UNITED NATIONS	HS
	Commission on Human Settlements	Distr. GENERAL HS/C/14/2/Add.2 19 November 1992 ORIGINAL: ENGLISH
Fourteenth session Nairobi 26 April - 5 May 1993 Item 4 of the provisional agenda        Activities of the United Nations Centre for        Human Settlements (Habitat): progress report of        the Executive Director
The Relationship Between Underemployment and Unemployment and Shelter Provision
Report of the Executive Director

I. Overview of Main Issues

[Back to Table of Contents]

A. Rationale for linking shelter and employment

[Back to Table of Contents]

1. The Global Strategy for Shelter to the Year 2000 (GSS) holds that the fundamental purpose of a shelter strategy is to ensure that the full human and economic potential of the shelter sector is achieved. To do so means more than focusing narrowly on the shelter sector itself. Instead, strategies must be formulated with respect to how the performance and processes of the sector are connected to broad social and economic concerns. Formulating a strategy in the absence of information regarding the impacts of shelter-sector policies on the rest of the economy is no more valid than formulating broad economic strategy in ignorance of the effects of those policies on the shelter sector. It is in this light that the formulation of strategies for the shelter sector should consider the links between the shelter sector and the overall economy, and in particular its employment potential.

2. In the context of this report, it is significant that the GSS calls attention to an increasing realization around the world that shelter and development are mutually supportive and interdependent: that "people, habitat, and development are part of an indivisible whole" (UNCHS, 1987: 6). Thus, it is important to understand the trends that shape the shelter sector and the interdependencies linking it with its overall economic and social context.

3. Policy-makers should be aware that housing investment is productive and an important source of income and employment. Housing investment typically comprises from 2 to 8 per cent of GNP and 10 to 30 per cent of gross fixed capital formation. Each of these ratios has historically risen with economic development. In particular, as economic development proceeds, the fraction of household income that must be devoted to food expenditures drops sharply, and the first area in which households tend to increase their spending is on housing and related services.

4. As the GSS recognizes and hopes to utilize, this increase in demand is translated directly into a favourable investment climate for shelter, and the shelter sector is able to bid successfully for resources in the competitive arena of economic development. The resulting investment contri butes both directly and indirectly to increasing national income. Income and employment generated by shelter con struction are amplified by forward and backward links to other sectors of the economy. These links are not only at least as capable of generating income and employment gains as other types of capital investment, but frequently operate with considerably shorter time-lags than other investments between the inception of investment and the realization of its full effects. Thus, during periods of economic recession the shelter sector should be regarded as an attractive sector for stimulating economy recovery. Moreover, in the current economic climate, shelter sector investments are attractive because their typically low import requirements imply that incremental investments generate a higher domestic multiplier than do investments that are import-intensive.

5. The GSS embodies the rationale that is at the root of this report: if policies affecting the shelter sector are favourable, the sector contributes to economic development, and the gains of economic development are translated into sectoral improvements. If, however, the policies work against general economic development, the links will fail and both sectoral and overall objectives will suffer.

"It is a well established fact that the poor derive most of their incomes from labour, the only income-generating asset at their disposal. But since provision of shelter is itself an activity calling for substantial labour input, one wonders if it is not possible to promote both housing and employment goals simultaneously through a common strategy" (Sethuraman, 1991: 300).

B. The shelter sector and enabling shelter strategies

[Back to Table of Contents]

6. Experiences during the 1970s and 1980s convinced international agencies that direct action by governments or other large institutions to build houses, especially for low-income households, is not the answer to the housing- supply issues of most countries. Although the houses con structed represented a considerable financial and organi zational effort, they were very limited in numbers when compared with total demand. In addition, they were of high quality (a government cannot breach its own building regulations) and, thus, were expensive. In the 1980s, only a tiny percentage of urban residents could afford typical State-built houses without subsidy. Shortages led to rationing that tended to favour the politically powerful, i.e., civil servants, and those with middle incomes who could afford the rents. Thus, there was inequity, inefficiency in reaching target populations, and a redistribution of housing assets towards the relatively well-off.

7. Enabling shelter strategies imply a shift from a direct government construction effort, which is often only a fraction of the scale needed to fulfil a government's own targets let alone the need, to the encouragement of indi viduals, small-scale enterprises (SSEs), and large con tractors in their endeavours in shelter-sector construction. As the construction industry in developing countries is relatively undeveloped, and the housing-construction sub- sector is dominated by SSEs, a shift to enabling strategies will almost inevitably increase the role of the informal sector.

8. As will be shown later in this report, informal-sector operations, dominated by SSEs, tend to be more employment-intensive that those operated in the formal sector. In addition, there tends to be a larger proportion of unskilled workers as the materials and technology involve simpler tasks. Thus, more people will work to provide the same quantity of output. As the International Labour Organization (ILO) Convention No. 122 of 1964 encourages all governments to promote full employment, and as labour-intensive production can be as price-efficient as capital-intensive production, it is argued that the increase in employment itself is sufficient reason to increase investment in housing through the enabling process.

9. As is the case for informal employment, housing in the informal sector is a response to the shortage of effective housing supply within the formal, controlled sector. The informal sector in housing, like its employment counterpart, usually benefits from the lack of regulation that enables cheaper construction technologies to be used. However, it also suffers from its illegality in that formal finance cannot be raised for construction or improvement, and few services will be extended to informal housing areas while they remain outside the legal domain. As the sector constitutes a larger share of urban areas every year, the need to recognize the validity of its contribution and to normalize informal areas with respect to servicing, finance, taxation, and other urban duties and privileges becomes more urgent.

10. In many countries, it is as much the policies adopted in response to macro-economic trends as the trends themselves that have resulted in significant declines in shelter investment and have contributed to a worsening of housing and infrastructure conditions. Such policies include the maintenance of overvalued exchange rates, imposition of import restrictions and tariffs on shelter- sector inputs, public investment cutbacks and reallocations, and policies of directed credit toward tradable goods and supposedly "productive" industries. The impact of these policies reaches far beyond the impact on the shelter sector per se. Not only is the direct and indirect value of output of the sector lost, but distortions included in the rest of the economy impose a high economic cost in terms of produc tivity losses, reduced savings and capital formation, disruptions in labour markets and increased inflationary pressures (UNCHS, 1991).

C. Employment, underemployment, unemployment and poverty

[Back to Table of Contents]

11. Productive employment is that which contributes to the production of goods and services, for the market and outside the market, for consumption or fixed capital formation. The periods chosen for determining a person's employment can vary with particular circumstances. Under ILO Convention No. 122, Member States agree to pursue a policy designed to promote full, productive, and freely- chosen employment. This policy aims at ensuring that there is work for all who are available for and seeking work; that such work is as productive as possible; and that there is freedom of choice and the fullest possible opportunity to qualify with and use skills. In the spirit of Convention No. 122, it is not sufficient merely to promote economic growth and assume that employment will follow, as there is no automatic link between economic growth and the reduction of unemployment.

12. Underemployment is often defined as the underutili zation of labour. Gilbert and Gugler (1981) maintain that it takes three distinct forms. First, it is related to fluctuations in economic activities during the day (e.g., at markets), over the week or month (e.g., in recreational services), or seasonally. These fluctuations result in idleness for the self-employed and laying-off of wage earners. Secondly, underemployment refers to the state when workers are so numerous that at all times a substan tial number is less than fully employed. In this case, a reduction in the number of workers would not lead to an aggregate loss of output. The most numerous example of these in developing countries is street vendors, many of whom have a stable output to known customers and still only eke out a living. The third type is hidden under employment in which a solidarity group continues to employ people rather than discharging them even when there is insufficient work to keep them fully occupied.

13. A definition of unemployment is extremely prob lematic in developing countries. Some would limit the unemployed to those who are actively seeking employment (World Bank, 1980). Similarly, the ILO definition embraces those who are without work but are available for and are actively seeking work (Hussmanns and others, 1990; Mayer, 1991). Others would include those who are available for work but have become so discouraged that they have given up looking. Being unemployed usually signifies being provided for by someone else. The unemployed are not necessarily numbered among the poor, and poverty is not limited to the unemployed and under employed. Particularly in developing countries, very low wages for unskilled workers in the formal sector place increasingly large numbers of employed people among the poor.

14. The definition of poverty has inspired much discussion. As Szal (1977) points out, a standard of comparison is required to identify who are the poor. Although non-economic aspects of poverty are seldom disregarded, they are difficult to gauge. Thus primary reliance has been placed on economic measures. These fall basically into three categories:

• (a) Absolute measures of poverty, which usually specify an amount of money that an individual or a family must receive to obtain a minimum standard of living;
• (b) Relative measures of poverty which are based upon percentages, e.g., a percentage of the total population or of total income. Such measures imply that the problem of poverty cannot be solved: statistically, there will always be the specified percentage at the bottom of income distribution;
• (c) Absolute/relative measures of poverty which combine absolute measurement with a relative standard.

15. The non-economic aspects of poverty involve basic human needs such as self-fulfilment, participation in society, clean air and a healthy environment, security, and freedom of movement. Since these are more difficult to quantify, they are often mentioned and then disregarded. The use of income distribution in developing countries has merits but data are often unreliable. Some countries have used the level of officially set minimum wages as the standard of poverty. However, minimum wages should be looked upon as a policy instrument to reduce poverty rather than as the method of setting the poverty standard.

16. When considering support to the poorest groups, it is useful to make a distinction between the working poor, who with an enabling environment would be able to make a living, house themselves, and have some level of services for which they could pay, and those who, through old age, infirmity, disability, and other disadvantages, can only have a rewarding life through being helped directly by the State. The matters discussed in this report are not focused on their plight or on their needs as they must be provided for by the State as an act of humanity. Rather it is the former group that is discussed.

17. The working poor include groups of very different needs. A large number consists of households whose heads and other members work long hours at unremunerative jobs or in marginally profitable businesses. They have been joined, recently, by many workers in the lower grades of government and private formal-sector employment who have seen their incomes eroded in real terms. It is not uncommon for negotiations for minimum wages to fix a rate of only a fraction of what a household of six needs in order to eat basic food. Other groups may be condemned to poverty despite working as hard as they can. Female-headed households, in which the woman struggles to combine child-rearing and income-earning, constitute a significant proportion of households in many urban areas in developing countries.

18. Employment opportunities for the urban poor remain severely limited, particularly in the formal sector. In the case of women, the situation is often compounded by lack of easy access to credit, low educational levels and lack of marketable skills. This is coupled with the fact that government support to the informal sector tends to be geared toward artisan, technical and skill-oriented activities, such as motor mechanics and furniture-making, which are male-dominated areas.

19. The impasse which the poor workers seem to be in concerning the improvement of their income reflects on their demand for housing which depends on what house holds are able and willing to pay. In the past, provider- based housing projects have tended to be unsuccessful in reaching the poorest groups. Even in upgrading, which is cheaper per head than sites-and-services schemes, the poor have often been ignored or pushed out. This is partly due to the problems of recognizing and meeting the needs of tenants who are a majority of the poor in many cities. In addition, as affordability criteria have been unrealistically high, the poor have preferred to trade their benefits and leave. This upward filtering of housing (more prosperous households occupying housing intended for lower-income households) is seen by Strassmann (1977) as a failure in planning for housing.

20. The informal sector is particularly important in the large urban centres and accounts for half or more of the urban employment in many developing countries. In terms of output perhaps a third or more of what is produced in urban areas can be attributed to this sector though this ratio varies substantially between countries. It is this sector inwhich more and more poor and new job seekers are finding opportunities to earn an income and many of these income- earning activities originate in and/or continue to operate from residential premises. The sector's role in poverty alleviation in developing countries is now widely recognized (ILO, 1990: 1). The sector has made a substantial contribution to national economies. Besides providing employment and incomes to the poor it has been a major source of human resource development since it serves as a training ground for millions and enables them to acquire productive skills at low cost, without any public expenditure. Despite low incomes, many workers have mobilized considerable savings both in cash and in kind to develop their businesses. Since all the capital investment in this sector is financed through participants' own savings, there is no burden on the public sector. In many branches of economic activity the informal sector coexists with the formal sector and competes successfully; in other areas the two are complementary with the informal sector distributing goods and services produced in the formal sector or providing inputs and services through sub- contracting (ILO, 1990).

II. Human Settlements Development and Employment Generation

[Back to Table of Contents]

21. While housing in the past has been regarded mainly as a consumption good, shelter is increasingly acknowledged as a productive investment. The fact that investments in housing are recognized as capable of generating income, as well as influencing the productivity of the occupants at their work, has an important implication for development policies. It implies that housing is not only a goal of such policies, but even more important: it is a tool of development policy. This chapter will look at the linkages between housing and human settlements development and the economy at large. It will start by analysing the direct benefits of housing, and then discuss the forward and backward linkages of human settlements development (the backward linkages are discussed at length in section III.C). Finally it will take a closer look at the macro impacts of human settlements developments.

A. Direct benefits from shelter

[Back to Table of Contents]

22. There is a universal notion that improved shelter has far-reaching direct benefits for the population. Research efforts have been made to demonstrate claims that shelter improvements:

• (a) Raise productivity at work;
• (b) Reduce absenteeism from work;
• (c) Raise the level of health;
• (d) Increase the productivity of education;
• (e) Reduce the incidence of social deviation.

23. However, it has proved to be difficult to substantiate these claims, as the influence of improved shelter cannot be isolated from other factors influencing the welfare of people. It is, therefore, not surprising that no conclusive evidence has resulted from these efforts:

"...in no single case did better housing make matters worse for the rehoused populations studied. On the other hand, there are remarkably few cases where location in new housing generated unequivocally favourable results" (Burns and Grebler, 1976).

The fact that there is no final proof does not, however, mean that favourable results from housing improvements do not exist or are not important. Logically, the relations must exist, for how can people work if their health is weak owing to bad living conditions? The problem is that the impacts, especially on productivity, are hard to quantify in practical research.

24. Many slum dwellers live in an environment that totally lacks drainage, sanitation, suitable access, solid-waste removal and power supply. Their only service provision may be access to small quantities of grossly polluted water. In effect, this corresponds to what has been termed as a "zero baseline" service level (Cotton and Franceys (1991)). Any improvement in services in such settlements is likely to result in some benefit to the inhabitants. By improving the level of service in an incremental fashion, to give several relatively small benefits in health, safety, social well-being and convenience over a period of time, affordable improvements in the quality of life can be gained without the major, and unaffordable, step to conventional service standards.

25. In addition to the benefits that occupants accrue from improved housing, others receive benefits as well. Employers of people whose productive capacity rises or whose absenteeism decreases benefit. Governments of countries with some form of social security will also benefit as improved shelter can be expected to diminish fire hazards, reducing municipal outlays for fire brigades and payments in support of victims. As the health of the population improves, less needs to be spent on hospitals, medicine and social security payments. The negative effects of crime and deviant behaviour and the cost of combating them will also be lower than before.

B. Multiplier effects

[Back to Table of Contents]

26. Any investment in housing, has an effect on the national income of a country that goes beyond the direct investment itself. This is measured by a multiplier, in this case, an investment multiplier. It is defined as the ratio of the change in national income to the initial change in the sectoral investment. Thus, an initial investment of $1000 may cause an increase in the national economy of $2500, in which case, the multiplier would be 2.5. The additional $1500 is added through indirect increases, as consumption expenditure is generated as a result of the $1000 spent. In the case of housing, the builders will earn money that they will spend on food and other products produced in the country, they will buy raw materials for the building and hire transport to move them. The occupants of the house will buy furnishings and fittings, and pay for maintenance, all of which creates paid employment and the use of materials. Insofar as these activities produce chains of consumption within the country, so the multiplier effect will increase.

27. Some of the consumption induced by an investment will involve non-local produce. The transport operation will use imported vehicles, fuel and lubricants; the worker may buy an imported radio or hi-fi. At each round, there is likely to be some leakage as imported goods are bought or, less common in developing countries, money is saved. Thus, if an investment involves imported materials and the workers spend their income on imported goods, the multiplier would be close to one; if local materials were used and the workers consumed only local produce, leakage in that round would be lower and the multiplier would be higher. In this report, income multipliers are used to express the additions which investment in housing creates in developing countries; in other words, how much income is created per unit of expenditure on housing.

28. Grimes (1976) estimates, on evidence from Colombian research, that the income multiplier for low-cost housing construction is about 2: for every one unit of currency spent directly on house construction, an additional unit of currency is added to the economy.1 He further estimates that about seven additional jobs are created for every US$10,000 spent on housing construction. This is higher than for manufacturing and close to that of the economy as a whole. In the Republic of Korea, the income multiplier was also 2 but 14 jobs were created for every US$10,000. Similar results can be gained in India, Mexico and Pakistan. As much of the labour used in construction is low-income, unskilled or semi-skilled, much of the employment benefit affects the lower echelons in society.

29. Although low-cost house construction appears to perform quite well in comparison with multipliers in other sectors, there is more to it than that. It can be argued that if growth rates for the housing sector can be accelerated (as proposed in the GSS) under conditions of economic stagnation (which undoubtedly exist in many developing countries), the multipliers (especially from backward linkages) may play a more important developmental role than the notional multiplier of about two would suggest. First, the increases in housing and infrastructural investment are likely to trigger an increase in investment in building-materials manufacture (both in new plant and increasingly efficient operations as orders for materials increase in scale and regularity), transport and marketing. Secondly, the first and second round increases in individual incomes would largely be obtained by semi-skilled and unskilled labourers who have little propensity for buying imported goods.2

30. Increases in residential development are also known to have net multiplier effects through their inducement of an array of services to be established in the vicinity to provide the basic necessities of life to the new residents. The implications of economic activity in the home are covered in some detail in section II.C.

C. Forward linkages

[Back to Table of Contents]

31. Despite regulations in many countries to the contrary, it is not uncommon for people to use their home as a workplace. In pre-industrial societies, this was the norm. The beginnings of the industrial revolution in Europe can be traced to the move from home-based manufacture to factories. However, many processes still went on in the home and distinctive house types based on the needs of a home-based industry can still be found. Until recently these small shops and cottage industries were the dominant mode in the world, often part of sophisticated and complex production and distribution systems.

32. Fass's study of informal settlements in Port-au- Prince, Haiti, found that the use of dwellings for making, storing, and/or selling goods for the market was so universal that he decided to treat the dwelling unit as a piece of productive infrastructure rather than as a predominant part of household consumption.3 A house in the city allows low- income people to have some control over their lives; otherwise they are like a "proletariat dependent on the sale of their labour to others" (Peattie, 1979: 1019).

33. The design of dwellings to allow for economic activi ties has rarely been considered in official housing projects. It is, however, quite common for residents to alter their housing units to accommodate some form of home-based economic activity. The standards of hygiene and space which result from economic activity in the home might not be acceptable to officialdom but they are what the inhabitants can afford. When, and if, their earning power increases they might want to improve the standards. A major contributing factor to the growth of earning power, especially for the working poor, is found in the opportunities for generating income in their settlements. A key point is that settlements are not static, they tend to evolve and develop in response to the income-earning capacity of the residents. If this capacity is enhanced, residents will be able to undertake all types of improvements to their living conditions, including improvements to their houses.

34. These issues should be kept in mind when deve loping policies that include resettlement or similar activities. Low-income housing and employment opportunities within and around it tend to form an environment in which the poorest members of society can eke out some sort of living, and in which low-paid workers can supplement their formal incomes. It follows, therefore, that policies and programmes that seek to improve housing conditions through relocation to resettlement areas are likely to disrupt the economic activity patterns set up in the former environment. For the relocated households this disruption may be worse than any potential gains in housing and infrastructure standards, at least in the short and medium term.

35. In addition to direct impacts through employment in construction and backward linkages in employment in other sectors, additions to the housing stock create considerable forward linkages, not only in providing space to work but also in the number of jobs in such industries as textiles, furniture, and household fixtures. In addition, an array of service trades establish themselves around housing areas, many of which provide jobs for secondary-income earners whose contributions are increasingly important as formal- sector wages decline in real terms. It takes very little capital to start a shop in a room in the house. Many low-income households thus make their first foray into earning a living this way. Thus, the link between housing and employment in the informal sector is both close and symbiotic, in that economic activities enable housing improvements, and the latter improve employment prospects and productivity. This section examines the characteristics of the enterprises which occur in residential areas, especially those carried out within the home itself, and their contribution to household economies.

1. Frequency and distribution of home-based enterprises (HBEs)

36. It can be argued that, whenever housing production is being valued, its physical value is only one aspect. Housing provides an input to the production process, not only through the stream of benefits arising directly from being well-housed, but also from the opportunities housing gives for commercial activity, storage, small-scale manu facturing, service industries, and retailing. Thus, the argument follows, to value a house simply on its physical worth is only as logical as valuing a factory the same way.

37. Strassmann (1986) has estimated that between a tenth and a quarter of all dwellings in the cities of developing countries have an enterprise on the premises, usually one that the occupants prefer to unrelated outside work. His study of Lima, Peru, shows that their location is non-uniform; they tend to be concentrated in certain neighbourhoods. In the city as a whole, one in nine dwellings included an HBE; in unauthorized settlements they occurred in one in six dwellings, while in conventional residential areas, HBEs were found in only one in 16 dwellings. The characteristics of neighbourhoods determine where the greatest proportion of households have HBEs, what type they tend to be, and where they are most lucrative. The HBE core of the informal sector not only has ease of entry, small scale, labour intensity, and unregulated competitiveness; its strength lies in the ease of shifting labour, funds, equipment, materials, and space from making one product or service to another, from the market to the family and/or to dwelling expansion itself.

38. In areas where transport, proximity to formal-sector enterprises, availability of space, and other neighbourhood characteristics allow, enterprises with a city-wide clientele tend to be found. These will be light manufacturing (food, clothing and textiles) and a variety of services, including medical and dental clinics. To function here, the HBE has to be able to compete with nearby large or modern enterprises. Larger enterprises require space, good access to utilities, and many skilled workers among the resident population. All these factors point toward messy and noisy types of manufacturing that might not be as profitable or welcome elsewhere: making furniture, metal goods, and leather products (Strassmann, 1986).

39. Raj and Mitra (1990) found that each type of HBE is linked to particular income groups and each income group in turn enters specific kinds of HBEs. From their survey in Delhi, they note that movement of households from lower levels in the hierarchy to higher ones is not common. Many reasons were advanced for this, one being that each type of HBE has its own market conditions, requiring a given level of resources and mix of competitive conditions. Although, in theory, HBEs in manufacturing could expand, there are barriers to entry into higher levels of operation by the nature of their establishment in the informal sector and in residential areas. In addition, the wider market is more competitive, those engaged in HBEs tend to lack negotiation skills or working capital, or they do not have the production environment to achieve the desired results.

40. This case study revealed an average monthly household income of Rs.1475 (ranging from a low Rs.200 to a high Rs.5900). A worker engaged in HBEs earned, on average, Rs.777 per month. HBEs were responsible for 56 per cent of the total income. The more specialized retailing and services activities tend to be the most lucrative, followed by storage and hawking. Renting is the least profitable activity, both in total and in return per unit area, but it is the most common activity. It is more than twice as common as any other economic activity. The attraction of its meagre income probably has everything to do with its almost total lack of time commitment.

41. While the net income from HBEs may not be all that spectacular, the rate of return on total capital invested in HBEs is phenomenal. At the lowest rung of the investment ladder, the rate of return is as high as 20 to 50 times the investment made. Invariably no taxes are paid on the establishment or workers' income and no costing is made for work-hours devoted by family members.

42. Strassmann's study from Lima (1986), revealed that the least profitable HBEs were more likely to be operated by women. When this observation is added to the fact that female-headed households tend to have fewer members in the work-force, this implies that not only is the average income of women engaged in HBEs less than males, female- headed households have significantly lower-incomes than other households. Male-operated industries with a city-wide market in Lima were found to give the highest income, followed by retail trade, restaurants etc. with a city-wide market. Both are more profitable than making leather, wood, and metal products. Least profitable remain those female HBEs: laundries or medical services and light manufacturing for the local neighbourhood.

2. Disadvantages of HBEs

43. The housing conditions of households with HBEs are generally better than the overall conditions, indeed they can be the reason for and enabler of settlement upgrading.4 Without HBEs, both the income and the incentive for making improvements would be lacking, and housing and neighbourhood conditions throughout the city would be worse. The disadvantages of HBEs tend to be a result of the more general malaise in housing and in employment conditions rather than the result of working in the home per se. HBEs combine opportunities for housing problems and worker exploitation to arise in the same place. However, it would be unhelpful to point to the disadvantages as a reason for condemning HBEs, rather they should be dealt with as housing or employment issues within those general sector policies.

44. The first group of disadvantages arises particularly from the isolation in which the workers carry out their tasks. While factory workers can benefit from group solidarity in order to campaign for better working conditions, home-based workers are less able to improve their lot. The exploitation of workers inherent in the outwork system is now generally recognized. It has been said that

"it allows the manufacturer to pay wage rates which imply an intolerable level of exploitation of the worker, frequently lengthening the working day, and forcing the worker to work in conditions which present not only safety but health hazards to herself and other members of the family" (Young, 1981).

Its persistence and spread has been attributed to the large- scale existence of poverty and surplus labour in developing countries. The meagre earnings of outworkers are often crucial for family survival. The majority of outworkers is made up of women, who are particularly vulnerable to this form of labour exploitation.

45. The second group of disadvantages concerns the effect of economic activities on the residential environment. These are often referred to as externalities. An externality is an effect which a particular form of land use or activity has on neighbouring uses and occupants. A new golf course is likely to have positive externalities on house prices as people would like to live near a large, well tended open space frequented by the élite. In contrast, a chemical works will have some positive externalities (employment) and some negative (pollution). The plight of the people of Bhopal in India, requiring jobs but suffering terrible consequences of chemical pollution, expresses the two sides of externalities in a stark and gripping manner.

46. HBEs provide convenient services and employment for occupants of the neighbourhood. However, they may have considerable negative externalities that may be tolerable for some and intolerable for other residents. The fumes from fish-smoking, charcoal-burning, and other smoky or smelly enterprises may cause nuisance. Fire and boiling liquids can cause dangers beyond their immediate users, especially to children. Tinsmiths and other metalworkers, car repairers and engineering activities cause noise. All activity tends to increase the need for access by larger vehicles than those simply used for the occasional passenger journey.

47. There is a need to assess the dangers and nuisances caused by HBEs, not least in order to set some order of magnitude on the main reason why planning authorities resist them and keep them in uncertainty about their continued existence. Using the sort of techniques assembled by Tipple and Willis (1991), analysis of the costs and benefits of various activities could be carried out in order to recommend policies which would decrease the incidence of negative externalities while maintaining the viability of HBEs.

D. Backward linkages

[Back to Table of Contents]

48. The sources of income multipliers involved in the production of shelter are referred to as backward linkages, they take place before completion of the housing and infrastructure product. Backward linkages are thus measures of the demands created by one economic sector (in this case shelter) for the products of other sectors. They are mainly involved in the building operation and, before that, the production of materials, transport and other activities leading up to construction. Moavenzadeh (1987) shows how more than 50 per cent of total building cost in Kenya and Mexico is used for buying the outputs of other sectors. These inputs include all the materials and the means of transporting them to the site. In general, non-metallic minerals, metals and metal products, and wood and its products are the main intermediate inputs to the construction sector. The inputs from trade and other services are also significant. The construction sector's own inputs are mostly limited to labour costs.

49. The backward linkages from the habitat sector are acknowledged to be larger than in most other sectors.5 This is so even though the data collected from official sources do not include informal-sector activities which is especially strong in construction. It must be noted that different kinds of buildings imply different capital-labour ratios and different skill mixes. The consequences for local employment and use of resources thus differ as well. The beneficial effects of multipliers seem to be inversely related to the cost of housing. As low-income housing (especially that constructed by the informal sector) uses greater proportions of locally produced materials, higher labour to capital ratios, smaller amounts of imported machinery, vehicles, fuels and lubricants, and more unskilled and semi- skilled labour, the economy of the country benefits more from its construction than from higher-cost housing. Furthermore, since the informal sector generally uses fewer imported items than the formal sector, it is likely to maximize backward linkages. Table 1 shows how backward linkages and labour-intensities of construction depends on the choice of materials.

50. The indirect employment effects of the expansion of cement-intensive and steel-intensive houses will remain below those of an expansion achieved with the intensive use of indigenous materials, such as brick and wood. Moreover, in the latter case the benefits will accrue to the small-scale formal sector and especially to the informal sector, both of which tend to be labour-intensive. Just as direct-employment impacts are greatest from self-help projects, so are indirect effects. The exception to this may be large, middle-income countries where the formal sector imports a relatively small portion of its materials. In such countries, the indirect employment created by more expensive housing projects may be considerable.

[Back to Table of Contents]

51. Based on these arguments, it is not surprising that the common neglect of indirect-employment effects leads to an unrealistic and too pessimistic picture of the employment that may be created by shelter investments. Even more importantly, the extent to which investments lead to increased production and employment-generation depends on the way the money is spent. Public or private projects applying high standards and complex techniques seem to induce the least employment in most countries. Govern ments which wish to increase employment, especially among the poorer sections of the population by means of shelter investments, should thus invest in those that are directed to the poor.

52. The technology adopted may also have immense influence on the labour-intensity. In addition, policies on the actors who work in the shelter process can greatly affect the employment-intensity of construction and the effect of investments on the poorest workers in society. Self-help housing and upgrading schemes are particularly effective in using under-utilized labour in different development stages.

E. Macro-level impacts

[Back to Table of Contents]

53. World Bank estimates show that construction and infrastructure generally constitute around 50 per cent of gross fixed capital formation in developing countries (Moavenzadeh, 1987). A typical developing country will spend about 70 per cent of its annual budget on construction and infrastructure. Furthermore, about 40 per cent of international loan funds are spent on construction and infrastructure.

54. The macro-economic effects of shelter policies are often misunderstood by housing specialists and macro- economists alike. It is, therefore, useful to identify the macro-economic effects of various shelter policies. Examples of policies with harmful macro-economic effects may be said to be those involving heavy subsidies and direct government production of housing services. Good policies (from the macro-economic point of view) are those that improve the supply of the key inputs to housing (a skilled labour force, a sound and functioning regulatory framework, and the efficient supply of building materials, land and finance at prices people can afford) and ensure their efficient use.

55. The productivity of the shelter and infrastructure investments arise, at least in part, from being investment in an asset which yields a flow of services over time. To label such investment as "consumption", as has been quite common in the past, is incorrect. The fact that housing can be considered a social sector does not mean that it is "non- productive". Housing investment produces a flow of housing services which can, in turn, be considered an intermediate input into the production of other goods and services. Housing policies that are internally sound from an economic point of view will also be efficient from the macro-economic point of view. Conversely, sound macro- economic policies are a precondition for the health of the shelter sector.

56. In recent years, many governments have embarked on structural adjustment programmes in which housing policies have been a part. Typically, structural adjustment focuses more on overall targets for reducing absorption (including government expenditure) and less on how it is to be reduced. Governments can choose to cut elsewhere, but housing will often bear a large portion of the cost of structural adjustment.

57. This might seem to be in line with a key issue of structural adjustment policies, the expansion of the traded- goods sector (that is, goods which are exported and goods which substitute for imports). This requires, among other things, that relative prices for such goods, and for foreign exchange, reflect their real source cost. As housing is a non- traded good, and structural adjustment may seem to imply that expansion of the traded-goods sector must entail contraction in the non-traded sector (in this case, housing), housing seems to be the obvious choice for attack. However, if resources are not fully and efficiently employed (as is the case with labour in most, if not all, developing countries), there may be scope to expand the traded-goods sector without contracting the non-traded sector. Since labour is underutilized in all developing countries, employment-rich human settlement developments should be an integrated part of structural adjustment policies. In addition, and perhaps most importantly, housing and its associated infrastructure are a basic input for the production of any other goods, tradeable or not.

58. Shelter policies that include such distortions as subsidies have undesirable macro effects. Heavily subsidized housing finance has contributed to increasing general interest rates, fuelling inflation, and increasing budget deficits.6 Some housing programmes are consistent with structural adjustment (those that improve financial intermediation, for example), while others are not (simple housing allowances). In general, projects that improve the efficiency of input markets (in land, finance, materials, skills and labour) and move towards appropriate regulatory frameworks are consistent with structural adjustment. In fact, they may be required for successful adjustment.

1. Income

59. As the demand for housing stimulates the demand for labour in the construction and building-materials industries, its effect on income-production in the economy can be very marked. In micro-economic terms, housing is a significant component of household consumption and savings. In developing countries, expenditure on housing, on the average, accounts for between one seventh and one fifth of all consumer expenditures. In addition, for the majority of households, this investment is one of the primary objectives for savings. The house is regarded by many as the most important hedge against inflation, on which a family can protect what little wealth it has. The housing sector is thus a major factor in income both for citizens and for the countries (Chatterjee, 1981). Furthermore, as the construction industry is a significant source of employment for unskilled migrants to the urban setting, the effect of increasing that employment is generally redistributionary, helping the poorer households more than the richer.

60. Efforts to reduce construction costs can stimulate demand for low-income housing by enabling the poorest households to participate. What is perhaps more important is that by increasing the participation of the working poor in urban areas in the production of construction materials it is possible to improve the incomes of households and hence the effective demand for housing and related services. In addition, the positive effect exerted by provision of shelter space on residents' ability to get work, either by being close to employment opportunities or by having space in the house and a ready market for their services in the neighbourhood, cannot be underestimated, particularly for the second and third earner whose income may raise the household out of poverty.

2. Balance of payments

61. If expenditure is switched from one sector to another, the effect on the balance of payments reflects whether the shift is from a higher import-using sector to a lower, or vice- versa. In general, housing is quite a low user of imported materials. In the aggregate, only about 20 per cent of housing investment in developing countries is devoted to the purchase of imported materials. Further, there is considerable scope for encouraging the substitution of imported materials by local ones in many countries. This will not only reduce the direct import costs but also reduce the energy (usually a high foreign-exchange user) needed to produce the materials. In general it can be said that the cheaper the housing units are, the more likely they are to consume few imports. Thus, the production of housing, especially low-income units, does not present a problem for a country's balance of payments.

3. Prices

62. An increase in shelter investment will produce an increased demand for inputs. Insofar as these are available, and bottle-necks in supply are absent, the rise in prices will depend greatly on the elasticity of supply. Where there are many unemployed or underemployed unskilled workers, labour prices (wages) for unskilled construction jobs (the majority in low-technology building) are unlikely to rise very much. Increases in demand are thus unlikely to induce wage rises and so fuel inflation. Increased investment in more expensive housing, however, is likely to have a contrary effect. As skilled labour is more important in the more expensive housing, and its supply is usually scarce, an increase in high-cost house-building is likely to have inflationary effects (Klaassen and others, 1987).

63. Building materials are usually in short supply in developing countries, especially if they are imported or employ a complex technology in production. Increased demand induced by a rise in housing construction would thus drive prices up. However, the effect of this can beminimized by increasing the proportion of local, unsophisti cated materials and by encouraging larger numbers of smaller-scale producers, who would use less capital- intensive methods, more unskilled labour, and accrue lower transport costs and, thus, produce more cheaply than large, capital-intensive operations. If the increase in shelter-sector activity is at the expense of other sectors which use the same materials, prices will only rise or fall in response to different proportions of materials in the various sector demands. In economies where prices rise and fall there is likely to be little inflationary effect. If, however, prices are not free to fall, the increase in prices of the more demanded goods will fuel inflation. Again, more expensive housing will have greater inflationary tendencies than cheaper housing.

64. If housing demand is satisfied from credit rather than from switches from other sectors, inflation is likely to be higher than the above cases. However, credit is usually unavailable to poor people and they must finance shelter out of savings or forgone consumption in other sectors. Their activity is likely to be less inflationary than formal construction involving institutional credit and an increase in the money supply. In general, the inflationary effects of low-income shelter production are only very modest, while the effects from high-income shelter may be more serious. In either case, bottle-necks in the supply of building materials are likely to worsen the situation.

III. Scope and Modalities for Increasing Employment Opportunities in Poor Communities through Shelter, Infrastructure and Services Provision

[Back to Table of Contents]

65. The main theme of this report is that increasing employment and income-generating opportunities should not simply be an "optional extra", but a major consideration in the development of the most suitable approach to develop ment. This chapter, therefore, examines the potential for reducing costs and improving efficiency in the provision, maintenance and management of those urban services which, at the same time, increase employment opportunities.

66. In their ability to take part in housing and infra structure provision independently of, and sometimes despite harassment from public authorities, the low-income groups may be seen as a potential workforce, willing to contribute labour for no monetary reward, in the interests of some improvement in their living conditions. Critics of self-help efforts argue that it is wrong that the poor have to labour and contribute in order to enjoy the facilities which the better-off receive as of right. A natural question to ask should then be: in what circumstances is it reasonable to expect people to contribute free labour, and under what circumstances should they be paid a wage? These are issues which have occupied the ILO in its capacity as a regulator of working conditions on behalf of governments, employers and workers.

67. As suggested by the ILO, a clear distinction between "major works" and "minor works" is helpful in creating a workable approach to employment-intensive and sustainable development in the urban sector. UNCHS (Habitat) concurs with this view for the shelter sector. In the urban context, major works are those which are based on wage labour, whereas minor works are those within which a labour or cash contribution from client groups is acceptable (Lyby, 1992). While discussing these issues, two conventions on human rights should be mentioned: No. 29, The Forced Labour Convention, of 1930, and No. 105, Abolition of Forced Labour, of 1956. These were established in order to curtail abuse in cases in which people were forced to work for no payment in tasks for which they would normally be entitled to a wage. In Convention No. 105, the most important case of forced labour which concerns us here is that used "as a method of mobilising and using labour for purposes of economic development".7 The Convention sought to suppress this in any form and by 1992, 111 Member States of the ILO had ratified the Convention.

68. Some of the poorest countries tend to resort to "voluntary" unpaid labour owing to their lack of resources and as a means of increasing political commitment to the country. They argue that self-help work is a means of paying taxes through labour rather than cash, and if the work were not done this way, it would not be done at all. The alternatives of higher effective taxation, or the repercussions of not having the drains or streets or other penultimate improvements may create more hardship than contributing the labour.

69. In response, the ILO constituents have agreed that "minor communal works" should be exempted from the above definition as long as they are really both minor and communal; in other words, if they are of direct benefit of those who contribute their labour. Thus major works and minor works are defined as (Lyby, 1992):

• (a) Major works include roads, drains, sewerage, electricity etc., which belong to the public domain proper;
• (b) Minor works comprise local stormwater drainage in flood-prone areas, small-scale paving of access streets and footpaths, on-site sanitation, community buildings, or locally-based waste-disposal systems. All these are items of direct interest to the inhabitants and tend to recur as high-rated priorities in local surveys around urban areas in developing countries. The building of houses by self-help and mutual aid would fit into this definition of minor works if the people who build them also live in them as owners, or with a share in the equity equal to the value of their labour inputs.

70. It should be noted, however, that the distinction between major and minor works is only concerned with whether the workers should be paid rather than being expected to contribute labour free of charge as a community service. The major/minor works distinction does not govern who does the work and, therefore, receives the payment. The main argument of this report is that local, under- or unemployed people should be employed rather than non-local (or even foreign-paid) people. Thus, the wealth accruing from the employment would enrich the local area (i.e., have large local multiplier effects). Table 2 outlines the potential roles of various actors within the major/minor works distinction.

[Back to Table of Contents]

71. There has been a tendency for public works in developing countries to move the boundary between major and minor works in an upward direction; increasing the scale and "publicness" of the utilities provided. While European and North American home-owners are quite willing to control the environment within their own property boundaries, policy-makers in developing countries have accepted the need to allow individuals or groups to dig drains and lay pipes within the public land around their plots. In some squatter settlements, residents have taken the definition of minor works even further up the scale continuum by fitting settlement-wide water mains and electricity supplies (usually clandestinely connected to the city mains).

72. Major works have considerable potential for employment-creation if the latter is recognized as a priority in the selection of technology, contract procedure etc. These will be discussed in more detail below. It is noteworthy that employment creation through labour-based approaches to public works is self-targeting.

A. Employment potential in the housing process

[Back to Table of Contents]

1. Potential of the formal and informal sectors in the construction of new houses

73. There is a tendency to assume that formal housing should be built by formal-sector enterprises whereas this is not necessarily the case in reality, nor does it need to be an aim of policy. On the contrary, there is a strong argument in favour of the involvement of small-scale informal construction enterprises in the execution of housing projects. They use more unskilled labour, fewer imports and less hard currency.8

74. In countries where labour is abundant, increased construction activity is one sure way to increase employment. However, there are considerable differences in employment potential depending on whether investments are directed to housing low-income or high-income households. Low- income housing developments in the formal sector tend to be more labour-intensive than high-income housing. It has been shown in Kenya that the labour to materials ratio is 45:55 for low-income housing whereas it is 30:70 for high-income housing (Syagga, 1989). Further, it has been shown that small firms producing a single unit of housing generate most employment, but that their costs were 52 per cent above those of a medium-sized firm producing at a low volume. The medium-sized firms, however, have the lowest share of labour costs, use the fewest work-days per unit, and have the lowest costs for this type of work. They therefore create more jobs and units for a given expenditure than do large firms (Strassmann, 1985).

75. It should be noted that the above figures refer to "formal" housing, meaning that the units, built to the high standards specified in the building regulations, are served with running water supply, power etc., and thus include such facilities as kitchen, bathroom and toilet. This implies a significant share of skilled and supervisory, as against unskilled, labour, and use of relatively capital- and import- intensive methods and materials. A shift to self-help or low- income housing will help the informal sector with its greater labour-intensity, use of indigenous techniques, unskilled labour and small firms. Studies have shown that the cheaper the conventional housing, the greater is the proportion of total construction cost that goes to labour.9 Employment opportunities also vary with the type of public subsidy to housing. It has been shown that mortgage subsidies to lower-income groups tend to create more employment than subsidies for upper-income groups.10

76. It appears that, in the formal sector, lower-cost houses tend to consume more labour per unit produced than higher- cost houses. If the lower-cost house had been built by the informal sector, however, the pattern would be different because, even though materials are relatively inexpensive, housing in the informal sector appears to consume less labour per unit produced. This seeming anomaly is explained by the fact that the construction process in the informal sector is less demanding on skilled labour than in the formal sector.

77. It has been shown that in informal housing (basic minimum shelters without private services), only a fifth of the total construction expenditure may accrue to labour directly engaged on site (Sethuraman, 1985). At the same time, the housing costs are only a sixth of the formal-sector cost for the same floor space. The lower unit cost implies, however, that informal housing generates more employment per unit of expenditure than does formal-sector housing while producing six times as many dwelling units in the process.11

2. Sites-and-services schemes

78. The concept of sites-and-services is based on two suppositions; first, that it is cheaper (or at least that financing is more easily managed) to build houses gradually, and, secondly, that the house-owners will invest their own labour in the construction. Both these suppositions have lately come under attack.12 The second assumption is of direct interest to this report. It is founded on the idea that surplus labour, which was assumed to exist in project areas, could be transformed into productive employment leading to additional fixed capital.

79. Yet, experience has shown that there is not as much surplus labour in project sites as originally anticipated. Experience from El Salvador, the Philippines, and Zambia suggests that paid employment (in number of work-days) is about twice as common as unpaid. More than 80 per cent of the households concerned used hired labour.13 These findings thus cast serious doubts on one of the assumptions which underlie sites-and-services projects, that the opportunity cost of families' own labour is zero or near zero. Evaluation of a number of projects suggests that the beneficiaries not only lack "free" time, they also lack the necessary construction skills.14 The potential for paid employment in sites-and-services schemes is thus considerable.

3. Upgrading of existing housing

80. Because of its incremental nature and small scale, the renovation and extension of houses has traditionally attracted more small-scale contractors than new-built housing development. For a medium- to large-scale company, the overhead costs of doing small contracts are just too high for profitability. Small companies and individual tradesfolk, however, have thrived on the low- technology and low front-end cost environment in which renovations and extensions are carried out. Capital equipment can be limited to scaffolding, small-scale machinery, wheel-barrows and hand tools. Water and power are usually available on-site from the existing building, security of materials and equipment are ensured because the site is occupied. In general upgrading of existing housing is carried out, mainly by self-help or small-scale contractors, in a very labour-intensive fashion.

4. Maintenance of existing housing

81. The employment potential for maintenance traditio nally increases as the housing stock expands. However, the extent of the income multipliers is largely dependent on technology. Traditional technologies in developing countries, largely originating in subsistence agricultural communities, have minimized capital cost through accepting considerable maintenance commitments. Housing technolo gies inherited from industrialized countries have, however, minimized maintenance through high capital cost. If the materials are durable, the argument goes, lifetime cost is reduced because maintenance can be minimal. In a society where labour is scarce and expensive, this is a very cogent argument. In developing countries, however, where capital is very scarce and expensive, but labour (even in cities) is plentiful and relatively cheap, the argument for reducing maintenance by dint of increasing initial quality and cost (and, thereby, increasingly substituting capital for labour) loses some of its logical appeal. However, low-maintenance materials and technology have been adopted in many developing countries despite the scarcity of capital and abundance of labour. The technology used in such con structions are of such a nature that it is very difficult or impossible for local artisans to undertake major maintenance works. The potential for maintenance using local labour, and, therefore, the scale of income multipliers, is thus minimal.

82. There is a large stock of government-built housing, built on the high-capital cost, low-maintenance principle, which is approaching, or has passed, its design lifespan. Substantial portions of this is passing into private hands at highly discounted prices through the sale of government housing to its occupants. Thus, maintenance automatically passes from government direct labour to small-scale enterprises (SSEs). It may be appropriate to redirect maintenance of all government housing built in conventional materials to SSEs, based, initially, on current employees being encouraged to become independent contractors with some guarantees of initial contracts and some training in business management.

B. Employment potential in infrastructure and services delivery, maintenance and improvement

[Back to Table of Contents]

83. In the past, the provision of mains infrastructure to and within housing areas has been regarded as the task of central and local government whose land (roadways and service reserves) is used for their passage and who can have both the broad view of regional or city-wide infrastructural requirements and the resources to tackle large-scale and widely distributed capital and maintenance works.

84. However, as it becomes increasingly hard to keep the potable water flowing, dispose of sullage water, solid waste and stormwater, and keep the traffic moving, the need to involve communities, not only in crisis maintenance but also in planning and provision, is being widely recognized. This is not only born out of a movement towards democratization of decision-making, but also from sheer necessity as fiscal austerity freezes formerly routine local-authority tasks, especially in the context of unprecedented growth in demand for services.

85. Discussions on the potential advantages of alternative approaches to the provision of infrastructure have centred on their comparative advantages in terms of costs and improved management. In few of the debates has increasing the use of labour-intensive methods to expand employment opportunities been given as much consideration as questions of efficiency and cost-recovery. In fact, efficiency in provision and management is usually associated with reducing rather than increasing the labour force.

86. The obvious connection between reduction in the capital cost of the provision of urban services and the maintenance of these services can be taken a step further by looking at the potential for services to be extended, managed and maintained using labour-intensive methods at the neighbourhood level. To do this, it is important to think through the separation of each service into what can be undertaken by the public sector and what might be undertaken by the private or community sectors (UNCHS, 1989). In this discussion, there appears to be no automatic gain in employment to be made simply by transferring the functions of the local authority or government to a large private agency whose use of capital-intensive methods might be close to that of the original authority. The argument is more concerned with moving operations down the scale of enterprises to those which tend to use labour-intensive methods, switching parts of major works to SSEs and to community contracts.

1. Roads

87. The ILO has promoted labour-based technologies for roads since 1974, mostly in rural areas. As a result, the labour-based approach is now accepted for rural roads; there is no more argument over the fact that the labour-based approach is cheaper than equipment-based methods. The arguments about whether the works done are of sufficient quality are mostly settled because the stress of ILO policy has not so much been on encouraging labour-intensive methods in all circumstances, but on using the technology which is appropriate to achieve the quality required within a general presumption in favour of employing low-income people. As a rough generalization, it has been found that labour-based methods are cheaper than equipment-based when wages are US$3-4 per day, or below. A recent ILO- sponsored US$50 million programme of labour-based road building in Kenya produced an average of 6000 jobs over a 10-year period with 12,000 people employed at its peak.15

88. Compared with the size of the labour force in a country, labour-based road construction does not create large numbers of permanent jobs. Two of the most important constraints are the lack of small contractors familiar with labour-based techniques in infrastructure works (as opposed to house construction, production of sand-cement blocks etc. where many are active) and the lack of suitable and enabling practices in the government to encourage small contractors to become involved in infrastructure works. However, it does save money, allows what little money is available to be utilized in an efficient manner, and ensures that as large a proportion as possible of the capital spent on infrastructure reaches down into the lower-income groups. It is, thus, a poverty-alleviation measure rather than a work-generation programme.

89. The maintenance of urban services can also generate a variety of needs depending on the technology used in its provision and that used in the maintenance task. The concept of cost-in-use has been preferred by many municipal engineers to that of initial cost. In simple terms, the argument is that the choice of technology in road construction, like that in housing above, comes down to a balance between capital cost and maintenance cost over a notional economic lifetime.

2. Operation and maintenance of water supplies, drains and sewerage

90. Much of the above is equally applicable to water supply and sanitation (subject to the design and the capa bility of labour-based contractors). Major sewage works may require the use of heavy equipment (to lay the large main sewers leading to the outfall), but the smaller sections of work (smaller sewers leading to the main sewer) could be constructed using labour-based methods, depending on the designs adopted. This is especially relevant in digging trenches and backfilling where mechanical diggers would normally be used. There is thus a vast potential for individual or consortia of SSEs, and community contracts under the overall control of a major contractor (probably the municipal or regional authority) acting as an orchestrator of these smaller tasks rather than as executor. The same could be applied in the construction of water supply.

91. Communal services which cannot be individualized, such as drainage, need to come under the responsibility of a some grouping of households or a community development committee. They can be responsible for construction and maintenance and be paid for the work, or receive discounts on bills, or both, as payment. Within a neighbourhood, it can be argued that drains leading from small groups of houses, and those within the groups, would fall within the minor works category and qualify for unpaid, community labour. For trunk services, community contracts appear to have many advantages and few disadvantages not only for construction but also for operation and maintenance. Any engineering problems encountered are unlikely to be beyond the capacity of SSEs or community contractors except where land is steep or particularly liable to flooding. However, revenue collection should be done by a separate grouping as otherwise there will be seen to be a conflict between representing the community and policing it.

3. Community-based low-cost sanitation

92. In the past, labour-based, low-technology systems of sanitation have tended to mean the haulage of raw excrement away from houses to be tipped outside the city. The amount of employment this type of sanitation system creates is quite substantial, and it is self-selecting towards the poorest of the poor. However, conditions are so unamenable to significant improvement, that the handling of raw excrement is not to be encouraged as an employment alternative.

93. There are, however, labour-intensive and ecologi cally-sound systems which separate people and their excre ment until the latter is safe to handle. These composting systems, such as the "ventilated indirect pit" (VIP) latrine and its various regional versions, allow sanitation to be removed from a centralized system based on large invest ments in pipework and digestion plants. Instead the work can be done within the community with the simplest of local materials. Instead of a complex flushing and removal system, requiring a city-wide network and large (and foreign- exchange led) per capita investment in sophisticated plant, the excreta can be removed after a safe composting period by a person with a shovel and barrow to be used locally for compost.

4. Waste management

94. It has been estimated that city authorities in developing countries spend 30 to 50 per cent of their budgets on solid-waste management. Despite this, most do not manage more than keeping up with the backlog and few see general improvements in the environment without some external funds specifically allocated for cleaning up the city.

95. There are major resource and employment gains to be made from recycling refuse. While this can be organized on a large scale, it can equally well be done through individuals and SSEs, each specializing in a particular function or type of waste. Thus, waste-collection can be contracted out area by area; it can be dumped or deposited for sorting at sub-city or city level. Its sorting and pre-processing (washing, baling etc.) are ideally suited to SSEs as they are highly labour- intensive and involve little capital. Even the processing of, say, baled plastic into granules for sale to casting industries, or the casting of simple items, can be done by SSEs.

96. It should be kept in mind that there are social problems for people engaged in waste disposal. They are often regarded as inferior by the majority of the population, and are sometimes ostracized. In addition, some producers of wastes do not like to pay for the waste-collection, either because they have little spare money or because they believe that the system is corrupt. It is thus not difficult to understand why people prefer to dump, bury, or burn waste themselves.

5. Transport

97. The construction of transport systems might appear to be, of necessity, dominated by heavy engineering using large capital equipment. Yet, railways have frequently been built using mainly manual labour, particularly in the United Kingdom and the United States of America during the last century and the early part of this. Even urban systems may be amenable to labour-intensive methods as the Calcutta Metro System demonstrates. This must also take into account the cost increase due to prolongation of the time for completion of the programme.

98. Private and informal-sector participation in public transport, and in the transporting of goods is a feature of developing countries for many years. The dolmus of Istanbul; tro-tros in Ghana; and matatus in Kenya, represent the vibrancy of private transport. Although many of these modes of transport are illegal, both by the fact that they may ply for hire without licenses and that they are likely to be less than road-worthy under road traffic laws, they are a fact of life and provide the poor with a means of getting to work and carrying their goods and chattels which publicly-run transport systems alone fail to do.

99. Road safety continues to be a major concern, not only because of the generally overladen and badly maintained condition in which these vehicle operate, but also from their conduct on the road. As with other informal-sector transport operations, the quasi-legal status of matatus has led to problems over parking for picking up and setting down passengers. In addition, drivers and other workers must put in long working hours for fairly meagre wages, thus adding the problem of fatigue and increased human error to the passenger-risk.

100. There is a huge employment potential in smaller, human-powered vehicles in both urban and rural areas. The humble cycle, and its three-wheeled offspring the cycle- rickshaw, has the capability of transporting a variety of loads and gainfully employing unskilled people. In Dhaka alone, there are an estimated 200,000 licensed cycle-rickshaws,16 each of which will have two drivers working shifts. Thus, direct employment is provided for 400,000 in a city in which its workforce is unlikely to be much above 2 million adults. In addition, the manufacture and repair (not to mention the lavish decoration) of rickshaws and their parts is obviously a vigorous industrial concern carried out in many small workshops, often within residential plots.

C. Employment-generation potential through backward linkages: the building-materials industry

[Back to Table of Contents]

101. While backward linkages for construction in general usually represent a value which exceeds the value added by the construction sector itself, alternative production techniques have different multiplier effects depending on the scale of production and the origin of equipment and materials. From a socio-economic point of view, technologies associated with the largest multiplier effects should, all other things being equal, be favoured over those with limited effects.

102. While building materials cost developing countries between 3 and 5 per cent of their GDP per annum, they account for 5 to 8 per cent of the total value of imports. While there are important economic reasons for replacing imported materials with locally-made ones, many conventional building materials are made in processes which are highly sophisticated and consume large amounts of energy. Turn-key technology transfer, in which a European building-products factory is set up in Africa, has proved unhelpful for a number of reasons:

• (a) The enormous demand for energy;
• (b) The large potential output of high quality (and expensive) products for which demand is only developing slowly;
• (c) The tiny employment generated in the production process.

103. Small-scale, relatively labour-intensive building- materials manufacturing technologies are generally asso ciated with larger multiplier effects than large-scale, capital- intensive technologies for the following reasons:

• (a) Most, if not all, the equipment may be manufactured locally instead of being imported, thus generating additional employment and incomes;
• (b) Small-scale plants use local fuel while large-scale plants often rely on oil or electricity. If small-scale plants are adopted, the gathering and processing of fuelwood or other local materials will generate additional employment;17
• (c) The marketing and transport of materials produced by small-scale units generate more employment than those of the output of large-scale plants.

104. Most of the imported materials currently in wide use in developing countries can, in principle, be replaced by locally-produced, cheap materials. The problem is that knowledge about the production and use of innovative building materials which can be locally produced is limited to a few laboratories and research institutions. The gap in technology transfer, within a country and between countries, for wide-scale adoption of innovative materials is a result of several related factors (none of which are discussed in this report).

105. Production in small units close to markets both creates local employment and reduces transport costs. Furthermore, transporting building materials can be a real problem, because of their generally low value-to-weight ratio: in Botswana, Honduras and Sudan (as in many other countries) the cost of transporting cement 160 kilometres exceeds its value (Moavenzadeh (1987)). Thus, the economies of scale available from a centralized production system may be more than lost if markets are widespread and its transport infrastructure undeveloped.

106. In addition to transport advantages, locally-based small-scale production units can train labour cheaply and on the job, with women and youths having their opportunities. In addition, the capital investment per worker is only 1 per cent of that required for large-scale production (JUNIC, 1987). In Sri Lanka, the capital investment in the manual production of tiles was less than 0.6 per cent of that required to produce galvanized-iron sheets in a large-scale industrial process (UNIDO, 1985).

107. There are too many different building materials to deal with in a report like this. The lessons from low-cost brick-making can, however, serve as an example. A study from Colombia has shown that small-scale units in the brick- making industry, using relatively labour intensive techniques require from 10 to 20 times more labour than large-scale plants per unit of production (see table 3).

[Back to Table of Contents]

108. Other studies have revealed that the labour-capital ratio in small-scale brick-making units is over 100 times greater than that for large-scale manufacturing plants.18 In other words, for a fixed sum of capital, 100 times more jobs can be created in labour-intensive brick-making than in large-scale mechanical plants. Table 4 shows that, from a foreign-exchange and capital-investment point of view, small-scale brick-making technologies are much more appropriate than large-scale, automated plants. Capital investments in large-scale plants are 6 to 100 times larger than those for small-scale plants. Similarly, the import component of these investments is 5 to 15 times larger for the automated plants than for small-scale units.

[Back to Table of Contents]

D. Scope for public-sector support to labour-intensive work in shelter, infrastructure and services provision

[Back to Table of Contents]

109. The absorption of labour is not an end in itself. The increase of labour inputs to housing and shelter-based urban works programmes should be done on the understanding that they provide better productivity than other methods, when the multiplier effects have been calculated.

110. The choice of technology can have significant effects on employment both in construction and in maintenance of public services. The adoption of local materials and technologies, and the construction of simpler houses, creates more employment per dollar invested in construction than other approaches. As local authorities and national governments are major actors in the construction sector, their potential for influencing employment through switching from one technology to another could be considerable. Most government buildings could contain substantial labour-based components and could maximize maintenance rather than capital cost if there were a will to do so. Classrooms, clinics, administration and payment offices, depots, and other buildings can be built in labour-based materials both for their own sake and as examples to legitimize otherwise unaccepted technologies.

111. If employment-generation was positively valued, a change of technology could increase labour requirements for direct labour operations by local authorities without increasing cost overall. However, although direct public works activities provide considerable opportunities for increased income-generation, it can be argued that there are sufficient problems with them currently without adding the problems consequent on employing more workers. Local authorities in urban areas are usually so unequal to the task of keeping the services provided and maintained that added complications are unlikely to be welcome. In addition, many local authorities are such poor collectors of revenue that their direct activities are very small indeed and the potential of major proportional increases in employment is unlikely to provide much real employment. Thus, the chief means by which public authorities can support labour-intensive methods is likely to rest on their attitude towards small scale- enterprises.

112. UNCHS (Habitat) has in the past called for governments at the national level to support the role of SSEs in the construction sector (for buildings and services), reducing the advantages larger enterprises have through preferential import practices and creating a favourable regulatory, legal and research environment to boost their productivity (UNCHS, 1989). Similarly, local authorities have been called upon to create a conducive local environment for the growth of the SSE. Bye-laws and licensing requirements should support the use of local supplies of raw materials, their contracting and selection procedures should include SSEs using labour-intensive methods as acceptable bidders for public works programmes, and there should be a deliberate policy of using local contractors and locally-manufactured building materials for public buildings.

113. SSEs have a number of strengths which include:

• (a) They tend to use labour-intensive methods;
• (b) They have knowledge of local conditions, work within local neighbourhoods and can offer a service based on customers' special requirements;
• (c) They can develop from a very small scale, often in the home, and can give employment to local skilled, unskilled and unemployed labour;
• (d) They can use a variety of local materials and a minimum of imported inputs.

114. Table 5 summarizes the feasibility of transferring responsibilities traditionally held by local authorities to SSEs.

[Back to Table of Contents]

E. Potential for improving the legislative, administrative and planning environments

[Back to Table of Contents]

115. Recent UNCHS (Habitat), ILO and World Bank work has focused on the need for particular actors in the develop ment process to concentrate on the functions for which they are best suited. Central governments have been seen as best able to create a wider policy environment in which SSEs are encouraged; to withdraw regulations unhelpful to them, to labour-intensive works, and to employment in the home; to give such enterprises chances to carry out government- sponsored projects; to encourage research leading to greater efficiency; and to support the foundation of cooperatives and trade organizations which will strengthen the negotiating power of SSEs. At the same time, local authorities should ensure that the narrower, local policy environment is also supportive. This involves devising appropriate bye-laws and enabling building regulations; licensing and other requirements and practices which allow and support local raw materials, contracting by SSEs, and labour-intensive employment (UNCHS, 1989).

116. The GSS proposes that the first requirement for a healthy shelter sector should be a set of well-chosen policies at the national level within an effective macro-economic strategy, being implemented by the formal sector, government, non-governmental and community-based organizations, and the informal shelter sector - to increase production of housing and to improve existing housing. However, these policies must address the supply side of development. Demand for housing in developing countries tends to be much greater than supply; people are consuming little housing largely because not enough is available. The inputs of land, finance, labour, materials, and a regulatory framework should be the currency traded in order to increase the capacity of developing countries to improve shelter and employment at the same time.

117. The challenge to policy-makers is to maximize both the employment-creating potential of the informal sector and the degree of social protection and regulation extended to it. It can only be integrated into the economy as a whole if it is treated as part of an overall economic policy. A strategy which seeks to attain the two objectives of an increased integration of the informal sector into a regulated economy and an enhanced capacity of the sector to generate incomes and employment for larger numbers of people, needs to consist, therefore, of broad policy and institutional reforms in addition to programmes of direct assistance to the informal sector.

1. Assisting the informal sector

118. The disadvantaged position of the informal sector cannot be remedied by simply creating a less discriminatory policy environment or by throwing public money at it. It will also require a wide range of special measures to overcome certain inherent weaknesses of the informal sector itself, and to strengthen the productive capacity of informal-sector units. Without such measures, it is difficult to see how the informal sector can compete, or establish mutually beneficial complementary relationships, with the modern sector, and thus become more fully integrated into the mainstream of the economy.

119. In particular, it is important when designing policies to distinguish between, on the one hand, the relatively viable small enterprises in the informal sector which appear to have a potential for growth and employment creation, and, on the other hand, the precarious jobs performed on an individual basis, such as many street-vending activities, which clearly have no such potential.

120. In a recent ILO seminar to determine the most appro priate ways to promote the informal sector, the recommendations included the following (ILO, 1990):

• (a) To adapt the legal and institutional framework to the sector's characteristics taking care of its internal dynamism and its mode of self-organization;
• (b) To propose economic measures which will ensure the creation of a favourable environment for the development of small crafts by setting aside for them some markets including those currently supplied by imports;
• (c) To provide flexible technical and financial assistance focusing on appropriate technology development, product improvement and diversification, control over market networks and strengthening of local financing;
• (d) To establish a national structure aimed at discussing and coordinating employment promotion and to set up a national fund specifically geared to financing projects for the informal sector;
• (e) To decentralize, as much as possible, the assistance to this sector by transferring skills to local communities.

2. Legal, institutional and economic framework

121. In order to address the supply issues, the institutional structure for the delivery of urban services will need to be streamlined. Currently, institutional weaknesses are a serious constraint to the delivery of infrastructure. Too many ineffective agencies, sometimes with overlapping or competing interests, an inadequate framework for encouraging and supporting community participation, and a lack of motivation for efficient performance-oriented service delivery have been identified in the GSS as constraints to the supply of shelter services. Strategies suggested by the GSS to overcome these problems include the following:

• (a) Streamlining local-government institutions and strengthening their organizational capacity to deliver urban services;
• (b) Introducing reforms in organizational structures and mandates to encourage community involvement;
• (c) Increasing organizational efficiency, perhaps through the use of performance agreements that allow private enterprises to provide some services under governmental supervision of quantity and quality specifications.

122. The GSS further recommends a radical reappraisal of laws and regulations affecting the shelter sector, especially that which is currently classified as informal. Reforms of land legislation, planning and building regulations, property leases etc. should be undertaken by governments with their likely economic impact taken as a major determinant of their acceptability. In the short term, governments should examine their regulations and administrative practices with a view to amending, simplifying and streamlining those which present obstacles to growth and employment in the informal sector.

123. If one looks at building regulations, it was quite common in the past that governments used a top-down approach, imposing rules and regulations, based on a "blueprint" of how life should be, rather than a response to how life is. Such housing solutions not only reflect official interpretations of demand but can be a drain on the income of residents without giving them what they need. Yet, the reality is that most housing is provided by people outside the controlled environment of planning. The approach required in order to assist the efficient supply of low-income housing is not simply one of allowing people's participation in implementing some sites-and-services or upgrading scheme; it is providing support and resources to allow lower-income groups to undertake what they see as the priority in improving their housing and living conditions (UNCHS, 1985).

124. Inappropriate building codes have negative effects on the housing situation of the urban and rural poor. They have led to the low quality of building materials and construction techniques used in informal low-income settlements. Most existing codes favour import-based construction materials which, despite their high cost or scarcity, are often imposed as the only choice available to the low-income builder. Thus, attempts to introduce improved traditional building materials and to develop the employment-intensive backward linkages in the building-materials industry offered by residential development have been hampered (UNCHS, 1987).

125. It may be argued that the informal housing sector is a result of the building regulations in the past. What is needed for the future are regulations that can progressively extend measures of regulation to the informal sector. Only by doing so can the negative effects of the informal housing markets be alleviated. With this progressive spirit some standards could be encouraged as a first step towards regularization of the informal sector. Three types of such standards would appear to deserve priority attention:

• (a) Basic human rights such as freedom of association, freedom from forced labour, and freedom from discrimi nation;
• (b) The most exploitative types of employment relation ships in which many children find themselves trapped (often at a very early age) such as bonded labour; the employment of children in particularly hazardous occupations or industries; and the employment of very young children;
• (c) Occupational safety and health.

126. Meanwhile, it would seem counter-productive to inhibit economic activities within the home. In fact such activities should be encouraged. As noted above, regulations in operation in residential areas often prohibit the use of houses for income-generation. However, this report demonstrates that the separation of home and workplace is counter-productive for many of the world's poorest people who stubbornly refuse to behave in a manner congruent with town-planning law, preferring to increase productivity (often just to scrape a living) than to move the workplace out of sight of the living room.

F. Potential of community-based organizations

[Back to Table of Contents]

127. A basic principle of the GSS is the mobilization of the full potential of all actors in the shelter production and implementation process. Yet, the GSS spells out that the final decision on how to house themselves should be left to the people concerned. The introduction of enabling shelter strategies implies a change in roles and responsibilities among the actors involved. The roles and responsibilities of the various actors have been outlined in table 2 above. The potential for community action does, however, deserve an in- depth examination.

128. In its simplest form, the distinction between major and minor works regards the community as having no role in major works. However, the success of the community contracts programme in Sri Lanka demonstrates that there is potential for community organizations to bid for major works projects as well.

129. Community construction contracts are written agree ments between a government agency and a community to carry out the construction of simple public facilities in their area. The community is regarded as a partner in the development process. Before this can happen, the community should be represented by a legal entity, in Sri Lanka's case a community development council (CDC). It should be involved in the planning and design of the proposed facility, and must open a separate bank account for the funds involved in the contract. The National Housing Development Authority (NHDA) of Sri Lanka pays the community the cost of the construction (materials and labour) plus a small profit, more or less equal to the costing of a commercial contractor. It is then left to the community to decide how to organize the work. It has to buy the materials, but can ask NHDA for advice. The community can decide on the rates for its own skilled and unskilled labour and on the use of remaining funds (or profits) for other community activities. A technical officer of NHDA supervises the construction, checks its quality and gives technical advice. In some cases, the CDC management may be too weak and inexperienced to handle a fairly large contract. In that case, an NGO can assist the CDC in carrying out the contract and NHDA funds will be channelled through the NGO. In the period January 1986 to December 1988, a total of 83 community contracts was awarded to 70 different CDCs. The range of works included residential roads and drains, toilet blocks, water supply standposts, community buildings and repair and maintenance of buildings and infrastructure.

130. An additional way of looking at works which are suitable for community activity can use the "ultimate level" of service as the indicator of the end of public-sector responsibility and the beginning of autonomy within the community. In the full-servicing standard, where each plot is provided with a connection to publicly maintained infrastructure lines, this ultimate level is the plot boundary (or even within the dwelling; the place where the service is metered for charging). This is, undoubtedly, the ideal, which all would prefer if they could afford it. However, the reality of the late twentieth century is that the majority of citizens cannot afford it and most governments cannot cope with its provision. In the absence of the best, therefore, it is reasonable to argue that second-best solutions not only should be accepted, but should be welcomed as a way forward to gaining the major improvements in living conditions which can result from limited improvements or no servicing at all.

131. In the context of community responsibility for provision, maintenance and revenue collection, the ultimate level can be moved upwards to the boundary of the area occupied by the group whose members agree to act in concert. The area concerned may be a compound house, a house group formed around a single cul-de-sac or alley, or even a whole neighbourhood or settlement. Thus, the public authority could contract with a community to provide piped water to a point on the 100 mm main at which the supply is metered. From that point on, the community can provide the level of service which its members demand (i.e., need and can pay for), and can accrue the profits (or carry the losses) contingent on revenue collection. The public sector can, instead of paying for the service provision, simply subsidize the community's own provision with a grant to cover the cost of installation. For very poor communities, whatever subsidy option is chosen on social-welfare grounds can be applied at the ultimate level, by charging sub-economic prices for the water delivered to the penultimate meter, or the wastes collected from the dump.

132. Thus communities, represented by community-based organizations (CBOs), have a role to play in assisting in the implementation of local-authority strategies, especially as local-authority resources are insufficient for all the tasks they must fulfil (UNCHS, 1991). CBOs are ideally placed to undertake the day-to-day implementation of policies affecting their neighbourhoods, to act as intermediaries between individuals and the local authority, and with finance agencies (in which role they can act as honest brokers, or as the corporate entity with which proceedings for distraint may be joined), and to encourage groupings of artisans or entre preneurs who can negotiate successfully for infrastructure provision or contracts.

133. The advantages of the community-based approach to servicing in housing areas may be summarized as follows (Joshi, 1992):

• (a) It helps towards empowerment of the local govern ment structure;
• (b) It assists in creating communities;
• (c) It exerts pressure on government agencies;
• (d) It creates awareness in the servicing fields;
• (e) It gives new hope to residents of poor housing areas;
• (f) Each activity illuminates more on how the integration of community activities with the formal sector can be worked out.

134. There is also a role for communities to play in financing shelter improvements, both through organizing savings and through acting as an intermediary in the borrowing process. There is a long tradition of community savings schemes. The role of intermediary has also been important as an interface between individuals and formal financial institutions. The need for the intermediary arises from the presumption of bankers that low-income house holds with no collateral are a poor risk. If these low-income households are to break out of the need to use cash only or borrow from the usurious private money lenders, the brokerage of the community as an intermediary may be essential.

IV. Examples of Recent United Nations Efforts at Integrating Employment-Generation and Shelter Provision

[Back to Table of Contents]

135. A number of international agencies have already made efforts to address the issues outlined in this report. The four cases presented below are examples of programmes addressing the relationship between employment-generation and shelter delivery.

A. The UNCHS/Danida Training Programme for Community Participation in Improving Human Settlements

[Back to Table of Contents]

136. This Programme is aimed at helping governments to incorporate community participation into their national housing strategies for improving shelter conditions in poor urban communities. The Programme was initiated in 1984 and is currently in its third phase. Activities have been concentrated in three countries: Bolivia, Sri Lanka and Zambia where non-formal training of local-government staff and community residents has been developed as a tool to establish national strategies for community participation. The training approach consist of:

• (a) Offering workshops in organizing and facilitating community participation;
• (b) Testing methods and tools for use by motivators in community participation;
• (c) Introducing methods to monitor and evaluate training courses;
• (d) Providing on-site advisory services for countries requesting assistance;
• (e) Organizing regional seminars to collect, exchange and disseminate information on community participation.

137. The training materials developed under this Programme have led to interactive forms of learning-by-doing, cooperative partnerships between government agencies and community organizations, and to a recognition of the poor as active and competent agents in managing their settlements.

138. As outputs of the Programme, the communities' organization bases have been strengthened so that they can now negotiate with the authorities19 on all aspects of service provision. Furthermore, the Programme has led to a significant strengthening of their traditional forms of cooperation and self-help. Perhaps the outwardly most impressive outcome of the Programme can be seen in the community contracts procedures which have been introduced in Sri Lanka (see section III.F). These procedures provide useful examples for community involvement in public works and housing programmes in other countries.

B. The UNDP/ILO Employment Generation in Urban Works Programmes

[Back to Table of Contents]

139. These programmes address the creation of employ ment opportunities and improving living conditions for the poorer strata in urban areas in the least developed countries. Their immediate objectives can be summarized as the