending, they are complementary to them. In combination they should go a long way toward ensuring that all members of society have opportunities to fulfill their potential as human beings.

The structural reforms singled out for discussion have been carefully selected. In some sense they are fundamental for if everyone capable of productive work has an opportunity of employment, if all households have a share in society's productive assets, if every child, woman and man is assured of enough food to sustain an active life and if the lame, the sick and the infirm are entitled to a minimum of economic security then it can truly be said that a basis exists for rapid and sustained human development. This does not imply that in some countries other reforms, such as a reform of the system of governance, may not merit higher priority, and to that extent our list should be regarded as illustrative only, but the issues raised here are ones which every society sooner or later must confront.
 

Guaranteed employment

Employment is fundamental to a human development strategy. Employment provides people with a source of income, it contributes to the output of goods and services and it gives workers a sense of dignity by enabling them to participate in the activities of the community and to contribute something of value. 58 Equally important, employment is necessary to reap the full benefits of investment in human capital. The returns to human development expenditure will fall dramatically if those who embody human capital are unable to put their energy, skills, knowledge and initiative to productive use.

Well conceived macroeconomic policies can help to create an economic environment conducive to rapid employment creation and a higher average productivity of labour. The structure of incentives, discussed in section 2, obviously has a role to play. Three things are worth a special mention here. First, it is important that policy makers ensure that there is a positive real rate of interest in the formal sector capital markets so that borrowers have an incentive to substitute abundant labour for scare capital. Second, an overvalued exchange rate should be avoided so that there is no incentive to substitute imported foreign equipment for domestic labour. Third, obstacles to the smooth functioning of the labour market should be removed, including the elimination of discrimination against women and other groups, the removal of all barriers to entry into markets and the weakening of factors which restrict access.

These microeconomic policies have obvious implications for the volume of employment. In addition, there are trade and budgetary policies. We have placed considerable emphasis on the importance of grasping global opportunities and the role a human development strategy can play in making this possible. Of course the opportunities must be there to be grasped and the international environment for job creation must be favourable. The developed countries, which dominate global trade, have clear responsibilities for maintaining an open trading system and a bouyant world economy. Given the international environment, however, it is up to the government of each developing country to ensure that its trade regime encourages an efficient use of the country's existing human capital and provides encouragement to further human capital formation. This can usually best be done by exploiting actual and potential comparative advantages in employment intensive and human capital intensive activities. In addition, the government should manage its budget in such a way that the pressure of aggregate demand on total available resources is neither so great as to cause rampant inflation nor so small as to cause excess productive capacity, depressed investment incentives and open or disguised unemployment of labour. These are the core employment policies and other policies and programmes discussed below should be regarded as supplementing where necessary these core policies.

Beyond macroeconomic policies and the structure of incentives, the expenditure policies of the state are important. First, the need to create employment should be taken into account when determining the composition of expenditure. Some activities are inherently more capital intensive than others--for example, air transport is more capital intensive than secondary education--and differences in employment intensity should be considered when allocating resources among various activities. It fortunately is the case that those government activities which promote human development--discussed in section 3--often are relatively labour intensive and hence there is no conflict between the need to choose an appropriate composition of public expenditure and the need to generate employment.

Second, in cases where government departments and state enterprises are insulated from market forces, steps should be taken to ensure that methods of production reflect the real cost of labour and are not excessively mechanised. This point applies in a great many fields where there is some choice among alternative techniques. Examples include construction, materials handling, transport and irrigation. Third, it is equally important that in countries which have a large public sector investment programme, the criteria used to evaluate public sector investment projects take account of the opportunity cost of labour and the desirability of generating employment. Especially in projects financed by foreign aid, capital may be underpriced and a strong implicit bias against labour may inadvertently be introduced. The same problem can arise when funds to finance state investment projects are raised through taxation and allocated through administrative and budgetary processes. The implicit price of capital may appear to the implementing agency or ministry to be low or even zero, since it is allocated a lump sum as part of a project budget, and hence a strong bias against the employment of labour may be introduced.

In a large majority of developing countries the future growth of employment will originate in the private sector and much of this will be concentrated in small scale enterprises, whether in the informal sector or not. Particular attention should therefore be paid to ensuring that small enterprises are able to exploit their potential to the maximum. The future of employment lies not with the private oligopolies and large state monopolies that dominate the formal sector but with household enterprises, small privately owned firms and cooperatives in the urban and rural areas. The microeconomic policies mentioned above clearly are relevant here. In addition, the government should do what it can to remove any administrative obstacles--licenses, regulations, discriminatory taxes, city ordinances, police harassment--which prevent the expansion of household enterprises and other small businesses. Beyond this, it may be possible to provide positive assistance in the form of improved access to credit (but not a preferential interest rate), training and short courses (for example in bookkeeping) and marketing (perhaps by helping to organize marketing cooperatives). A bit of imagination and experimentation will be required, but the payoff to a successful initiative could be high.

Even if all these measures were adopted, however, the employment situation might be so serious, and the role of employment in a human development strategy so important, that a major structural reform in the form of a guaranteed employment scheme should be considered. There is nothing particularly novel about this. Many governments have experimented with public works projects as a way to reduce seasonal or year round unemployment while creating useful assets. 59 In China the mobilization of surplus labour within the commune system -- disbanded after 1978 -- was a key mechanism for rural capital construction which succeeded in large part in transforming the countryside. 60 And in the state of Maharashtra, India a successful employment guarantee scheme has been in operation in the rural areas since 1975. 61 In Bangladesh there has been a large food-for-work programme for about three decades and in the second half of the 1980s it provided approximately 15 days of work per year for households which owned less than 0.5 acres of land. 62

The point of departure of an employment guarantee scheme is recognition of the right of everyone to work, i.e. to guaranteed employment for performing unskilled manual labour. This right applies to women as well as men, to those who reside in urban as well as rural areas, and to all able bodied persons age, say, 15 or above. A programme for guaranteed employment would thus be open to anyone who applied. It would in effect act as the residual source of employment and would place a safety net under those most in need. The wage rate would be the same for women and men, although some regional variation could be introduced if the price of wage goods differed substantially from one part of the country to another or between rural and urban areas.

The programme should concentrate on the construction of productive assets, including investments in infrastructure projects, that can be expected in future to contribute to sustained higher levels of output and income. That is, the guaranteed employment programme should not be regarded as a short term measure to provide relief in an emergency or during a period of readjustment and restructuring; nor should it be regarded as an income transfer or welfare programme. Rather it should be seen as a mechanism for mobilizing otherwise unused resources for capital formation and growth. It is thus a component of a long term human development strategy and the employment scheme should be closely integrated into the government's overall policy framework.

In areas where unemployment is chronic, particularly in some urban areas, it may be necessary to organize the scheme on a permanent, year round basis. In many rural areas, however, the problem of unemployment is largely seasonal and the employment guarantee scheme would have to be designed to mesh into the agricultural cycle and supplement normal earnings from crop cultivation. But whether seasonal or permanent, the programme should emphasize labour intensive, low cost infrastructure and other capital construction projects. Obvious candidates include small scale irrigation works, soil conservation, afforestation and orchard planting, housing and water supply, and construction of primary schools, primary and health stations rural roads. The creation of assets intended to raise future incomes should be a primary objective. (i) the wage rate

Any able bodied person seeking employment and willing to do manual work should be guaranteed a job at a subsistence wage. The daily wage rate or piece rate should be set at a level which does not attract workers from other jobs, since the purpose of the scheme is to provide work to those who have no other source of gainful employment. In the rural areas this means that the wage should be set marginally below, say 10 per cent below, the wages received by hired agricultural workers in the lowest income regions of the country. Such a wage probably would reflect fairly closely the opportunity cost of labour and would allay any fears that the guaranteed employment scheme would raise labour costs and damage employment prospects in other sectors of the economy.

Projects should be designed in such a way that labour costs account for a high proportion of the total. As a guideline one might aim for projects in which the costs of employment represent about two-thirds or more of total project costs. The experience of Maharashtra indicates this is realistic. If for any reason work cannot be provided to those who seek it, and the government is in fact unable to guarantee employment, unemployment compensation should be paid at a rate equivalent to the daily wage rate on works projects. This will provide an incentive to government to design and implement useful projects since, if not, the workers will be entitled to a wage payment in any case. (ii) financing the programme

In a macroeconomic sense the programme should be self-financing in the long run, since if the investment projects are selected carefully they should generate a stream of future benefits the discounted value of which exceeds the costs of construction. Nevertheless there may be a problem of public finance if the government is unable to cover its outlays through taxation of either the additional income generated or the additional wealth created. In the rural areas costs could be recovered through a tax on land and in the urban areas through a value added tax on transactions. In practice, however, it is unlikely to be possible to ensure that the ultimate beneficiaries bear the full tax burden. This programme, like virtually all government activities whether related to human development or not, will have to be financed out of general government revenues, national or provincial.

The difficulty of doing so should not be exaggerated. The large employment guarantee scheme in Maharashtra, for example, accounts for only eight per cent of total state expenditure. It thus comes down to a question of priorities. If the provision of employment to all is sufficiently important, it should be possible either to raise additional public revenue or reallocate public expenditure from low priority to high priority activities. (iii) distribution of the benefits

If a guaranteed employment programme is a success, a number of valuable assets will be created. These assets will raise both rural and urban land values and hence the incomes and general well being of those who own the land. Since the landholders will in virtually all countries be better off to begin with than the unskilled and otherwise unemployed labourers who work at subsistence wages to create the assets, the effect of the employment programme, paradoxically, could well be to increase inequalities in the distribution of income and wealth. This clearly would be an unintended consequence of the programme and specific remedies should be introduced to counter it.

The workers who participate would receive two types of benefits from an employment scheme. First, they would receive subsistence wages during the period when the work is undertaken. In Maharashtra more than three-quarters of the beneficiaries are landless workers or small farmer households; about 40 per cent of those employed are women. The subsistence wages in some cases account for between one-third and two-thirds of total household incomes. Thus the direct benefits accrue largely to the poor and are not negligible when seen from their perspective.

Second, to the extent that the assets created under the scheme generate a permanently higher demand for labour, the workers should enjoy in future some combination of higher market wage rates and more employment opportunities. Most of the benefits however are likely to accrue to landowners in the form of lower costs (e.g., if farm-to-market roads lower the cost of transport), higher yields (e.g., from irrigation projects), higher land rents and hence higher land prices. Moreover, the benefits to landowners are likely to vary more than proportionately with the size of holding. This would occur, for instance, if a project lowers transport costs and the marketable surplus increases with the size of farm.

One way to overcome the problem of an unequal distribution of benefits would be to impose a progressive land tax. A more imaginative way would be to transfer ownership of the assets created by unemployed workers to the workers themselves. This could readily be done by forming a multi-purpose cooperative or worker-managed enterprise, the shares in which would be proportional to the number of days worked on the construction projects. The cooperative would then be responsible for maintaining the assets, managing them and distributing profits among the shareholder-cooperators. Unemployed workers would thereby gradually be converted into assetholders, although they should be free to cash in their shares with the cooperative if they wished.

Not all assets created by the employment programme could be managed in this way, but considerable scope exists for simultaneously transforming idle labour into physical capital and workers into shareholders in cooperative enterprises. The impetus this would give to human development could be quite considerable. For example, a cooperative could be organized around an irrigation project and water sold to farmers; a timber, firewood or fruit cooperative could be organized around a tree planting project; a toll-charging company could be organized around a bridge building project; a fishing or duck cooperative could be based on an artificial pond, and so on. These individual cooperatives could then be grouped into a multi-purpose cooperative that would have overall responsibility for managing the collectively constructed assets. Such an approach could have an enormous impact, provided of course the projects are selected carefully. (iv) project selection

The projects selected for a guaranteed employment programme should ideally satisfy several criteria. First, we have already emphasized that they should be highly labour intensive since a primary purpose of the scheme is to act as the residual source of employment for those who otherwise would be unemployed and without an income. Second, again as previously emphasized, the projects should result in the creation of permanent assets which can raise future output and incomes. Third, whenever possible the projects should be located near the homes of the workers, be they residents of rural or urban areas. Fourth, insofar as possible, the projects should form the nucleus of a cooperative or workers' enterprise to ensure that the benefits of the project accrue largely to those who work on it. Alternatively, fifth, projects should be located in areas where local organizations of peasants and workers already exist which can take responsibility for the management and maintenance of the assets created. That is, the projects should be used to encourage participation of the otherwise unemployed and to empower them. 63 Finally, failing this, projects should be located in areas of the countryside or parts of cities where the distribution of land ownership is relatively equal.

If these criteria of selection are satisfied a guaranteed employment programme should be able to reduce unemployment substantially, alleviate poverty in the short run by providing a subsistence income to those most in need, contribute to a faster rate of growth by creating productive assets and, in the long run, improve the distribution of income and the ownership of productive wealth. In addition, an employment scheme can serve as a catalyst for institutional change, facilitating participation at the local level in economic affairs and empowering those who at present are the poorest and have the least amount of power. (v) administration

Overall financial administration obviously will have to be the responsibility of the central or a provincial government, from where the funds presumably will originate. Selection and implementation of projects however should occur at the local level. Those seeking employment should have designated places at which to register, near to where they live, and it would then be the responsibility of the local authorities to provide employment within, say, two weeks. If no work is offered within that period the local authorities would be required, as suggested above, to pay all registered workers an unemployment allowance. This would provide an incentive to the authorities to make certain that work is in fact offered, since in any event they would have to pay labour costs, and it would also make certain that the poor receive a subsistence income even if the authorities are unable to provide productive employment.

The projects themselves could be implemented by the local government, by peasant organizations, by trade unions or by other non-governmental organizations. The objective should be to involve as many institutions as possible in the identification and design of suitable employment creating projects and in the actual organization of the work. The only condition should be that those entrusted to implement the programme are technically competent to undertake labour intensive investment projects. A monitoring system should be introduced by the local authorities to prevent corruption and misuse of funds.

The guaranteed employment scheme could perhaps begin on a small scale as an experiment. Once experience is gained it should then be extended to the entire country. The results of such schemes in the two largest developing countries, China and India, indicate that they can contribute a great deal to human development. It is likely that many other countries can gain from the experience of China and India and apply the lessons to their own specific conditions.
 

The distribution of productive assets

Even if everyone seeking a job is able to obtain productive employment, the pace of human development is likely to be very slow if the distribution of income generating wealth is highly unequal. The reason for this is that the flow of income originates from the stock of natural, physical and human capital. Even if human capital is spread fairly evenly across the entire population, e.g., because of public provision of essential services such as education, health, sanitation and clean water, income as a whole will be distributed unequally, and human development frustrated, if the ownership of natural and physical capital is concentrated in a few hands. A household's income depends on the amount of capital it possesses (human, natural and physical) and on the rate of return earned on its capital. If the household owns neither natural nor physical capital, it will be forced to rely entirely on its income from human capital and experience throughout the developing world demonstrates that in most cases this is not enough to prevent large numbers of people falling into poverty.

The connection between poverty and the distribution of wealth is seen most vividly in the rural areas. The poorest of the poor almost always are members of households that are assetless. Indeed the absence of productive assets is a sure sign of poverty, often dire poverty. Landlessness usually is the primary cause of economic distress in the countryside, but landlessness typically goes hand-in-hand with an absence of other productive assets, namely, livestock, implements and other equipment, barns and other immovable assets. Particularly in Latin America and parts of Asia, land ownership tends to be highly concentrated and the large landowners tend to possess most of the other productive assets as well. It is hardly surprising that many of the great agrarian rebellions of the twentieth century have occurred in these two regions: Mexico and Bolivia, China and the Philippines.

In countries which have not had a sweeping agrarian reform poverty and income inequality are closely associated with inequality in the distribution of land. In Bangladesh, for instance, small farms (those up to 2.5 acres in size) account for 70 per cent of all farms, yet small farms account for only 29 per cent of the cultivated land. At the other extreme, farms over 10 acres in size account for only five per cent of all farms but 26 per cent of all land. 64 The distribution of land in the other countries of South Asia is broadly similar. In India in 1982, two-thirds of all farms were an acre or less, yet they accounted for only six per cent of the cultivated land; 11 per cent of the farms were larger than 15 acres and they accounted for 61 per cent of the land. 65 In Pakistan landownership is heavily concentrated: in 1976 farms larger than 50 acres accounted for 58 per cent of the land in the Sind province and about 82 per cent of the land in the Punjab, the richest province in the country. 66

Self-reliant human development in the rural areas can often be achieved most quickly through the careful design and implementation of a land reform. A redistribution of property rights in land and water, accompanied as necessary by technical assistance, credit to small farmers and improved marketing, is very likely to improve allocative efficiency67 as well as the distribution of income and there is no evidence that it reduces the long term rate of growth of agricultural output. Indeed the experience in Asia and elsewhere indicates that rapid agricultural growth does not necessarily lead to higher incomes of the very poor or to human development in general, and where growth has been translated into higher incomes of the poor, the incomes of the non-poor usually have risen relatively faster, with the result that inequality has increased. The reason that the benefits of growth do not reach the poor or fail to reach them in full measure is that landless labourers, tenant farmers and small peasant cultivators participate only marginally in the benefits of technical change, the source of long run agricultural growth. Their lack of productive assets means that at best they have only a weak link with growth generating mechanisms (e.g. investments in physical and natural capital) and because of their high dependence on the remuneration of labour for their livelihood, they do not share in the higher profits and rents that accompany growth.

In principle it is possible to imagine massive and sustained state intervention intended to compensate the poor for their lack of productive assets. Policies to redistribute income, provide health and education services, mother and child care programmes, and ensure adequate nutrition, etc. obviously have a role to play but they cannot be expected to carry the full burden of great inequality in the distribution of wealth. First, the costs would be too high, for in effect the state would be attempting to compensate for the inability of large segments of the rural population to provide for themselves. Second, as long as the distribution of productive assets is unequal, the distribution of primary incomes will be unequal and those who receive the larger portion of the primary incomes can be expected continuously to resist government efforts to redistribute part of their income to others via taxation. The rich resist, often successfully, the removal of handsome subsidies that favour them. They will resist even more strongly attempts to take away part of their income. Thus on political as well as economic grounds there is a case for a once-for-all redistribution of assets.

Land reform in particular is likely to accelerate human development because of its effects on agricultural production and the employment of labour. Output per hectare tends to be higher on small farms than on large because small farmers use their land more intensively (have a higher cropping ratio) and utilize more labour intensive techniques of production. A redistribution of large landholdings in favour of the landless and those with very small holdings will bring about not only a more equitable distribution of income but also a higher level of income and output. In addition, because small farmers economize on land and physical capital relative to labour, employment per hectare also should increase. Thus two of the major causes of rural poverty, namely landlessness and underemployment, can be largely overcome by a comprehensive land reform.

In some countries water reform is almost as important as land reform. State owned tube wells could be "privatized" by turning them over to groups of landless people who would then have the right to pump and sell irrigation water and the responsibility to maintain the wells and pumps. The beneficiaries could be organized into cooperatives similar to those recommended for the management of assets created under guaranteed employment schemes. One local non-governmental organization has experimented with such an arrangement in Bangladesh and its experience has demonstrated that the idea is viable. With a bit of imagination it might be possible to design similar arrangements for the management and control of surface waters, thereby redistributing the ownership of part of the natural stock of capital to those who possess few productive assets.

It is sometimes argued that because of high population densities in some countries, the amount of land potentially available for redistribution is so limited that a land reform is not worthwhile. This view is not correct: whenever landownership is highly unequal there are gains to be had from redistribution, although where land is in very short supply it might be argued that cooperative or communal forms of organization become relatively more attractive. Even where land-man ratios are exiguous, however, human development can be advanced in rural areas by allocating to each family at least enough land to erect a house, cultivate a kitchen garden of vegetables and fruit and raise small livestock (chickens, ducks, pigs, rabbits). This is the bare minimum to which everyone should be entitled.

Where economies of scale can be exploited, small farmers should be encouraged and assisted to develop institutional arrangements--mutual aid societies, specialized cooperatives, irrigation associations, communes, etc.--which enable them to pool resources and undertake activities at the most efficient scale of operation. As we have seen earlier in this section, the most valuable asset of the poor often is their own labour power and rural employment programmes can mobilize seasonally available labour to create new physical and natural capital--irrigation and drainage facilities, earth roads and dams, small bridges, and land improvements such as hillside terraces, orchards and fuel wood planting projects. In many cases the assets created could become the property, individual or collective, or those whose labour created them. This would have the advantage of providing a permanent source of income for the poor and an incentive to protect that source of income through regular repair and maintenance of the asset. Apart from pooling labour, small farmers can join together to purchase and operate agricultural machinery that is too costly for an individual farmer and requires a larger area for efficient operation than is typical of a small farm. Such machinery cooperatives can help to raise the productivity of labour by overcoming any diseconomies of small plots that may arise. Similarly, credit and marketing cooperatives may have a role to play in supplementing the services offered by the private sector.

Land reform has been most successful in countries such as Japan, China, South Korea and Taiwan province, China where the rural population is well organized and is able to implement the reform after it has been adopted as national policy. Thus the experience of land reform is consistent with a wider proposition that informs this analysis, namely, that human development proceeds most rapidly when the people concerned participate fully in the process. In the specific case of land reform, participation is critical for success since land reform is impossible unless it is accompanied by a change in the balance of political forces. Once land reform has been implemented, the beneficiaries--landless labourers, tenant cultivators and small owner-operators--should be encouraged to create new institutions which permit them to play an active role in economic affairs and in the political life of the community.

Among capitalist economies, Taiwan and South Korea are notable for putting priority on a redistribution of productive assets before the growth of output. In both cases a land reform was the key redistributive measure. In Taiwan the land reform occurred between 1949 and 1953. Nearly a quarter of the cultivated land was transferred to the poorest 48 per cent of the rural population and this redistribution was equivalent to about 13 per cent of Taiwan's gross national product at the time. 68

The land reform in Taiwan was comprehensive and led to the creation of an egalitarian small peasant farming system. The degree of local participation in the reform process was high: authority to implement the reforms was entrusted to each village and farmers were encouraged by the government to form associations to provide credit, storage, marketing and extension services. In addition, irrigation cooperatives were formed so that farmers could manage the irrigation systems that served them. As a result of all this, a considerable degree of economic democracy prevailed at the local level although the national regime was politically authoritarian. The land reform, however, shaped the subsequent development trajectory in Taiwan and helped create a relatively egalitarian society with a capacity for broadly based human development. One indication of this is the evolution of the Gini coefficient of the distribution of household income. In 1953 it was 0.56; by 1964 it had fallen dramatically to 0.32 and it then fell further to 0.28 in 1980. Inequality then began to increase slightly, but by 1987 the Gini coefficient still was only 0.30, i.e., the distribution of income still was one of the most egalitarian in the world. Land reform contributed to this process in two ways, viz., by removing major inequalities in the countryside and by increasing the opportunity cost of rural labour (and hence real wages) of those who migrated from the countryside to the cities. The effect of land reform on urban wages was one of the key factors which led to a dramatic rise in the share of wages in national income.

The land reform in South Korea during the late 1940s and early 1950s was similar to that in Taiwan. Prior to the reform, land ownership was highly concentrated but operational holdings were distributed fairly equally among tenants. Indeed about half of South Korea's farmers were tenants and were effectively assetless. Thanks to the reform, 80 per cent of all land under tenancy and virtually all land owned by absentee landlords was transferred to small peasant farmers. The beneficiaries, in turn, were subject to an ownership ceiling: they were not allowed to possess more than three hectares of land. 69 In both South Korea and Taiwan sweeping land reform was made easier politically by the weakening of the landlord class--either because the landowners were Japanese or because indigenous landowners collaborated with the Japanese colonial authorities. When the Japanese were expelled at the end of the Second World War the landowning class found itself vulnerable to attack and without defenders. There were thus particular historical circumstances that made structural reforms possible in these two cases. Obviously the situation was and is very different in other countries. This points not to the exceptional difficulty of introducing structural change and institutional reform but to the need to consider the uniqueness of each situation and to exploit whatever opportunities exist.

Unlike Taiwan, the distribution of income in South Korea began to worsen from the mid-1960s. The main reason for this was widening income differentials between rural and urban areas. In the immediate post-land reform period rural incomes were roughly the same as urban, but by 1970 they were only 67 per cent as high. 70 The government responded to growing rural-urban inequality with a series of measures--greater investment in rural areas, better terms of trade for agriculture, and the launching of a New Village Movement to raise household incomes and improve living conditions in the countryside.

But South Korea's problem of rising inequality between the cities and the countryside has occurred in many other countries that have experienced industrialisation and rapid growth. This common problem suggests that one should take a broad view of the redistribution of productive assets. Development policy often has concentrated on promoting industrial expansion, with the result that agriculture has been severely under-capitalised, the countryside has lacked basic physical infrastructure (roads, power, communications) and rural development has withered for lack of finance. The outcome often has been deprivation, poverty and an absence of human development in the countryside. If human development is to become a reality in the rural areas, governments must reverse their policies, stop the squeeze on agriculture and stem the flow of capital from the countryside. This will require a substantial redistribution of effort and of public investment in favour of rural areas. The social rates of return on such investments in natural, physical and human capital are likely to be high.

The land reform introduced in China, beginning in 1978, was similar in some respects to the earlier reforms in South Korea and Taiwan, although the initial conditions obviously were very different. In each case the final outcome was an egalitarian agrarian structure centred around small peasant holdings. In the case of China, however, the reforms of the late 1970s were not a once-for-all measure but merely the latest stage in a long sequence of reforms that began in the 1940s--a sequence that included the formation of mutual aid teams, small and then large production cooperatives and the commune system. The most recent reforms included a dismantling of the communal tenure system and reorganization of brigade and commune level enterprises. In the process of reform some of the advantages of the commune system were lost, e.g. the pooling of labour for investment, the provision of basic services and the maintenance and expansion of some capital assets such as irrigation facilities. On the other hand, the incentive system was improved, output became more diversified and small scale rural industry enjoyed exceptionally rapid growth. Both agricultural and industrial output increased and average household incomes in the rural areas rose dramatically.

The distribution of wealth that emerged after all these changes is remarkably equal. The Gini coefficient for the distribution of net worth in rural China was 0.31 in 1988, which surely represents one of the most equal distributions of wealth in the world. 71 The largest component of total assets, namely land, was distributed among households on an equal per capita or in a few cases on an equal per worker basis. Hence the distribution of land after the reform was extraordinarily uniform and this virtually guaranteed that the overall distribution of wealth would be egalitarian. In addition, unlike other countries, the distribution of fixed productive assets (livestock, farm and non-farm machinery and tools, buildings) actually has an equalizing effect on the distribution of total assets in China. One reason for this is the way specialized economic activities have developed in the rural areas. Some households have moved out of farming in order to concentrate on industrial and service activities. They may possess little or no land or farm capital, but they do own non-agricultural assets, including their own homes. The assets of these specialized households are comparable in value to the assets of farm households and hence marked differences in wealth between farming households and those engaged in non-farm occupations have failed to emerge. As long as this pattern persists rural China may be able to combine an equitable distribution of wealth and income with increased diversification of production and rapid growth of total output. This, in turn, would augur well for future human development.
 

Food security

Food security is of course essential for life and even where there is enough food to prevent death by starvation, individuals may not consume sufficient food to enable them to lead a full life or achieve their maximum productivity. In this sense they may be undernourished. There is abundant evidence that growth alone does not necessarily result in better nutrition. 72 As in developed countries, there is in developing countries many a slip between the cup and the lip, between a rise in aggregate average incomes and an improvement in the nutritional well being of the population. On the other hand, where a permanent improvement in nutrition does indeed occur, in the form of a higher weight-for-height ratio, there is evidence that this results in a higher productivity of labour and in higher wages. 73 Thus food security, and specifically adequate nutrition, is one of the ends of human development while also being a means to human development.

It does not follow from this however that food security need be an explicit objective of public policy. In principle, food security could occur as an indirect result of policies intended primarily to achieve other objectives such as faster growth or a more equal distribution of income. Indeed if the policies and programmes previously discussed are fully implemented--primary health care (including family planning), basic education (including health and nutrition education), guaranteed employment and a relatively equal distribution of income and wealth--then the problem of chronic food insecurity should be a relatively minor one. Growth within such a framework would be equitable; the poor would have sufficient purchasing power to enable them to acquire a nutritionally adequate diet and the problem of hunger, at worst, would be reduced to infrequent threats of famine.

Unfortunately, however, we live in a far from perfect world and governments may need to intervene to ensure that everyone has access to a minimum amount of food. The form of intervention can vary enormously--from food stamps in the United States, extensive food rationing in China, food supplementation schemes in Tamil Nadu, India to food-for-work programmes in Bangladesh--and the most appropriate form of intervention will depend on a number of factors, including the nature of the nutrition problem and the range of other related policies already in place. 74 Even assuming a fairly ambitious public food distribution programme, the cost to the government need not be very large, provided the government has adopted a human development strategy and at least some of the basic components of the strategy have been implemented.

One way of estimating the costs of famine prevention, following Amartya Sen, would be to calculate the proportion of national income required to "re-create" the income of potential famine victims so that they could purchase the amount of food necessary to prevent serious hunger. Sen argues that "the costs of such public action for famine prevention are typically rather modest even for poor countries, provided they make systematic and efficient arrangements in good time." 75 What these "systematic and efficient arrangements" might be will be discussed in a moment, but first let us continue with a consideration of costs. Sen assumes, as an upper limit, that 10 per cent of the population will be potential victims of famine and because they are the very poor, they would normally account for no more than three per cent of the country's gross national product. Their normal share of food consumption would be no greater than four or five per cent of the national total. "Thus the resources needed to re-create their entire income, or to resupply their entire normal food consumption, starting from zero, do not have to be very large." 76 Moreover, since famine victims seldom are completely wiped out but usually have some resources, the net resource requirement that will have to be provided by the government will be less than three per cent of GNP or five per cent of total food consumption. A famine prevention or food security policy thus is feasible if a sufficiently high priority is given to it.

The best way to ensure systematic and efficient arrangements for food security is to introduce a self-targeting food rationing scheme. 77 The purpose of the scheme would be to provide the entire population, in rural and urban areas, with guaranteed access to a minimum amount of a staple food (usually a foodgrain) at a price which even the poor can afford. Thus the scheme, first, would be permanent and not mounted hastily on an ad hoc basis during an emergency. Second, it would be universal in coverage and not restricted to particular groups identified in advance as needing public assistance. Third, the rationed staple distributed under the scheme would be sold at a subsidized price. The extent of the subsidy could vary from time to time depending on circumstances and indeed could fall to zero if there were no serious threat of hunger.

Fourth, the amount of the staple each individual would be entitled to purchase at a subsidized price would be determined in such a way that at the aggregate level the total of rationed supplies plus purchases obtained on the free market would cover the minimum needs for food. The quantities distributed under the scheme could vary from time to time, falling when the situation of the poor improves and increasing when the possibility of famine rises. Finally, the rationing system should in practice strongly discriminate in favour of the poor by distributing a staple food that normally is purchased predominantly by low income groups. In practice this means supplying "inferior" cereals, i.e., those with a negative income elasticity of demand, or staples with low quality grades not acceptable to buyers at the top end of the market. It is this last feature of the scheme which contains the self-targeting element. Although everyone, rich and poor alike, is entitled to a given quantity of the rationed staple at a given price, only the poor are likely to take advantage of the entitlement since it is only they who consume the staple in large quantities. Self-targeting, in other words, would help to limit the size of the programme and keep total costs low. Moreover, as growth proceeds and the incidence of poverty declines, sales of low quality products and staples with a negative income elasticity of demand will fall, the need for food rationing will diminish and the scheme will in effect become self-liquidating. Attention can then shift to less ambitious programmes designed to eliminate residual malnutrition in well defined and very small groups of the population.
 

Economic security

Throughout this essay two aspects of human development have guided the discussion, namely, human development as an end in itself and human development as a means for achieving increased well being. A third aspect of human development, also requires attention, namely, its role in preventing severe deprivation. The objective in this case is protection rather than promotion. That is, the objective is to sustain people's capabilities to lead a minimally acceptable life. 78

Individuals and families in all societies are vulnerable to unexpected events, e.g., the death of an income earner, unemployment, sudden physical disability, incapacitating illness, or destruction of a crop (by flood, drought, storm, pests) on which one's livelihood depends. People are also subject to long term disabilities such as congenital handicaps or the infirmities of old age. The question that arises in developing countries and elsewhere is whether public policy can play a useful role in mitigating the effects of catastrophe, and if so, given the other possible claims on public resources, whether it would be wise to allocate revenues to that purpose.

It should be said straight away that a central objective of policies to ensure economic security should be to create the conditions in which individuals and families can help themselves. The human development policies discussed earlier in this monograph do precisely that. If everyone is assured of basic health care services, food security, a job and an equitable share of the country's income and wealth, the conditions in which self-help can flourish will have been created. The residual needs for protection should be rather modest and the cost of meeting those needs rather low. Self-help combined with self-selection (in the cases of guaranteed employment and food rationing) should suffice for the great majority of people.

Some might argue that the residual economic insecurity can be left to the market for solution: no further public policy measures should be necessary. If the expected flow of people's income over time fails to coincide with the flow of desired or planned expenditure, the difference can be reconciled through the capital market. That is, people can save (or lend) their surplus funds when expenditure is less than total income and they can borrow funds during periods when income is less than desired expenditure. Thus actual expenditure can be brought into balance with "permanent" income even in periods when "permanent" and actual income differ. Similarly, the market can significantly reduce the consequences of catastrophe. Insurance markets, after all, are designed to cope with risk: life insurance is concerned with the risk of death, crop insurance with the risk of drought, and similarly with marine insurance, automobile insurance, earthquake insurance, unemployment insurance. Provided the credit and insurance markets function efficiently, the residual economic insecurity that remains after the basic human development policies have been put in place should be of little consequence.

The rub however is in the proviso. Capital markets, particularly in developing countries, work far from perfectly and it is the lowest income groups who are most vulnerable to unexpected events and who have least access to credit and insurance markets. This suggests that in the first instance measures to increase economic security should focus on the poor. That is, the poor face a case of market failure and consequently there are grounds for governments to act in effect as lender and insurer of last resort. Whether this can be done at an acceptable cost is an empirical matter to be determined in each country, but in principle the net transactions costs to government--the subsidy required out of general tax revenues--for providing residual economic security in instances of severe deprivation should be very modest indeed.

If the market cannot be relied upon to solve the problem of economic security, perhaps some might argue that the solution is economic growth. Severe deprivation has disappeared in developed countries, it might be claimed, because average incomes are high and people consequently are able to provide for themselves when unforeseen events occur. Public intervention, according to this view, is not necessary.

This view, however, is a caricature of reality. Growth in the rich countries certainly has helped reduce severe deprivation, but growth alone has not been sufficient. Public policy in the advanced capitalist and socialist countries alike, in the form of a wide range of welfare measures, has played a prominent role both in raising the well being of the weakest and poorest members of society and in protecting the accustomed living standards of the population from unexpected as well as predictable declines. The achievements of the welfare state should not be underestimated. 79 Even so, deprivation remains in many rich countries such as the United States in the form of high unemployment (especially among minority groups), poverty (especially among children), hunger and lack of access to medical assistance.

Thus relatively well functioning markets and decades of growth of per capita incomes have not eliminated the need for governments to assume some responsibility for economic security, even in very rich countries. Indeed public expenditure on the welfare state, i.e., spending on all cash benefits plus public expenditure on health care, varies from 11.9 per cent of gross domestic product in Japan to 28.3 per cent in the Netherlands. 80 The major benefits, which of course vary from one country to another, are medical and sickness benefits, unemployment compensation, old age pensions, family allowance and social assistance to the needy. These welfare programmes clearly have succeeded in reducing poverty and inequality significantly and they have thereby increased economic security. It is estimated, for example, that in Australia and the United States, two countries that spend relatively little on welfare programmes, the number of people living in poverty was reduced by 49.3 and 22.6 per cent, respectively, as a result of the tax and transfer system. In contrast, in Germany and Sweden, two high spending countries, the reduction in poverty was 76 and 82.2 per cent, respectively. 81 Similarly, in a ranking of eight countries by the degree of inequality post-tax-and-transfer, the United States ranked as the most unequal, followed by Australia, whereas Sweden ranked as the most equal, followed by Germany. 82

The ideal way to achieve economic security no doubt would be through a combination of growth and what has been called a "social security strategy".83 Growth would provide the resources while the "social security strategy" would ensure that some of these extra resources are used to reduce deprivation. The question this raises is whether developing countries can afford a social security strategy or whether, to put it slightly differently, there is a conflict between accelerating growth and providing social security. Several points can be made about this alleged conflict.

First, social security, or as we prefer to call it, economic security is an objective in itself. It is one of the ends of human development. Growth is merely one means--an important means to be sure--towards achieving that end. Second, many of the components of a "social security strategy" have already been discussed above and defended as efficient means of promoting growth. They are not in conflict with growth but are an integral part of a programme to accelerate growth. We refer to expenditures on primary health care, water and sanitation facilities, guaranteed employment (as an alternative to unemployment compensation), redistribution of assets (as an alternative to tax and transfer programmes designed to reduce poverty and inequality) and food security. These expenditures can be justified in a growth context because they have high rates of return, i.e., rates of return at least as high as investments in physical and natural capital.

These are essentially measures intended not to promote human development but to protect people from the vicissitudes of life. Under this category come expenditures on old age pensions, widows' benefits, assistance to the permanently handicapped and disabled. The only potentially costly item on this list is old age pensions. Assistance to the disabled will almost certainly be a minor item of expenditure, except in countries which have suffered high casualties in a war. Widows' benefits, too, will not be costly since young widows will have income from a job (under the guaranteed employment programme) plus some productive assets (from the asset redistribution programme); old widows should be entitled to a pension. Hence the residual need to help widows is likely to be small and will consist primarily of assistance to widows with young children.

Developing countries should seriously consider, perhaps in stages, financing and implementing a pension scheme in which all working people participate, not just urban residents employed in the formal sector, as is often the case. The age at which women and men become eligible for a pension and the monthly payment obviously will have to reflect the demographic, economic and fiscal circumstances of the country, but in principle it should be possible to provide a measure of economic security to the elderly even in very poor countries. Thirdly, if this is too ambitious in the first instance, then one can begin by providing a pension to the lowest income groups, rather than to the upper income groups as is almost always the case.

Fourth, public expenditure on a pension scheme can be justified in part on economic grounds. It is now widely recognized that one of the motives for having children is provision of economic security in old age. Children, once they become adult workers, are a form of insurance for their parents--not just insurance against the infirmities of old age, but insurance against many of the catastrophes that can occur without warning. Children, thus, are a substitute for economic security measures that protect the livelihood of people, and in particular they are a substitute for old age pensions. It follows from this that state funded pension schemes in developing countries are likely to result in a reduction in the demand for children, a decline in the fertility rate and a slower rate of growth of the population. This in turn is likely to lead to a faster rate of growth of income per head. Thus measures such as pensions that are intended to increase economic security, to protect people's incomes and capabilities, also contribute indirectly to promoting human development. Protection and promotion often go hand-in-hand.

Lastly, there is the empirical observation that in fact a number of developing countries have given priority to expenditures on economic security. Among the socialist developing countries there is China and Cuba and among the more market oriented countries there are the well known cases of Sri Lanka, Costa Rica and Kerala, India. In each of these five cases public expenditure has contributed to impressive gains in human development and in none has expenditure on human development resulted in unacceptably slow rates of growth. If there is a conflict between growth and economic security, it must be a very mild one. Probably no such conflict exists.

Some corroboration for these views is provided by a study of a less celebrated case, namely the state of Tamil Nadu, India. 84 In Tamil Nadu, as in India as a whole, protective measures for economic security are available to those who work in the formal sector, i.e., for the government or for large scale industry. The entitlements include old age pensions, maternity benefits, injury and invalidity benefits, survivor benefits and a number of health and sickness benefits. In most of India, however, comparable entitlements are not available to the poor, and specifically to those who work in the rural areas or in the informal urban sector. Thus expenditure on economic security is regressive, favouring those who are relatively better off.

Tamil Nadu is an exception. In addition to the standard programmes for formal sector workers, the state has a history of providing some economic security to the unorganized poor and in 1989 the various entitlement schemes were liberalised and expanded to provide a fairly comprehensive range of benefits. A distinctive feature of Tamil Nadu's programme is that only those identified as "destitute" are eligible to receive public support. The support for which they are eligible, however, includes an old age pension (starting at age 65), a pension for the destitute who are physically handicapped (starting at age 45), pensions for destitute widows (age 40) and deserted wives (age 30), maternity assistance for women from poor households, survivor benefits for women (or men) who have lost the primary breadwinner and a marriage grant for girls from poor households.

The benefits necessarily are meagre--the old age pension, for example, is Rs. 50 per month--and not all eligible recipients in fact receive their entitlements, but a good start has been made and economic security for the most vulnerable groups in the community undoubtedly has increased. The cost in 1989/90 was very modest, only 1.5 per cent of total state expenditure. Thus it certainly is feasible to provide a measure of state funded economic security to the poorest of the poor. This support, however, should not be seen as standing on its own but rather as supplementing support provided by the family and traditional local institutions and charities.

Based on the experience of Tamil Nadu, it has been estimated that it would be possible to provide a basic minimum package of protective economic security for India as a whole at a cost the country could readily afford. 85 Only those poor persons classified as destitute would be eligible for the five types of benefit included in the basic package, namely, an old age pension (from age 65), a pension for widows and deserted wives (from age 40), maternity assistance, disability payments (from age 20) and compensation for occupational injuries. This basic package for the very poor need cost no more than 0.3 per cent of gross national product or one per cent of the combined current revenues of the state and central governments.

Tamil Nadu, a major state of India with a population approaching 50 million, shows what can be done to increase the economic security of those on the margin of survival. What is feasible in Tamil Nadu (and Kerala) surely is feasible in the rest of India. And if economic security is a feasible human development objective in India, one of the poorest countries in the world, then it is equally feasible in all other developing countries. The time has come to put the prevention of severe deprivation, the protection of people's capabilities to lead a minimally acceptable life, high on the list of priorities for public action.