Report No. 4148-KE P Kenya Economic Development and Urbanization Policy (In Two Volumes) Volume lI: Main Report December 13, 1985 Eastern and Southern Africa Projects Department Water Supply and Urban Development Division FOR OFFICIAL USE ONLY Document of the World Bank This report has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS Currency Unit = Kenya Shilling (KSh) 1/ KSh 1.00 = US$ 0.07 2/ US$ 1.00 = KSh 13.9 ABBREVIATIONS AND ACRONYMS CBS Central Bureau of Statistics CDC Commonwealth Development Corporation COL Commissioner of Lands DDC District Development Committee EABS East African Building Society GPT Graduated Personal Tax GRF General Rate Fund HABITAT United Nations Center for Human Settlements HDD Housing Development Department HFCK Housing Finance Corporation of Kenya HRDU Housing Research and Development Unit ILO International Labor Office iCC Joint Consultative Committee KIE Kenya Industrial Estates LGLA Local Government Loans Authority MFP Ministry of Finance and Planning MLG Ministry of Local Government MLS Ministry of Lands and Settlement MTC Ministry of Transport and Communications MWD Ministry of Water Development MWHPP Ministry of Works, Housing and Physical Planning NCC Nairobi City Council NCCo National Construction Corporation NHC National Housing Corporation NSSF National Social Security Fund PPD Physical Planning Department RIDC Rural Industrial Development Center SLK Savings and Loan Kenya Ltd. UNDP United Nations Development Program USAID United States Agency for International Development FISCAL YEAR Central Government and Local Authorities - July 1 to June 30 1/ The official unit of currency in Kenya is the Kenya shilling (KSh). However, in accordance with the practice of the Kenya Government, revenues and expenditures of the central Government and the local authorities are expressed in Kenya pounds (KE) at the rate of K£ = 20 KSh. 2/ The dollar-shilling conversion factor given here is based on the approximate exchange rate in effect in mid-1984. FOR OFFICIAL USE ONLY KENYA ECONOMIC DEVELOPMENT AND URBANIZATION POLICY Table of Contents Page O. INTRODUCTION ............................... o................... 1 I. ECONOMIC DEVELOPMENT AND THE URBAN SECTOR .................... 2 A. Population and Urbanization. . 3 B. Spatial Distribution of Production and Employment .6 C. Income Distribution in Urban Areas...... 8 D. Urban Development and Kenya's Macroeconomic Strategy .... 10 E. Summary .1.... ....... . 12 II. THE URBAN POLICY FRAMEWORK .................................. . 12 A. Urban Spatial Policy .............................. 13 B. Urban Employment Policy ................................ 16 C. Urban Investment Policy .............................. 18 D. Summary and Conclusion .............................. 25 III. URBAN DEVELOPMENT: THE INSTITUTIONAL FRAMEWORK ... 26 A. Existing Institutional Arrangements .... 28 B. The Investment Planning Process . . . 32 C. Agency Tasks . .. 37 D. Implementation ... . .. . ... 40 IV. LOCAL AUTHORITY SERVICES AND FINANCING ... 41 A. Overview of Local Authority Revenues and Expenditures ... 43 B. Recurrent Financing of Local Authorities . .... 48 C. Capital Financing of Local Authorities ... 56 V. HOUSING POLICY: INSTITUTIONS AND FINANCE .. . 59 A. Performance of the Housing Sector . .. 59 B. Urban Land Policy ... ... 66 C. Housing Finance and the Role of Public Sector Agencies .. 72 D. Conclusions . ...... 76 VI. SUMMARY OF RECOMMENDATIONS .... 78 The principal author of this report is Richard B. Westin. The initial research for the report was gathered by a mission consisting of R. B. Westin, B. A. Gouveia, F. Johnson, K. C. Sivaramakrishnan, P. Gidman, N. Jorgensen, and S. S. Yahya; this information was updated in the course of several subsequent missions in Kenya. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. STATISTICAL ANNEX Tables: Page 1.1: Population of Kenya, 1962-79 ................................... 84 1.2: Urban Centers with Population Exceeding 20,000 (1979) ..... ..... 85 1.3: Distribution of Towns and Their Population by Size, 1969-79 ......................................... 86 1.4: Urban Growth Projections to Year 2000 .......................... 87 1.5: Estimated Rural-Urban Income Distribution, 1976 ................ 88 1.6: Wage Employment and Earnings in 22 Towns, 1982 ................. 89 1.7: Percentage Distribution of Wage Employment by Town and Industry, 1982 .. ....... . .. . ........... .... .............. . . . . . . . 90 1.8: Growth Rates in Wage Employment and Population by Major Towns .....................**..**..***.**...**..t*..** 91 1.9: Nairobi: Household Income Distribution, 1974 ................... 92 2.1: Urban Investments in Central Government Accounts, 1980/81 ...... 93 2.2: Sewerage Coverage in Major Towns in Kenya, 1979 ................ 94 4.1: Expenditure of Locel Authorities, 1972-1983 .................... 95 4.2: Municipal Councils: Recurrent and Capital Expenditure on Main Services, 1975-83 .. ........................ 96 4.3: Municipal Councils: Economic Analysis of Expenditure, 1975-83 ......... . .. . . . .. . . . .. ................... .......... . 97 4.4: Town and County Councils: Economic Analysis of Expenditure, 1975-83 ................*...**. 98 4.5: Municipal Councils: Economic Analysis of Revenue, 1972-83 ...... 99 4.6: Town and County Councils: Economic Analysis of Revenue, 1975-83 ............... ...... 100 4.7: Recurrent Revenues and Expenditures of Larger Municipalities ... 101 4.8: Summary of Fund Balances, 1981 ...... ........................... 102 4.9: Source of Financing for Recurrent Expenditures ................. 103 5.1: Percentage Distribution of Urban Dwelling Units by Sector, 1979 ................ ........................................ 104 5.2: Percentage Distribution of Private Housing by Type and Tenancy Status, 1979 ............................... 105 5.3: Value of Residential Building Plans Approved by Nairobi City Council, 1974-82 ............................... 106 5.4: Rent Index for Nairobi, 1974-82, by Income Groups .............. 107 5.5: Sources of New Residential Plots ............................... 108 5.6: Annual Urban Land Requirements, 1981-85 ........................ 109 5.7: Local Authority Housing Stock, 1969-80 ......................... 110 5.8: Housing Units Completed by the National Housing Corporation, 1976-83 ................................ ......... 49* .... .....o..... 111i CHARTS 3.1: A Process for Developing a Program for Urban Services .......... 33 3.2: A Process for Developing Housing Programs ...................... 35 5.1: Alternative Land Development Models ..... ....................... 68 MAP Kenya: Urban Areas 1979 - IBRD No. 15860 PREFACE The original material for this report was gathered by a World Bank mission which visited Kenya in April 1981. This material was updated in the course of several subsequent missions which visited Kenya for other purposes, and a draft of the report was discussed with the Government in November 1983. The report was subsequently revised and updated to incorporate changes in the institutional framework for urban development (Chapter III) and the structure of local authority revenues (Chapter IV), and reflects the state of knowledge as of the end of 1984. 0. INTRODUCTION 0.01 Discussion of urbanization policy in Kenya requires more than simply counting the number of people one can expect over the next ten years and trying to identify the resources to provide a basic level of urban services. Kenya is rapidly changing from one of the less urbanized countries of the world to a country where the urban population and the urban economy play an important role in the development of the country. In the process, Kenya is developing a decentralized urban structure which can no longer be characterized as a single dominant capital in the midst of a mainly agricultural economy. Urbanization in Kenya, because it is beginning from a small base, is undergoing changes that are increasing rapidly in complexity. The need to provide suitable mechanisms to guide this multi-polar multiple-actor urban economy presents the Government with at least as many challenges as finding the resources to program for its growth. 0.02 This report draws as its theme the growing complexity of urbanization in Kenya and the need to develop an institutional and financial base that can channel this growth in productive ways. Kenya's policies toward urbanization have been articulately stated in a number of documents and are basically sound and compatible with Kenya's level of economic development. Moreover, as discussed in more detail in this report, these policies have generally been applied in a consistent fashion in the Government's choice of urban investments. In choosing as the focus for this report the supply mechanisms for urban growth, it is in full recognition of the Government's accomplishments in defining the directions for urban growth and initiating the steps to move growth in this fashion. 0.03 Chapter I of this report begins with an overview of the economic and demographic basis for urbanization in Kenya and the probable impact on urbanization patterns of current changes in Kenya's macroeconomic policies. Chapter II moves to urban development policies and reviews the record of what Kenya has sought to bring about in this field in the past ten years. On this basis, the next three chapters provide the substantive discussion of the institutional and financial framework for urbanization in Kenya, with Chapter III focusing on central-local government relationships and investment programming, Chapter IV covering local authority services and financing, and Chapter V discussing housing institutions and finance. Readers of this report will notice that relatively less attention has been paid to defining specific investment priorities and quantifying the magnitude of the resources required to provide them. These topics have been covered in depth in at least two major reports, and it was not felt to be necessary to repeat the recommendations of these studies except where they differ from more recently available data.1/ Instead, this report has built on previous studies to consider the implementing mechanisms that will be required if Kenya is to sustain and expand its accomplishments in the urban field as growth continues. 1/ See "Human Settlements in Kenya," Physical Planning Department, Ministry of Lands and Settlement, 1978; and "National Urban Development Strategy for Kenya: Rural-Urban Balance," by H. W. Richardson, A. Khanna, and 0. A. Sampaio, World Bank Consultant's Report, 1978. -2- I. ECONOMIC DEVELOPMENT AND THE URBAN SECTOR 1.01 In the first half of the 1980s, Kenya is facing serious constraints on its economic development. In the immediate period, the Government faces substantial budget deficits and a shortage of foreign exchange. In the medium-term, there is a need for structural adjustments in Kenya's economy to improve economic growth and restore balance of payments equilibrium. In the longer-term, Kenya's population growth rate, currently in excess of 4% per annum, constitutes the country's most serious obstacle to long-term improvements in living standards for its population.2/ The adjustment to these problems requires decisions so difficult and far-reaching as to dominate Kenya's economic development strategy for at least the next decade. Indeed, the touchstone for most sectors will be whether their development plans are consistent with efforts to solve these basic macroeconomic constraints. 1.02 Kenya's strategy to alleviate its economic constraints has several aspects. The first priority is revitalization of the agricultural sector, which is and will continue to be the main source of support for the majority of Kenya's population. This will require direct action in areas such as land policy, agricultural marketing, and investments to improve productivity, as well as more fundamental adjustments in pricing to improve the terms of trade in favor of agriculture. The second sector in need of reform is the industrial sector, which has developed behind high tariff barriers and other restrictions. For Kenya to expand its industrial base, it will have to move from a reliance on protected import-substitution industries toward export-oriented industries tied more closely to its agricultural resource base. Kenya's energy sector also requires rationalization, primarily through more comprehensive planning and careful economic scrutiny of proposed investments. Finally, Kenya's population policy must face the fact that present fertility rates, even if substantially curtailed over the next twenty years, will yield a doubling of Kenya's population by the turn of the century.3/ 1.03 It is within the context of these efforts that the present and future role of the urban sector in Kenya must be evaluated. Urban areas cannot demand a larger share of development resources solely on the basis of their poor population. Many of the urban poor, even if living in substandard and unhealthy conditions, are relatively better off than their counterparts in rural areas. Nor can urban areas demand more resources as prestige symbols of modern sector development. Economic restraint demands that investments in urban areas should show concrete benefits in terms of their contribution to the nation's development, and that urban areas should 2/ For a detailed discussion of Kenya's economic prospects, see "Growth and Structural Change in Kenya: A Basic Economic Report," World Bank, August 31, 1982. 3/ "Kenya: Population and Development," World Bank Country Study, July 1980. -3- tap their own resources to the greatest extent possible to pay for their needs. On the other hand, most economic activities in Kenya either take place in urban areas or rely on services and materials provided through urban areas. Moreover, even optimistic estimates of the population holding capacity of Kenya's agricultural sector indicate that urban areas will have to absorb a large share of the expected increase in the population. The responsibility for coping with these demands will devolve on an urban institutional and financial base which is currently ill-prepared to receive it. 1.04 For the fact is that Kenya is on the threshold of a substantial increase in urbanization. Kenya's urban growth rate is already one of the highest in the world, and population pressure on scarce agricultural land can be expected to keep rural-urban migration strong. At a minimum, Kenya's urban population and labor force can be expected to triple by the year 2000, and the share of the urban population will rise from more than 15% at present to between one-quarter and one-third of the total population. Kenya does not have the option of neglecting urbanization or hoping it will take care of itself. On the other hand, East Africa in general and Kenya in particular is one of the last regions of the world to enter the urban revolution. The mistakes of planning and problems of neglect which produced the chaos of Lagos and Kinshasa elsewhere in the continent have not yet occurred in the cities of Kenya. With foresight and realistic policies, urbanization in Kenya can be a positive force for economic development, if productive employment can be found for new urban residents and basic services provided for their needs. While problems of short-term economic management have tended to draw attention away from longer-term problems such as urbanization, urban development remains as one of the major issues which Kenya must face. 1.05 In this chapter, the economic and demographic pattern of urban development in Kenya is reviewed and put into the context of the current macroeconomic strategy. The first three sections provide the statistical background on urbanization in Kenya, focusing on trends in population, employment, and incomes. Section D considers the implications of Kenya's macroeconomic strategy for urban development, and section E pulls together a qualitative assessment of the economic outlook for urban areas over the next two decades. A. Population and Urbanization 1.06 Urbanization in Kenya has a long history among the Swahili culture of the coast. Centers such as Mombasa and Malindi played prominent roles in Indian Ocean medieval trading patterns and provided the primary points of contact for Europeans until late in the 19th century. The interior of Kenya, however, did not develop urban centers prior to colonization, and settlement patterns consisted of pastoral tribes and separated farms with small centers established as trading points. Because of this, Kenya is one of the less-urbanized countries of the world, although not atypically so for East Africa. The urban pattern which exists today predominantly reflects the development of British colonization and trade, rather than traditional African population and agricultural patterns. Nairobi, for example, is located in an environmental transition -4- zone which was not densely populated prior to colonization, while western Kenya, the most densely populated area of the country, did not develop any major urban center during the colonial period. 1.07 Table 1.1 presents information on the growth of Kenya's population since shortly before independence. The overall rate of population growth in Kenya has been increasing since 1962. This has been matched by an even faster acceleration of urban growth, reaching 7.9% over the decade 1969-79. This has caused the urban population of Kenya to more than double in ten years, and the share of the urban population to increase from 9% at independence to 15% in 1979. 1.08 Several factors must be accounted for in obtaining a reliable estimate of urban growth. One factor is an increase in the number of small towns with a population large enough to be classified as urban. Between 1969 and 1979, the number of urban centers expanded from 47 to 90 on this basis. If the growth rate is computed solely on the basis of towns which were urban centers in both 1969 and 1979, the urban growth rate is 6.9%, which is close to the value of 7% used in most projections by the Government of Kenya.4/ Another factor has been an expansion of municipal boundaries to include existing rural settlement areas. It is common for squatter settlements to develop on the fringes of municipal boundaries, often by the informal subdivision of agricultural land. Because of demands from these inhabitants to be brought within municipal boundaries and provided with services, most towns in Kenya had large expansions in their boundaries during the 1970s, with corresponding inflation of their apparent growth rates. 1.09 Table 1.2 presents data on the population and area of urban centers with more than 20,000 inhabitants in 1979. Several of these towns, notably Kisumu, Machakos, and Meru, had exceptionally large expansions in their boundaries during the 1970s, with corresponding inflation of their apparent growth rates. In order to provide a more accurate picture of urban growth, Columns VI-VIII of Table 1.2 provide information on relative growth rates where it has been possible to adjust for changes in municipal boundaries. From these data, we see that between 1962 and 1969, Nairobi and Mombasa experienced the fastest urban growth rates (with the exception of Malindi), while the other towns grew at rates at most slightly above the overall population growth rate. In part, this reflected the greater ease of movement of the African population after independence, which initially focused migration into the larger towns. It also reflected the new Government's employment policies which expanded African representation in government and industry and encouraged the growth of import-substitution industries in Nairobi. On the other hand, the secondary towns, many of which were linked to colonial agriculture, did not experience much of a population increase in the 1962-69 period. In fact, outmigration of settlers was enough to cause a decline in the population of Eldoret. 4/ Of the 47 urban centers listed in 1969, five do not appear on the provisional list of urban centers for 1979, even after accounting for known name changes. The indicated growth rate of 6.9% is based on 42 towns which are common to both lists. -5- 1.10 During the 1970s, these urban growth patterns were reversed. While the growth rates of Nairobi and Mombasa fell, the growth rates of the secondary towns increased to very high levels.5/ In Table 1.2, the average growth rate of the urban population in the secondary towns (excluding Nairobi and Mombasa and the towns for which growth rates could not be computed) was 7.4%, a rate which will double their population in ten years if it is sustained. This expansion of secondary towns is brought out clearly in Table 1.3, which presents the size distribution of urban centers in 1969 and 1979. We see that the proportion of the population living in towns exceeding 100,000 in population, even with the addition of Kisumu to this category, fell from 70% to 57% over the intercensal period. This was off-set by a large expansion in the population share of the second category of towns (20,000 to 100,000 population) from 7% to 25%. 1.11 The 1979 census figures show that urban growth in Kenya is not highly polarized towards Nairobi and Mombasa. Nairobi's growth rate of 5% during the 1969-79 period, while relatively high by international standards, has been moderate in comparison to the growth of the secondary towns. This result is in striking contrast to experience elsewhere in Africa, where urban growth has often been concentrated heavily on the capital city. The 1979 census data do not provide enough information to establish why this pattern has occurred. However, what appears to have been important in influencing urban growth in Kenya during the period 1969-79 is the strong rural-urban linkages of the economy, which have focused migration more into the secondary towns. It is also likely that the relatively neutral policies which Kenya has followed toward the rural-urban income distribution, which are discussed in more detail in the next section, have been important in preventing a shift in migration toward the largest cities. 1.12 Demographic studies do not foresee any significant slowing of Kenya's overall rate of population growth in the near future.6/ Hence, the outlook for urban areas will depend critically on the abiTity of the agricultural sector to absorb a large increase in population, which will require major investments to expand arable land and difficult political 5/ The population figure for Nairobi in the provisional 1979 census is significantly below most estimates prepared prior to the census. To some extent, this reflects the dispersal of Nairobi's growth into peri-urban centers such as Ngong, Kikuyu, and Thika. Other possible sources of error in the census figures (such as undercounting in squatter areas) could not be investigated with the data on hand. 6/ "Kenya: Population and Development," op. cit., pp. 37-40. -6- decisions concerning land redistribution.7/ Even if these policies are successful, urban areas can expect to absorb significant population growth over the next two decades. Table 1.4 compares two sets of urban growth projections, one based on the 1969 census and the other on the 1979 census. Drawing on the 1969 census projections, the Government's Physical Planning Department has predicted that Kenya's urban population in the year 2000 will be in the approximate range of 8.6 million, representing 28% of a total population of 31 million, with an average urban growth rate of 6.5% p.a.8/ As shown in Table 1.4, this projection of the overall urban population is supported by the 1979 census, but the expected spatial composition of the urban population is quite different when the 1979 data are used. For example, based on the 1964-69 intercensal growth rate, Nairobi was expected to grow to a population in the range of 3.5 million by the year 2000, representing 41% of the urban population. The 1969-79 intercensal growth rate indicates a more reasonable estimate for Nairobi in the range of 2.3 million, with an urban population share of 27%. This does not negate the seriousness of the problem of handling urban growth in Nairobi, as even the lower estimate predicts that Nairobi's population will triple in the next two decades. The revised estimates do indicate, however, that it is necessary to focus additional attention on the growth patterns of secondary towns. B. Spatial Distribution of Production and Employment 1.13 Urban areas in Kenya contribute to gross domestic product in an amount which is disproportionate to their population. This by itself is not surprising, as except for agriculture, mining, and certain other activities which are bound to the land, many producers prefer to be close to their markets and to sources of inputs such as labor and finance. Still, it is interesting to compare the relative productivity of rural and urban areas in order to assess the role of urban areas in an economic strategy to raise output. 1.14 Direct comparisons of productivity based on the location of economic activities are difficult to make. Instead, the approach used here is to examine the distribution of income between urban and rural areas. While this approach raises its own methodological problems (such as the treatment of urban-rural remittances), it probably gives a more accurate representation of relative urban-rural productivity and is less likely to show a "pro-urban" bias. Comparisons of the distribution of income between urban and rural areas for 1976 are shown in Table 1.5. From these data, we see that urban areas contributed a share of national product almost three times greater than their share of population. The main reason for this difference is the high average earnings in the urban modern private sector 7/ "Growth and Structural Change in Kenya: A Basic Economic Report," op. cit., pp. 50-59. 8/ "Human Settlements in Kenya," Physical Planning Department, Ministry of Lands and Settlement, 1978. -7- as compared to the low average earnings in the agricultural sector from which the majority of Kenyans earn their living. 1.15 Productive activities are not distributed evenly across the urban areas in Kenya. Table 1.6 presents information on the distribution of urban wage employment and earnings in 22 towns covered by employment surveys of the Central Bureau of Statistics.9/ The most striking feature of this table is the dominance of Nairobi in modern sector activities. Nairobi accounted for 56% of urban wage employment and 66% of wage earnings in 1982, against a population share of 42% for the urban areas covered by the survey. Indeed, only six urban areas had wage employment of over 10,000 persons in 1982 (Nairobi, Mombasa, Nakuru, Kisumu, Thika, and Eldoret), and these six towns accounted for 87% of total urban wage employment and 91% of wage earnings. 1.16 The industrial break-down of wage employment in 1982 for the same 22 towns is presented in Table 1.7. From this, we see that almost all industries with the exception of transport and communications have relatively high concentrations in Nairobi, with construction and business-related services such as finance and insurance being particularly concentrated there. Mombasa has a concentration in the transport and communications industry, as would be expected by its role as a major port. Manufacturing shows the least relative concentration in Nairobi and Mombasa and the highest share of employment in the major secondary towns of Nakuru, Kisumu, Thika, and Eldoret. This pattern may reflect the success of Government initiatives to decentralize manufacturing out of Nairobi and Mombasa, or it may reflect various locational advantages for resource-based industries. 1.17 Table 1.8 presents growth rates for wage employment during the period 1970-82 for the six towns with the largest employment shares. Only Mombasa and Thika (which serves as a satellite manufacturing town to Nairobi) had growth in wage employment greater than their growth in population. (Eldoret achieved this target in the latter part of the period.) Nairobi's growth rate in wage employment was slightly below its population growth rate, in contrast to the usual assumption that employment growth is concentrated in Nairobi. Therefore, despite the impressive performance of the Kenyan economy in achieving an overall growth rate for urban wage employment of 5.1%, a declining percentage of the urban population was able to find employment in the modern sector. 1.18 The concentration of modern sector activities in a relatively few towns is not unexpected. The wage employment patterns discussed above 9/ In the comparisons of population and employment which follow, Machakos and to a lesser extent Meru are continually shown to be outliers in terms of their ratio of urban employment to population. As shown in Table 1.2, both of these towns were quite small in 1969 and grew primarily as a result of expansion of their boundaries to include large rural areas. Hence, the employment pattern for these towns is more rural than urban. -8- presumably reflect agglomeration economies in terms of access to infrastructure, labor markets, product markets, business services, and communications, which are readily available in only a limited number of towns. Indeed, Table 1.7 indicates that the smallest 16 towns in the survey attracted a significant share of wage employment only in services (including government), trade, and resource-based industries, all of which are less affected by the agglomeration economies of larger towns. 1.19 The magnitude of the urban employment problem in Kenya during the next twenty years is sobering. The World Bank has estimated the size of Kenya's labor force in the year 2000 to be in the range of 9.8-12.0 million, with a middle-range forecast of 10.6 million workers out of a total population of 31 million.10/ The urban share of this labor force is expected to be on the order of 3.0 million, as compared to a 1980 urban labor force of 0.9 million.11/ Therefore, the number of new workers in urban areas will more than triple in the next twenty years. This rate of growth exceeds even the most optimistic estimates of the labor-absorption potential of Kenya's industry. The only alternative to provide for this increase is small-scale industry and the informal sector. C. Income Distribution in Urban Areas 1.20 There are no satisfactory statistics on the distribution of income in urban areas. Relatively good information exists only for Nairobi, which is not representative of other towns due to the concentration of wage employment there. For urban areas outside Nairobi, data are available only on the distribution of earnings from wage employment, which is of diminishing importance in all but the largest towns. For lack of more comprehensive information, Table 1.9 presents the Nairobi household income distribution derived from the 1974 Household Budget Survey, which is the best available source of data on income and expenditures. 1.21 The absence of good income data for urban areas has spawned a wide variety of estimates on the extent of urban poverty. Estimates range from 52% of the urban population using a "basic needs" definition of poverty,l2/ to a value of 3% based on a rural index of nutrition. 13/ Clearly such a range of estimates is too wide to play any useful role in public policy debate. Nevertheless, it is important to have some idea of 10/ "Kenya: Population and Development," op. cit., Table 3.3, p. 68. I/ Based on the population aged 15-64 and available for labor. 12/ "Planning for Basic Needs in Kenya," International Labor Office, 1978, p. 25. 13/ "Growth and Structural Change in Kenya," op. cit., p. 42. -9-- the extent and nature of urban poverty, and on this issue the data compiled over the past ten years give a fairly consistent picture. 1.22 The usual definition of poverty is based on a threshold household income adequate to purchase the food needed for a minimum level of nutrition. In applying this definition to urban areas, allowance has to be made for the fact that the economy of urban areas is intrinsically more monetarized than in rural areas. For example, the extent of urban squatting in the sense of not paying rent is very limited; even the very poorest generally have to make payment in some form for their living space in urban areas. Depending on how much is allowed for these factors, estimates of "absolute poverty" range from the 3% cited above to 15% using the estimates of the ILO. Given the inadequacies of the data upon which these estimates are based, it is desirable to look at more direct measurements of urban malnutrition. Some indication is provided by a survey in squatter areas in Nairobi in 1972 which found that 28% of children under the age of three suffered from some degree of protein-calorie malnutrition.14/ A more recent survey in Thika which sampled households on a random basis rather than in squatter areas found that 18% of children under the age of five were mildly malnourished, with less than 1% being severely malnourished.15/ While these figures are lower than for rural areas (the corresponding figures for rural areas around Thika were 25% mildly malnourished and 4% severely malnourished), they do indicate that urban malnutrition is an important problem in Kenya. Based on these data, one would conclude that a significant number of urban households, at least in the range of 10-15% of the urban population, live in a state of absolute poverty. 1.23 By far the largest number of the absolute poor in Kenya live in rural areas. This is to be expected; given the rural nature of Kenya's economy, the cities have not yet become the last refuge of the desolute and homeless. Instead, Kenya's cities are viewed as centers of opportunity, places to which impoverished rural families can send a household member to supplement their income through remittances. It is estimated that 21% of the urban wage bill is remitted back to rural areas, providing a major source of financing for smallholder innovation.16/ Clearly the economic attraction of urban areas will be a major factor in keeping rural-urban migration high for the foreseeable future. 14/ Blankhart, D.M., "Human Nutrition," in Health and Disease in Kenya, ed. by Vogel et. al., East African Literature Bureau, 1974, pp. 409-428. 15/ Figures cited in "Kenya: Population and Development," p. 53. 16/ "Growth and Structural Change in Kenya," op. cit., p. 43. See also "An Analysis of Income Transfers in a Developing Country: The Case of Kenya," J.C. Knowles and R. Anker, Journal of Development Economics, pp. 205-226, 1981. -10- 1.24 This discussion of urban poverty illustrates the difficulty of defining poverty solely in terms of income levels. Urban poverty in Kenya is not only a problem of income; it is also an environmental problem caused by lack of access to services which should be affordable to the poor. It is estimated that at least 25-30% of urban residents in Kenya live in unplanned and over-crowded slum settlements, most of which lack even basic services. The health and environmental problems in these areas are well-documented. For example, even though malnutrition among children has been found to be less severe in urban areas, a recent CBS survey in and around Nairobi found that infant mortality is higher in urban areas because of environmental diseases. Urban squatters are particularly prone to diarrheal and respiratory diseases which are spread by overcrowding and poor sanitation. These factors contribute to a decline in productivity, shortened lives, and a poorer standard of living. 1.25 Experience with various projects in Kenya designed to provide services to the urban poor has shown that such services are affordable if sufficient effort is made to lower their cost. For example, Bank-financed projects in Kenya which provide an integrated package of shelter and urban services have shown that households down to the 20th percentile of the urban income distribution can afford a site and service approach to shelter, while households down to almost the bottom of the urban income distribution can be reached through upgrading of existing squatter settlements. While in the long run, the urban (and rural) poor are aided best by economic development which raises their income, in the medium-term a great deal can be done through the supply of basic services to ameliorate the problems of poverty and improve the lives of the least advantaged. D. Urban Development and Kenya's Macroeconomic Strategy 1.26 Urbanization is a long-term issue, and most attention in Kenya is currently focused on short-term problems of economic management. Nevertheless, the macroeconomic policies which are being introduced will have a profound influence on the pace and direction of urbanization. In this section, we will sketch out some of the effects which will result from Kenya's trade, pricing, and industrial promotion policies. 1.27 Kenya has avoided policies of subsidizing food prices and basic services which have affected the distribution of income between urban and rural areas in many other African countries. By and large, urban areas in Kenya are expected to pay their own way, and the Government's pricing and investment policies have generally been consistent with this approach. In several instances, the Government has taken more direct actions to improve the income differential in favor of rural areas. For example, throughout the 1970s, the Government deliberately restricted wage increases to 75% of increases in the cost of living. The impact of this policy was a fall in the average real wage rate by 12.5% between 1972 and 1979, a decline which primarily affected modern-sector urban workers. 1.28 Nevertheless, in a more subtle way, Kenya's trade and industrialization policies have worked to the disadvantage of the rural sector. Kenya's impressive increase in manufactured output, which grew at a rate of 10.8% p.a. throughout the 1970s, was stimulated by increasingly -11- high levels of protection and an emphasis on import-substitution oriented to a domestic rather than an external market. This created a cost-permissive atmosphere which contributed to a rise in the prices of domestically manufactured goods relative to other prices. In addition, the trade barriers led to an overvalued exchange rate which tended to underprice all tradeable goods, including agricultural products. These factors led to urban incomes in the modern private sector being higher and agricultural incomes being lower than might have been the case with a less protective trade policy. 1.29 The Government has begun to take various measures to improve the system of import administration and rationalize the protection offered by the tariff system. Combined with other measures to promote exports, these policies will help turn the terms of trade back in favor of agriculture. The effect of these policies on urban incomes will not be uniform, however. A number of writers have pointed out the dualistic nature of Kenya's urban employment base.17/ In the view of these researchers, the modern sector has been relatively immune to fluctuations in informal sector employment in Kenya and has been able to maintain a level of wages and capital intensity which is not typical of the economy as a whole. In consequence, the size distribution of firms in Kenya is anomalous for its stage of economic development, bearing more resemblance to a high-wage capital-rich country than to a low-wage labor-intensive economy.18/ As the Government shifts more to a policy of export-promotion and rationalization of trade barriers, one can expect a gradual erosion of urban wages in the modern private sector and increasing labor-intensity of urban employment. This will improve the incomes of the urban poor as well as those employed in the rural sector. 1.30 A second important factor of Kenya's macroeconomic policy as it affects urban areas is the spatial implications for the location of economic activities and population. The types of industries which have benefited most from Kenya's import-substitution policies are large capital-intensive industries such as vehicle assembly. Such industries tend to locate near major urban centers such as Nairobi, both to be close to sources of skilled labor and capital and also because they are often dependent on imported intermediate goods rather than Kenya's domestic 17/ See, for example: (a) Employment, Incomes, and Equality in Kenya, International Labor Office, 1971; (b) The Kenya Employment Problem, H. Rempel and W.J. House, 1978; and (c) "Poverty and Growth in Kenya," Annex 1 of "Growth and Structural Change in Kenya," op. cit. 18/ "The Industrialization Process: Growth and Structural Adjustment," Annex III of "Growth and Structural Change in Kenya," op. cit., p. 21. -12- resource base. In contrast, the change in trade policies is likely to promote the development of modern small industries with higher employment elasticity than large industries. As studies in a number of countries have shown, such industries are more likely to locate close to markets and areas of resource availability, such as in secondary towns. The employment effects of Kenya's trade policy will therefore help reinforce the diversified pattern of urban growth into secondary towns which is already occurring. E. Summary 1.31 The information presented in this chapter gives a clear qualitative view of what can be expected for urban areas over the next two decades. Urban growth will be rapid and decentralized, a process which will be further strengthened by the rationalization of trade policies and a shift to export-oriented industries. Relative urban incomes in the modern private sector are likely to decline as modern sector employment becomes more "open" to international forces. Small-scale industries and informal employment, on the other hand, will be aided by these policies and should play a major role in providing urban employment. Rural-urban linkages can be expected to remain strong, both through income remittances and by the diversification of employment and services throughout Kenya. 1.32 It would be easy to use the growth projections discussed in this chapter to paint a vivid picture of the consequences of rapid urban growth and the need for massive urban investments. Certainly the Government of Kenya does face a substantial problem in providing adequate investment funds to service economic activities and provide for the basic needs of its urban population. The number of urban centers which require increasingly more expensive investments will grow, and the cost of maintenance and administration will increase at least as rapidly. On the other hand, other sectors of the economy also have priority needs, and the difficulties caused by Kenya's adjustment to its economic conditions will constrain the ability of the Government to service them. As urban centers can be expected to grow at a rate faster than overall population growth, there is little chance for any real growth in Government investment per capita in urban areas over the next five to ten years. 1.33 In the next chapter, we will review the Government's urban policy response of the past few years. The focus will be on the performance and achievements of the Government's urban policy and whether the urban policy objectives remain valid in the light of changing economic conditions. On this basis, we will be ready to move on in later chapters to discuss what the Government must do to improve efficiency in the use of urban resources and to promote alternative sources of resource generation. II. THE URBAN POLICY FRAMEWORK 2.01 In contrast to earlier policy statements by the Government, the 1984-88 Kenya National Development Plan pays relatively little attention to problems of urbanization. Nevertheless, the basic policy objectives of the -13- Plan as they affect urban development remain essentially unchanged from past Development Plans, save for the addition of the new "District Focus," which will be discussed in Chapter III. In general, the Government's policy objectives in urban areas can be summarized under the three categories: (a) Urban Spatial Policy--Urban growth should be decentralized throughout the country in order to provide for the needs of rural development, promote regional equity, and avoid excessive population concentration in Nairobi and Mombasa. Particular importance will be placed on "gateway towns" in less-developed areas and on the development of urban areas in western Kenya; (b) Urban Employment Policy--Modern sector establishments will have primary responsibility for providing for the employment needs of the growing urban population, and an important secondary source of employment will be provided by growth of the informal sector; and (c) Urban Investment Policy--Urban investment policy will emphasize the alleviation of poverty and the fulfillment of basic needs of the population, with priority given to improving health and nutrition, education and training, housing, social services, and water. These objectives are basically sound. Moreover, they are consistent with the focus of Kenya's economic development policy described in the previous chapter. The Government has made good progress in achieving some of these objectives, but implementation has been neither even nor altogether successful and has become increasingly constrained by macroeconomic conditions. In this chapter, we will review the progress which has been made and assess the priorities for continuing action on these policies. A. Urban Spatial Policy 2.02 For purposes of spatial planning, the Government has designated four levels of service centers.19/ These are: urban centers (resident population in excess of 5,000 inhabitants), rural centers (resident population between 2,000 and 10,000), market centers (resident population below 2,000), and local centers (primarily a cluster of rural inhabitants). Of these classifications, only the first two correspond to the definition of an urban area which is used in the census (resident population in excess of 2,000). Nevertheless, the service center designation is useful for investment planning, as it indicates the type of services which each level of center can expect.20/ In all, almost 1,700 19/ "Human Settlements in Kenya," op. cit. 20/ E.g., Everything from water supply to roads to schools to post offices. -14- centers have been designated throughout the country as service centers, with the vast majority of these being rural-oriented market and local centers. 2.03 In defining the urban part of Kenya's spatial strategy, a recurrent theme has been the belief that Nairobi and Mombasa are growing too rapidly, at the expense of urban development in the rest of the country. As we saw in the previous chapter, this pattern of growth did take place during the 1960s, when the special circumstances surrounding independence focused migration initially on the largest cities. Since then, however, urban growth has decentralized in response to the predominant rural-urban linkages in Kenya's economy. While this does not negate the need for an explicit urban spatial policy, it does change somewhat the policy measures which are required. 2.04 Outside of identifying general urban spatial priorities, the 1984-88 Development Plan does not suggest any measures by which the Government will pursue its spatial objectives, other than by encouraging ministries to take spatial priorities into account in the allocation of funds for construction of social and economic infrastructure. In this regard, it is interesting to review the past investment decisions of the Government to see whether they contained a high degree of spatial selectivity. Table 2.1 presents the allocation of Government investment funds to urban areas based on the FY80/81 Development Estimates. From this, we see that the allocation of funds for basic urban infrastructure (water, sewerage, housing, roads, and other infrastructure) generally matched the population share of the different urban areas.21/ On the other hand, Government investments in small scale industrial promotion through KIE had moved to smaller municipalities due to the completion of industrial estates in the larger municipalities. Finally, the planned investment in regional and national infrastructure such as provincial hospitals and airports was skewed toward larger urban areas, particularly Nairobi. From these data, we would conclude that the Government allocated urban investment funds for basic infrastructure essentially along the lines of population share. Where Nairobi benefited disproportionately, it was for investments with a national orientation rather than a purely local focus, a fact that is consistent with Nairobi's status as the capital city. In other words, these data do not reflect either a "pro" or "anti" Nairobi bias in the allocation of investment funds. 2.05 Some critics have argued that Kenya's spatial strategy should be more selective and concentrate resources on a limited number of "growth centers." Because of the lack of a major urban center in western Kenya, 21/ The allocations in the Development Estimates in any one year predominantly reflect investment decisions committed in previous years. The FY80/81 commitments in Table 2.1 are heavily influenced by Bank-financed projects for water supply in Nairobi (approved in 1978) and Mombasa (approved in 1975) and for low-income shelter projects in Nairobi (approved in 1975) and in Nairobi, Mombasa and Kisumu (approved in 1978). -15- Kisumu and Eldoret are identified in the 1984-88 Development Plan as deserving particular attention. The rationale for the growth center argument is generally the following: (i) there is a need to provide an alternative "magnet" to attract migration away from Nairobi and Mombasa, and (ii) the growth centers will benefit from economies of scale and complementarity of urban investments.22/ However, urban growth has already decentralized away from Nairobi on a fairly wide pattern and is not concentrating on a selected number of points. Therefore, if the Government is interested in holding rural-urban migration in secondary towns, it is probably wiser to strengthen a range of secondary towns rather than concentrate resources in a chosen few. The other argument, complementarity of urban investments, is of a very long-range nature and is difficult to justify in the face of a constrained resource base. Given the current constraints on the Government's investment budget, it is doubtful that the Government has adequate latitude to mandate a substantial reallocation of resources between urban areas. In this situation, it is probably of more importance that Government scrutinize priorities for investments within the various urban centers based on individual rates of return, rather than attempt a significant redistribution of resources on the basis of macro criteria. 2.06 The major value of Kenya's urban spatial strategy is that it recognizes the importance of Kenya's secondary towns in the economic development process. As we have seen in the previous chapter, the development of secondary towns is compatible with the demographic trends of urban growth and supportive of the export-promotion thrust of Kenya's industrialization strategy. However, once the importance of secondary towns is realized, it would probably be a mistake to go further and concentrate a significant amount of resources on a select few. Certainly, the secondary towns will grow at different rates. Eldoret, for example, has a significant advantage over Kitale, less than 100 miles away, in terms of its location on Kenya's inter-city transportation network. This helps explain the large share of manufacturing employment in Eldoret and its relative dearth in Kitale. However, while factors such as these should be taken into account in deciding on the allocation of investment resources, it should be in the context of their effect on the productivity of individual investments, rather than to influence a specific urban spatial pattern. 22/ The "growth center" argument for Kenya is developed in most detail in "National Urban Development Strategy for Kenya: Rural-Urban Balance," by H.W. Richardson, A. Khanna, and O.A. Sampaio, World Bank Consultant's Report, 1978. The rationale for choosing Kisumu as the major growth center for western Kenya would stand reexamination based on the results of the recent census. Table 1.2 shows that Kisumu had one of the lowest urban growth rates in Kenya during the 1969-79 period, once the effect of boundary changes is taken into account. -16- B. Urban Employment Policy 2.07 The most important employment policies for urban areas are the macroeconomic policies which are discussed in the previous chapter. In order to generate enough employment for the large expected increase in the urban labor force, it is essential that economic incentives be changed at a macro level to make industrialization in Kenya more efficient and to provide for a higher rate of labor absorption. In order to facilitate this process, the Government can also pursue more direct policies than "getting the prices right." Possibilities for stimulating more employment in Kenya's industry have been analyzed in a recent World Bank Report.23/ In this section, we summarize some of the results of this study and relate them to Kenya's overall urban development strategy. 2.08 The proposed rationalization of the external trade regime will encourage the development of small industries in Kenya. It is helpful to make a distinction in this regard between modern small enterprises and artisan industries. Modern small enterprises manufacture a much wider range of goods than artisan firms and will probably experience a faster rate of growth. Because modern small industry depends on innovation in the use of skills and physical inputs, the number of producers is more limited and the profit potential is greater. Artisan industry, on the other hand, makes use of traditional production techniques, with the result that entry is more open and competition is often intense. The artisan subsector exists by the production of "basic needs" items, low-quality consumer and producer goods for which the only substitutes are of higher quality at a far higher price. In general, the modern small industry sector provides the best alternative for stimulating a large number of new African entrepreneurs. As was previously mentioned, modern small industry is also more likely to respond to spatial advantages in location. 2.09 The Government considers its major program for the development of modern small industry to be Kenya Industrial Estates (KIE). KIE operates full-scale industrial estates in five municipalities (Nairobi, Mombasa, Kisumu, Nakuru, and Eldoret), smaller Rural Industrial Development Centers (RIDC's) in eight secondary towns (Nyeri, Machakos, Kakamega, Embu, Kisii, Malindi, Voi, and Homa Bay), and more modest Industrial Promotion Areas (IPA's) near RIDC's in several towns. Besides providing workshop space on the different types of estates, KIE offers technical assistance and access to machinery through Technical Advisory Centers (in RIDC's) and Technical Service Centers (on industrial estates). KIE also administers a credit program to provide fixed and working capital to small entrepreneurs.24/ 23/ "The Industrialization Process: Growth and Structural Adjustment," Annex III of "Growth and Structural Change in Kenya," op. cit. 24/ "The Industrialization Process: Growth and Structural Adjustment," op. cit., pp. 58-65. -17- 2.10 Besides the KIE program, it is a common feature in the planning of low-income housing estates to include some provision for workshops and other small industries. The Bank-financed Second Urban Project was originally designed to provide almost 1,000 plots in three cities for the construction of workshops, five bulk workshop sites for development by charitable institutions, 2,000 market stalls, and 350 artisan kiosks. Materials loans were to be provided for the construction of worksheds, and full cost recovery was expected on all sites.25/ A further advantage of this approach is that the workshop sites are integrated into the planning of residential communities, providing the residents with easy access to employment. While these targets have since been scaled back substantially due to cost overruns on the project, they do indicate the possibility of involving organizations other than KIE in employment promotion activities on a large scale. 2.11 Although local authorities have developed workshop sites through low-income housing projects, experience has also shown that the employment planning aspects of these projects is often weak. Neither the National Housing Corporation (NHC) nor the local authorities have expertise in assessing the market demand for workshops, although they do have considerable experience in building design and urban planning. Consideration should be given to strengthening the employment planning aspects of low-income housing projects through the Ministry of Local Government, NHC or the Town Planning Departments of the larger municipalities. These concerns are indicative of the need for more inter-sectoral planning in urban development, a topic we discuss in Chapter III. 2.12 Besides workshop facilities, there are a number of other areas where urban employment concerns can be strengthened. One industry which has strong linkages to the local economy is the production of building materials for low-cost housing.26/ Organizations such as the Housing Research and Development Unit (HRDU) at the University of Nairobi and HABITAT (the United Nations Center for Human Settlements, also located in Nairobi) have done a great deal of research on the use of local materials to lower building costs. This could be integrated with a KIE project to promote such industries, achieving both increased employment and lower costs in housing projects. 2.13 Another area worth considering in more depth is employment in the construction of low-cost housing. The monitoring study of the Bank- financed site and service project at Dandora in Nairobi showed that low- income housing projects generate a substantial amount of employment for 25/ "Kenya: Second Urban Project," World Bank Staff Appraisal Report, March 23, 1978, p. 30. 26/ "The Industrialization Process," op.cit., Appendix, p. 4. -18- small contractors (called "fundi's" in Kenya).27/ The possibility exists of using an agency such as the National Construction Corporation (NCCo) to provide technical assistance to these small contractors. NCCo is a parastatal established in 1967 to promote the development of the local contracting industry in Kenya. It faces a number of problems, such as: (i) too broad a mandate, taking it into sub-contracting, equipment leasing, and construction credit, (ii) poor cost recovery because of a lack of detachment from its clients, and (iii) an overemphasis on a small number of larger clients. However, NCCo would appear to have the potential to provide substantial assistance to small contractors if its operations are simplified and better cost control is introduced. C. Urban Investment Policy 2.14 The 1984-88 National Development Plan states that "the alleviation of poverty and the fulfillment of basic needs have always been a major concern of Kenyan development efforts."28/ A great deal of effort has gone into defining a basic needs investment strategy for Kenya.29/ Because of this extensive documentation, we will mainly summarize the accomplishments in each sector as they relate to urban areas. 2.15 The interpretation of official statistics on the provision of urban services creates a problem about the coverage of unplanned housing developments. These "squatter" areas are estimated to contain roughly 25% of the urban population of Kenya, with the specific percentage by town running from 15% (e.g., Thika, Kisii) to almost 40% (e.g., Eldoret).30/ In general, the unplanned housing developments lack most urban services. Where safe water is available, it is usually off-site or inconvenient. Other municipal services such as solid waste collection are seldom provided. These unplanned areas are often not included in official service statistics, either because they lie outside municipal boundaries or because their existence is not officially recognized. Therefore, although the official statistics for urban service coverage are impressive, they can conceal or gloss-over serious service deficiencies. With this caveat, we turn to a review of urban investment policy. 27/ See in particular the MEDIS 6 report on self-help building groups and MEDIS 7 on employment linkages, Dandora Community Development Project, reports by Senga, Ndeti, and Associates, 1979. 28/ "1984-88 National Development Plan," Government of Kenya, 1983, p. 32. 29/ Two of the more comprehensive references are: a. "Kenya: Population and Development," op. cit., Chapter IV, pp. 78-119, and b. "Planning for Basic Needs in Kenya," op. cit. 30/ "Third Urban Housing Project", Interim Report No. 1, Consultant's Report to the Ministry of Urban Development and Housing, 1980. -19- 2.16 Water. Returning to Table 2.1, we see that water supply accounts for a major proportion of urban infrastructure investment, 41% in the FY80/81 Development Estimates. The Government estimated that in 1982, approximately 75% of the urban population had reasonable access to an improved water supply system.31/ The quality of the supply depends on the size of the community and is generally better in the larger towns. A large part of the investment in this sector in recent years has gone into major works to provide for the growth of Nairobi and Mombasa. At the same time, the Government has supported the provision of safe water in rural areas and has improved the systems in a number of smaller urban centers and municipalities. 2.17 The responsibility for urban water supply is shared between the local authorities and the Ministry of Water Development (MWD). Most of the larger local authorities (with the exception of Mombasa) operate their own water system. MWD owns and operates water supply systems in five municipalities, 110 urban, and 90 rural areas. In addition, a limited number of water schemes in small urban and rural areas are operated by agencies such as county councils and Kenya Railways. The allocation of water resources, pricing of water supply, and the operation of water undertakings are governed by the Water Act, the Public Health Act, and the Local Government Regulations. 2.18 The financial performance of the water supply sector has been mixed. Tariff structures approximating long-run marginal cost pricing have been introduced in both Nairobi and Mombasa. While these tariffs have raised substantial financial resources which could have been used for additional investments, the lack of appropriate financial controls has undermined the intention of these policies. For example, Nairobi City Council (NCC) established a separate water fund as part of a Bank-financed project and built up substantial reserves to cover its local cost obligations under the succeeding project. Because of persistent cash deficits elsewhere in the Council, the water funds were diverted to pay for other recurrent expenses, leaving an inadequate cash reserve to cover NCC's committed obligations. A similar situation pertains to the Mombasa project, where MWD has used excess revenues to cross-subsidize other water projects. In general, most local authorities charge only enough to earn a cash surplus on their water account but not enough to cover the replacement cost of their assets. MWD on its part does not have financial objectives of the type found in most water utility companies and generally seeks only to cover its cost of operations and maintenance. 2.19 Overall, the water supply sector has achieved reasonable performance on the investment side. The major issue in the sector is financial performance, which ultimately could become a binding constraint on the ability of the sector to maintain its expansion. There may be justification for introducing a uniform pricing policy (not necessarily the same as uniform prices) for water in Kenya. Even this would not solve the problem in local authorities where water revenues are diverted to pay other council expenses. This problem needs to be addressed as part of the 31/ "1984-88 National Development Plan," op. cit., p. 36. -20- overall problem of local authority finances, which is covered in Chapter IV. 2.20 Sewerage. There are about 22 public sewerage systems in Kenya serving about 25-30% of the urban population. The FY80/81 Development Estimates allocated 9% of urban investment funds to sewerage systems, indicating that planned expansion of the system was moderate. In the larger cities, coverage of the sewerage system can be quite good, with the exception of the unplanned housing areas. Table 2.2 provides information on the type of sanitation provided in the major cities in Kenya. We see that with the exception of Mombasa, most of the population living in planned housing developments in large cities has good access to sanitation services, either by conventional sewerage or by septic tanks. 2.21 In the unplanned housing areas in Kenya, the question of appropriate sanitation presents a serious unresolved problem. The cost of providing sewerage to these areas, particularly given their irregular physical layouts, can require a substantial investment of scarce capital funds. In addition, the hook-up rates to the systems can be quite low because residents are obliged to make a significant additional investment for on-site sanitation facilities in order to properly utilize the system. Given the very low incomes of most of the residents, it is doubtful that imposing sewerage charges on them represents a socially optimal use of their limited resources. Nevertheless, many of the local authorities have taken a very conservative approach to sewerage, insisting that full water-borne systems have to be provided. The Kenya Low-Cost Housing By-laws Study 32/ recommends a more flexible approach to this problem, allowing the use of various low-cost sanitation systems where soil conditions and population densities permit.33/ One obstacle to adopting these systems appears to be the fragmented review procedure for housing projects used by local authorities. The Medical Officer of Health in the local authorities, who is required to approve the health aspects of all infrastructure and building designs, is often not familiar with the broader socio-economic considerations involved in the design of low-cost housing projects. For example, protracted disagreements with the Medical Officer of Health for Nairobi City Council over sanitation options finally led to the deletion of major parts of the Second Urban Project which would have provided shelter options for the very poorest residents of Nairobi. 32/ "Kenya Low-Cost Housing By-laws Study; Final Report," (two volumes), Ministry of Urban Development and Housing, 1980. 33, There is a substantial literature available on low-cost sanitation options which has been produced by the World Bank Research Program in Water Supply and Sanitation. (Publications available from the World Bank Publications Unit, Washington, D.C.) The UNDP is currently funding a demonstration project in low-cost sanitation in Kenya, which will hopefully promote better understanding of the options available in addition to conventional water-borne sewerage. -21- 2.22 Housing. The Government has identified housing policy as one of the important components in its efforts to alleviate poverty and provide for the basic needs of the population.34/ The private sector is expected to provide for the middle and upper income groups, with the Government providing indirect incentives through the development of housing finance. For lower income groups, the public sector is expected to play a more direct role in providing shelter meeting minimum standards of basic needs services. This is to be accomplished through site and service projects and community upgrading and the regulation of minimum building standards for other housing through bylaws enforcement. Thus, Kenya's housing policy is broadly consistent with the philosophy of limiting public sector involvement to situations where the private sector response is not adequate. 2.23 Central to the concept of site and service development is an emphasis on home ownership as an efficient means of mobilizing individual savings for housing construction. Kenya's housing policy has been oriented almost exclusively toward home ownership, with an official goal of providing a minimum two-room dwelling with separate kitchen and bath for every household. Because of the high resource cost implicit in this objective, much of the discussion on housing policy in Kenya has focused on the possibilities for cost reductions and the lowering of minimum standards. Site and service projects are seen as complementary to this effort by making the individual homeowner rather than the Government responsible for providing an adequate structure over time. 2.24 In fact, site and service projects are not new in Kenya and have been implemented in a limited form since prior to Independence. Local authorities occasionally demarcated tracts of public land, laid out plot boundaries and road reserves, and allowed households to occupy the land and construct dwellings in return for an annual rental fee. Although only minimal services were provided at the beginning, the local authorities have extended services to these settlement areas as funds became available. This type of incremental urbanization of land is what later site and service projects were intended to improve on. The site and service concept became part of the formal housing policy in Kenya with the adoption of Sessional Paper No. 5 of 1966/67 on housing policy, which was based on the recommendations of a 1964 UN Mission by Bloomberg and Abrams. However, site and service projects were not implemented on a large scale until the mid-1970s. At that time, the National Development Plan called for the National Housing Corporation (NHC) to devote the major portion of its resources to site and service projects, and the World Bank agreed to finance two large projects providing site and service plots and community upgrading in three major cities in Kenya. The Bank-financed projects introduced a number of features such as assistance in community development and the provision of materials loans which have helped to facilitate housing construction. 34/ An excellent discussion of the evolution of Kenyan housing policy is contained in R.E. Stren, Housing the Urban Poor in Africa: Policy, Politics, and Bureaucracy in Mombasa, Institute of International Studies (University of California, Berkeley), 1978. -22- 2.25 In view of the financial stringency in Kenya's budget, there is little justification for public sector housing agencies to embrace more expensive shelter alternatives such as contractor-built houses. The FY80/81 Development Estimates allocated almost 46% of urban investment funds to low-income housing and associated infrastructure and land acquisition. However, there is evidence that even this amount of investment will not be able to keep up with the need for low-income housing, and the performance record and possibilities for improvements in this sector are examined in depth in Chapter V. 2.26 Urban Transportation. From the standpoint of urban growth and planning, roads are the framework on which cities are built. While there is little justification for Kenya to embark on an expensive urban roads program, moderate and well thought-out investments in urban roads are a necessity to provide for structured urban growth and avoid a possible stranglehold of congestion such as found in many other cities. However, because urban roads are not classified as a basic needs investment and have therefore attracted little donor attention in recent years, investment in urban roads has shown the sharpest decline of any major urban investment category. Where roads accounted for 13% of urban investment expenditures in 1964, Table 2.1 shows that in FY80/81 only 4% of urban investment funds were allocated to urban road construction. This neglect of urban roads is somewhat inconsistent with national policy, where approximately 25% of the FY82/83 national development budget was allocated to road infrastructure outside of urban areas. 2.27 The continuing decline of urban road maintenance is the most disturbing feature of current policy. This is directly linked to the faltering financial performance of the local authorities, which have cut back on road maintenance as their budgets have become tighter. For example, Nairobi City Council virtually ceased road maintenance in 1979 because of a lack of funds. In Mombasa, continuing neglect of road maintenance led to widespread failure of road surfaces during heavy rains in 1979, causing damage which has still not been adequately repaired. On the national level, Kenya has put substantial emphasis on highway maintenance through the Ministry of Transport and Communications (MTC).35/ Because urban roads fall under local authorities, this expertise has not extended into urban areas. As it is, Kenya may be approaching the 35/ E.g., "Kenya: Second Highway Sector Project," World Bank Staff Appraisal Report, April 9, 1984. -23- situation all too common in Africa, where road surfaces give out when you cross into the city.36/ 2.28 The traffic engineering quality of the urban road network in Kenya is generally adequate. Only Nairobi is beginning to experience serious congestion problems during the peak hours, and many of these problems could be alleviated with simple traffic engineering improvements. This would also contribute to energy conservation; it is estimated that the introduction of traffic engineering measures in Nairobi could cut gasoline consumption in the city on the order of 10-15%.37/ Unfortunately, budgetary restraint has severely limited the ability of the Government to consider road improvements. 2.29 Public transport in Nairobi and Mombasa is provided by a mixed mode system based on conventional buses and paratransit vehicles locally called "matatus". Matatus provide virtually all the public transport service within secondary towns and much of the inter-city passenger transport. Studies recently carried out in Nairobi have shown that the mixed bus-matatu system is economically efficient and provides a highly complementary form of service.38/ Buses and matatus are all privately owned and operated, and no subsidies are provided to the system. However, neither the bus company nor the matatu operators have adequate access to investment funds. The mandate for urban transportation policy is divided among several ministries, with no clear policy guidance in this field. Thus, while the situation in urban transportation is not yet critical in Kenya, continuing neglect in this area could eventually constrain the economic efficiency of the cities in the movement of goods and people. 36/ Under the Transfer of Functions Act in 1969, only the seven major municipalities in Kenya retained responsibility for urban road maintenance, with MTC taking over responsibility for all other roads including urban and rural roads previously maintained by town and county councils. In practice, MTC has limited its activities in the urban areas to the maintenance of classified roads (i.e., trunk roads). In consequence, the secondary urban road network in most smaller towns in Kenya is virtually neglected except on a piece-meal basis as the local authorities can afford. Interestingly enough, a recent in-depth review of road maintenance in Kenya (see "The Organization of Road Maintenance" by Melody Mason, June 29, 1981, a paper prepared for the Projects Advisory Staff of the Transportation, Water, and Telecommunications Department of the World Bank) determined that maintenance of secondary urban and rural roads has not improved despite substantial amounts of money invested in MTC, and proposes as an alternative approach "increased involvement of local administrative levels in programming maintenance work." 37/ "Kenya: Energy Assessment", World Bank Report, January 1982. 38/ "The Matutu Mode of Public Transport in Nairobi," by S. Kapila, M. Manundu, and 0. Lamba, Mazingira Institute, 1982. -24- 2.30 Education. Primary education is a major priority of the Kenya Government. Throughout the 1970s, the Government increased the access of all students to primary education by progressively abolishing school fees. This has caused a large increase in demand for enrollments, in addition to the places which would have been required for normal urban growth. 2.31 After the reorganization of functions in 1969, only the seven largest municipalities retained responsibility for primary education. Even for these local authorities, the central Government has taken over much of the control and financial responsibility for primary education. For example, the central Government introduced a grant for teachers' salaries in 1978 because the local authorities were unable to carry the rapidly rising costs.39/ Until recently, the Government also provided a grant to replace the loss of revenue from the abolition of school fees, although the amount provided has generally not been adequate to compensate the local authorities for the large increase in enrollments. The major problem for local authorities which have retained responsibility for primary education has been to find investment funds to build new schools. Unlike rural areas, self-help (Harambee) efforts have not been effective in providing the number of urban schools required. Because of this, school construction is a major constraint on the expansion of primary education in urban areas. In addition, the central Government is facing major problems in maintaining its recurrent expenditures for education because of the tight budgetary situation. This has meant that local authorities cannot be assured of grant funds to cover teachers' salaries if they do open additional schools. 2.32 There are no adequate statistics on the number of children denied places in urban areas because of a lack of schools. In Nairobi, many of the children in squatter areas, such as Mathare, Kibera, and Kwongware, do not attend school, either by choice or because there are not enough schools in the area. Nairobi anticipates a need for 15 new schools a year over the next several years, but it has not been able to identify the resources to build them. Some schools are being provided by donors as part of low-income shelter projects, but the coverage of these efforts is uneven and has not been able to provide for the back-log of children seeking enrollment. 2.33 Health. Because doctors and hospitals are concentrated in urban areas in Kenya, urban residents are often assumed to be adequately served. However, a number of surveys have demonstrated that health conditions in urban squatter areas are as bad or worse than conditions prevailing in rural areas. Nairobi City Council estimated in 1981 that it needed to expand the number of basic health clinics by 50% to provide adequate coverage of the urban area, although it could not identify the funds required to construct and operate the clinics. Most of the municipalities 39/ The extension of the grant for teachers' salaries was largely a matter of equity, as teachers' salaries are set by the central Government and the Government had eliminated the local authorities' main source of revenue to pay for salaries, the Graduated Personal Tax, in 1973. -25- have found that the local clinics which they provide are preferred to provincial facilities operated in the same town. Preparation studies for the Second Urban Project also identified problems in maternal health care and infant nutrition in most squatter areas. Based on this, Nairobi, Mombasa and Kisumu have been working on the development of special programs in child nutrition, with the assistance of Bank-financing. 2.34 Other Services. Many other services are necessary to provide decent living conditions in urban areas. These include services such as solid waste disposal, markets, street cleaning, and pest control. Almost without exception, these services have deteriorated in recent years as the financial position of the local authorities have become more constrained. Moreover, as these services are financed primarily from general sources of revenue, there is little prospect that they will improve in the near future. D. Summary and Conclusions 2.35 Two major points can be summarized from this discussion of Kenya's urban development policies. The first point is that the Government's urban strategy is supportive of current demographic trends and Kenya's macroeconomic policy. In particular, the explicit recognition of the role of secondary towns in holding rural-urban migration and supporting the diversification of economic activity throughout the country is an important complement to Kenya's external trade and pricing policies. The second point to be drawn from the discussion is that by and large, the Government's actions have been consistent with its stated policies. In general, one can conclude that Kenya's urban strategy as put forward in the various National Development Plans is a reasonable basis for future planning and provides a consistent statement of priorities in urban areas. 2.36 Urban areas in Kenya are in general better serviced and better maintained than most cities in East Africa. This should not distract attention from the fact that there are a large number of urban residents living in over-crowded and unsanitary conditions. However, given the economic constraints on Kenya, it would be unrealistic to argue that urban areas should receive a significantly larger share of investment funds. The required resources either don't exist or are needed elsewhere. At the same time, it would be equally unrealistic to ignore urban areas or to give up on the policy gains which have been made. Urban areas have an important support role to play in Kenya's economic development by absorbing a large share of the new population and providing for their employment and service needs. The ability of urban areas to do so has to be maintained as part of the long-term economic development of Kenya. This does not argue for bold new initiatives in urban areas, but it does justify sustained and reasonable support for the progress which has been made so far. 2.37 These considerations set the basis for the two major themes to be developed in the rest of this report. First, the immediate priority for the urban sector in Kenya is to consolidate the institutional and financial framework which supports urban development. Urban institutions have sustained rapid growth and change in the last few years, and they are going to face even more rapid growth in the future. There are problems of -26- inadequate coordination, divided responsibilities, and over-extended agencies which have to be dealt with in order to allow this growth to be efficiently managed. Even more importantly, urban institutions will have to look to their own resource base as much as possible. The claims on the central Government budget are numerous, and urban areas are not the first in line. External donor agencies can help somewhat, but these funds are limited and have a high opportunity cost to the country in terms of other foregone investments. This leaves urban areas with only hard choices; there are no others. 2.38 In order to carry out the task of providing for urban growth, the only reasonable alternative for Kenya is the development of local capabilities, local initiatives, and local resources, with the advice and encouragement of the central Government. Kenya has maintained a tradition of semi-autonomous local authorities, and the basis for the further development of local capabilities is one of the best in Eastern Africa. At the same time, Government actions in such areas as local revenues, budgetary responsibilities, and central-local relationships have often been detrimental to local authorities and inadequately thought-out in execution. The reasons for this are complex and include implicit political constraints on the development of autonomous governmental units. Nevertheless, it is clear that Kenya, in implementing its economic development strategy, needs viable local authorities to carry out its urban policy objectives. Responsible local authorities with a sound resource base and a capacity for sustaining local initiatives can be a positive force in accommodating urban growth and servicing the diversification of economic activities throughout Kenya. On the other hand, if the Government allows current trends to continue, local authorities can become a drag on Government development initiatives and a chronic sink of poverty and investment funds. 2.39 The remainder of this report will explore in more detail the various financial and institutional mechanisms which support urban development in Kenya. Particular stress is placed on possibilities to improve cost-efficiency and resource generation and the means by which the central Government and the local authorities can strengthen these processes. The concerns reflected in this report can thus be viewed as an integration of the social concerns of the 1970s, when new directions were charted in basic needs investment policy, with the financial discipline required in the 1980s to sustain the programs in the long-run. The challenge now is to consolidate earlier initiatives and find the resources to support them on the scale which will be required to provide for urban growth in Kenya. III. URBAN DEVELOPMENT: THE INSTITUTIONAL FRAMEWORK 3.01 Prior to independence, local authorities in Kenya had virtually the sole responsibility for managing urban growth. The involvement of the central Government through the Ministry of Local Government was primarily regulatory, focusing on the supervision of local elections and the review of annual budget proposals. During the late 1960s and throughout the -27- 1970s, the central Government began to take more interest in influencing the pace and direction of urban development. This involvement has been reflected in increasingly articulate discussions of urban development strategy in the National Development Plans. It was also expressed in discrete events such as the creation of the National Housing Corporation shortly after independence, the transfer of functions from local councils to the central Government in 1969, and increasing government control of particular sectors such as primary education, health, and water supply throughout the 1970s. At the same time, the central Government has failed to develop a consistent internal framework for the planning and financing of urban development activities through local authorities, with the result that implementation of the urban strategies proposed in the National Development Plans has often been poorly coordinated. 3.02 The need to improve the coordination of urban development planning was emphasized in a recent Government report on public expenditures.40/ Some of the implicit concerns which justify a more coordinated approach to urban development policy are the following: (a) The Government has no planning mechanism to ensure that its investment program in urban areas is internally consistent and compatible with the strategy laid out in the National Development Plan. This means, for example, that it is difficult to monitor compliance with the Government's national urban spatial policy objectives except on a sectoral basis. There are no formal mechanisms for ensuring inter-sectoral consistency between investments, such as programming for expansion of water and sewerage networks to be consistent with priorities for opening up new urban areas for low-income housing developments. (b) As more attention has been paid to integrated urban projects, such as site and service developments, it has been recognized that better inter-ministerial coordination is required to ensure efficient implementation. For example, the urban shelter and community upgrading projects in the Bank-financed Second Urban Project have required at various times inputs from ministries such as the Ministry of Works, Housing and Physical Planning for review of on-site shelter investments, the Ministry of Local Government for off-site investments and community facilities, the Ministry of Water Development for review of water and sewerage components, the Commissioner of Lands for land acquisition and valuation, and the Ministry of Finance and Planning for authority to award contracts and incur expenditures. Delay in obtaining the necessary coordination between these agencies has been a serious impediment to timely implementation of the project. (c) Finally, it was recognized that the planning of individual sectoral investments has to take account of the overall financial 4°/ "Report and Recommendations of the Working Party on Government Expenditures," Government of Kenya, June, 1982, p. 56. -28- viability and implementation capacity of the local authorities. As an example, the Bank provided financing through several loans to Nairobi City Council for investments in water supply and shelter. Although each individual project could be shown to be financially viable, the overall financial weaknesses of the Council put the projects into jeopardy in terms of availability of local counterpart funds, adequate revenues to sustain recurrent costs, and the recycling of surplus revenues back into the investment program. 3.03 In order to establish a framework for the analysis of urban policy coordination in Kenya, it is suggested that urban development as an institutional process should have the following objectives: (a) It should provide technical support to the Government's macro- level investment programming in terms of maintaining a pipeline of suitable investment projects and monitoring their consistency with the Government's spatial, economic, and social objectives for urban areas (the MACRO objective); (b) It should provide a focal point for inter-ministerial coordination of investment projects to achieve efficient implementation and inter-project consistency (the MICRO objective); and (c) It should encourage the development of local initiatives and local capacities to deal with urban problems within the context of a sound financial base (the INSTITUTIONAL objective). In this chapter, a structure is suggested for urban development planning which seeks to achieve these objectives within the current institutional framework. An important step in this process is the specification of an explicit mechanism for policy coordination. This is provided in section B, following a brief description of the current institutional structure in section A. A. Existing Institutional Arrangements 3.04 As in most countries, the responsibility for urban development in Kenya falls on a number of agencies. The 1979-83 National Development Plan listed thirty-six separate branches of the public and private sectors which have a role in the formulation or implementation of urban planning and development.41! Of these, the most important are the local authorities, central Government ministries such as the Ministry of Local Government (MLG), the Ministry of Works, Housing and Physical Planning (MWHPP) and the Ministry of Water Development, and parastatals such as the National Housing Corporation and Kenya Industrial Estates. 41/ "1979-83 National Development Plan," Appendix II, p. 516. -29- 3.05 Local Authorities. The Government's urban strategy assigns a major role in the implementation of urban investments to the local authorities.42/ Following the UK model, local authorities in Kenya are semi-autonomous governmental units with prescribed rights and obligations put forward in the Local Government Act of the Laws of Kenya. The Local Government Act recognizes four categories of local authorities: municipal councils (which includes the City Council of Nairobi), town councils, urban councils, and county councils. In 1984, there were twenty municipalities, of which Nairobi and six others--Mombasa, Kisumu, Nakuru, Thika, Eldoret, and Kitale--provide the broadest range of local services, including, inter alia, primary education, health services, road construction and maintenance, water supply, sewerage, housing, solid waste management, drainage, markets, and social services.43/ The remaining municipalities are smaller and have fewer functions, with no responsibility for primary education.44/ Town councils, of which there are currently seven, have a more limiteT range of functions. Urban councils come under the jurisdiction of county councils and are considered to be a transitional form in the development of local representation. There were seventeen urban councils in 1984, all of which had been formed since 1979. County councils, of which there are thirty-nine in number, are regional in focus and have boundaries which coincide with the district administrations. Neither county nor urban councils exercise any significant range of urban functions, having had the major share of their responsibilities--primary education, public health, and roads--assumed by the central Government in 1969. Therefore, the major scope for local representation in the urban development process will come from the municipal and town councils. Even within the larger councils, the needs, resources, and staff skills vary 42/ The report of the Working Party on Government Expenditures focuses on the deficiencies of local authorities in the current planning process, but it continues to affirm the importance of the local authorities in promoting urban development. See "Report and Recommendations of the Working Party on Government Expenditures," op. cit., pp. 55-59. 43/ Mombasa does not have responsibility for water supply, which is being provided through the Sabaki Coastal Water Project. 44/ Between 1981 and 1982, nine town councils were elevated to municipal status, as well as creation of a number of town and urban councils. This wholesale elevation of various councils has had serious implications for the financial viability of local authorities. To give an example, Voi urban council, a town with 7,400 inhabitants in 1979, was established in 1981 with an estimated total indebtedness in excess of KSh 30 million to pay for housing and sewerage projects under construction. License fees had to be raised significantly in the town, and the overall financial viability of the new council was in serious doubt. (See recommendation vii, p. 59, in the "Report and Recommendation of the Working Party on Government Expenditures," op. cit.) -30- considerably. Nairobi, as the largest council, has a development budget more than twice that of Mombasa and three times as much as all the other municipal councils put together. While this is partially explained by Nairobi's large population, even municipalities of comparable size, such as Eldoret and Thika or Nyeri and Kitale, have significant differences in their revenues and expenditures. 3.06 A distinct feature of local authorities in Kenya is that they function as housing authorities. Patterned after the UK, this arrangement goes back to colonial times when large municipalities like Nairobi, Mombasa, and Kisumu provided a variety of rental housing schemes for African workers. Since independence, the diminishing role of public rental housing and greater emphasis on ownership housing have altered the picture considerably. The present and future role of local authorities in housing is not very clear, and there is some ambivalence in the thinking of both local and central Government levels on this issue. This topic is discussed in further detail in Chapter V. 3.07 Ministerial Responsibilities. The Ministry of Local Government has the primary responsibility for overseeing the operations, financing, and management of local authorities. The Ministry of Works, Housing and Physical Planning has responsibility for housing policy and legislation and for the preparation of budget submissions and supervision of public expenditures on housing. The Physical Planning Department (PPD) in the same ministry prepares land use plans for urban areas and does general spatial planning.45/ In the Ministry of Lands and Settlement (MLS), the Commissioner of Lands (COL) approves the development plans for government land, which in many urban areas constitutes the bulk of the available space, controls the allocation and/or approval of acquisition of land for public purposes, and the registration of land titles. The Ministry of Water Development owns and operates water systems in a number of urban and rural locations. The Ministry of Health and the Ministry of Transport and Communications also have significant responsibilities for preparing, implementing, and maintaining development projects in urban areas. 3.08 Parastatals. The major parastatals involved in urban development programs are the National Housing Corporation, the Housing Finance Corporation of Kenya (HFCK), and Kenya Industrial Estates. NHC is the major recipient of government investment funds for public housing, which it on-lends to local authorities for the construction of public housing. It also supervises the construction of housing estates as an implementing agency for the smaller local authorities. The HFCK is solely involved in housing finance, with almost all its funds devoted to mortgage finance for 451 The Ministry of Works, Housing and Physical Planning is also assigned a mandate for "urban development policy," a function that has shifted between ministries several times since 1980 and has generally been associated with the Physical Planning Department. As the emphasis in this chapter is on the coordination and planning of urban investments, relatively little attention will be paid to defining the potential policy role for MWHPP in urban development. -31- middle and upper income housing. Housing finance in Kenya is dealt with further in Chapter V. 3.09 The "District Focus" for Development Planning. As of July 1, 1983, the Government officially adopted a district-level focus for future development planning. The purpose of the district focus is to increase the scope for local initiative in development planning and to achieve a better integration of development projects at the local level. As such, this decision to decentralize planning is compatible with the increased spatial diversity of urban growth which was discussed in Chapter II. While the primary objective of district planning is to improve rural development projects, the local authorities in each district are required to coordinate their development plans through the District Development Committees (DDCs). The larger urban local authorities (municipal and town councils) are also expected to place more priority on the coordination of rural and urban development in planning their investment programs. While this innovation marks an important step in defining local responsibilities, it will still be necessary for the central Government to ensure that national development policies are effectively translated into practice at the local level. The major instrument for achieving this objective will be the Government's Forward Budget planning process. 3.10 Coordination of Investment Planning. The principal instrument for coordinating Government urban investments at the ministerial-level is the Forward Budget. The starting point for this process is provided by the indicative five-year development budgets presented in the National Development Plans. These indicative budgets are intended to be used by the sectoral ministries to produce detailed investment guidelines for the planning period. In practice, the process of producing detailed guidelines takes place with varying degrees of efficiency. Performance monitoring of planned investments has been generally poor, and planning targets are often inadequately defined and overly-ambitious. The Ministry of Finance and Planning (MFP) has recognized this problem and is issuing a series of guidelines to improve project monitoring. 3.11 The process of determining annual development budgets is based on rolling three-year projections of development expenditures, which are prepared initially by the sectoral ministries and discussed with MFP. The Forward Budget review is carried out by two separate working groups for Sectoral Planning and for Estimates, each group comprising officials from MFP and the sectoral ministry concerned. The difference in the functions of the two groups is one of degree. The Sectoral Planning groups ensure that the budgets for the three forward years are consistent with each other and reflect the pattern of development established in the National Development Plan. They are also responsible for identifying new projects, modifying existing ones, and monitoring on-going projects in terms of economic and social criteria. The Estimates groups focus more on expenditure requirements and adjustments for ongoing and new projects. 3.12 From the standpoint of urban development, a major weakness of the Forward Budget process is that the working groups are organized ministry- wise. For example, there are separate Sectoral Planning groups for the Ministry of Water Development, the Ministry of Local Government, and the -32- Ministry of Works, Housing and Physical Planning. Although some coordination was maintained in the past by assigning the same staff members from MFP to the different working groups, the latest alignment of responsibilities has broken this connection. Given the multi-sectoral nature of urban development, there is a valid argument for establishing a common Sectoral Planning group for closely related subjects such as those handled by the urban-related ministries. As programs of urban development often involve several agencies, this arrangement appears essential to ensure that the Forward Budget exercise, intended as a management tool, does not deteriorate into a routine exercise. B. The Investment Planning Process 3.13 Because of the number of agencies involved in the planning and implementation of urban development projects, it is helpful to define an explicit framework for coordination. The inter-ministerial framework presented in this section draws together the tasks required of the principal ministries, parastatals, district development committees, and the local authorities. By focusing on the sequence of coordination, the framework provided here should remain valid even if, as is likely in any ministerial-type government, particular tasks and departments are later reshuffled to different agencies. 3.14 Although the role of local authorities may appear obvious in a discussion of urban development, the "prime client" nature of local authorities is important to remember in a discussion of institutional arrangements, as central Government decisions have often been made to the detriment of this relationship. Because of this, the institutional arrangements for urban development planning, the preparation of schemes, and the monitoring of implementation should recognize the local authority as the starting point of the investment process. This process will have to recognize the limitations of existing local authorities and work within a flexible framework to develop their capacities. Thus, major municipalities such as Kisumu, Eldoret, and Thika may have the technical ability as well as the volume of investments to sustain project preparation and implementation responsibilities internally. The role of the central Government agencies for these municipalities should therefore consist of technical review and oversight, as well as playing an advocacy role for the municipalities in the Forward Budget process. Smaller local authorities will generally need greater technical assistance from central agencies, and may in some cases need to turn all itfplementation responsibility over to a central agency. This type of flexible support role has been adopted by a number of agencies which deal with local authorities, including the National Housing Corporation and the Ministry of Water Development. What is required is not only a continuation of this facilitating role, but also an explicit recognition of the role of the local authorities in the investment planning process. 3.15 Chart 3.1 presents a framework for the coordination of urban development planning which is based on the objectives discussed earlier. The essence of the process is that it uses the urban area as the starting point and the local authority as the principal client. Several points are worth noting with respect to this framework. First, it calls for the -33- CHART 2: A PROCESS FOR DEVELOPING A PROGRAM OF URBAN SERVICES Monitoring and MLG Evaluation | Concerned | tatal | Implementation or or Local Authority Operation and Ministry ~~~~~~~~~~~ ~ ~~~~~MaintenanceI +~~~~~~~~ MLG or Formal Approval Appropriate Ministry of Loan Sanctio Inter-Ministerial inclusion of Committee, Forward Budget and MFP ProcessI JCC (MLG, MWHPP, Local Authority, NHC, or Approval of City/Urban other parastatal or Department Area Development Prog co ncerned ) Financial Analysis an MLG Relation to LocalAuthority ________ j j ~~~~~~~BudgetJ Initial Review of t District Schemes and Development 1 Aggregation at Urban + Committees Area Level _ Provincial m f o rnca Cosutats orlZaT rDevelopment Schemes Awistance or or or |uLocal (for urban services [ nitsnAthrte and facilities) sPh ycals 1 Planning or Local Authorities Development Dartment Plans Direct Responsibility - -- ~- Feedback Responsibility World Bank-23729 -34- preparation of a development program for urban services and facilities, either by the local authority if staffing is available or by a Government agency or consultant if necessary.46/ The purpose of this development program is to define in preliminary detail a set of investment proposals which would be consistent with the economic potential of the urban area and the needs of the local authority. The focus of the planning exercise would not be on the traditional land-use/master planning approach, but rather would emphasize the urban infrastructure and other facilities needed by the town to accommodate its expected growth. As such, the plan would consider improvements in basic trunk infrastructure such as water, sanitation, and roads, and also related services such as solid waste disposal, community facilities, schools, and maintenance equipment. Making the local authority responsible for preparation of the program would strengthen the commitment of the local community and provide for the development of local initiatives, as indicated in the INSTITUTIONAL objective discussed previously. In addition, review of the proposals at the level of the DDC should help to improve the coordination of rural and urban development planning at the local level. 3.16 Many of the local authorities, at least at the level of the municipal and town councils, can be expected to be able to prepare an indicative statement of their investment priorities. The process could be assisted by the establishment of provincial technical assistance units or by the use of consultants. However, for specialized investments where the local authorities do not have an internal capability, it will often be necessary to provide direct access to specialized assistance from the relevant technical ministries. For example, NHC has four different methods of sharing responsibility for project planning and implementation with the local authorities, each of which entails a different division of tasks between NHC, the local authority, and/or consultants. In order to incorporate this aspect of project preparation into the investment planning process, Chart 3.2 presents a modification of Chart 3.1 for the development of housing programs. A similar chart could be prepared for water and sewerage programs involving MWD. The main point is that different technical ministries can be involved in the initial project identification and preparation stages without modifying the investment planning process in any significant way. 3.17 After the preparation of the initial project proposals and review by the DDC to ensure coordination with other district-level investments, the proposed planning process provides for a financial review by MLG in all cases where the investment is to be financed by a debt obligation on the local authority. This stage is provided for in the Local Government Act, which requires local authorities to obtain a loan sanction for every proposed investment. However, the current process considers the financial viability of the projects on a case-by-case basis only. What is perhaps more important is to determine the impact of the package of investment 46/ This process has been initiated by MLG with the preparation of guidelines for local authorities to prepare an individual "Local Authority Development Program," in connection with the planning exercise for the 1984-88 National Development Plan. -35- CHART 3.2: A PROCESS FOR DEVELOPING HOUSING PROGRAMS Housing Policy Guidelines, Monitoring MWHPP and Evaluation, Indicative Housing Programl |7,~I] | Local |I| Implementation, Operation | | Or Authority and Maintenance l~~~~~~~~~ * MLG 4 Formal Approval I _ _ _ _ _ of Loan Sanction l Inter-Ministerial Committee, Inclusion in Forward l and MFP Budget Process JCC (MWHPP, Determine Funding MLG, NHC, Arrangements; Local Authority Inclusion in National Housing Program Financial Reviewand M LG | Relation to Loca il Authority Budget Preparation of Housing Schemes to be taken up by the Public Sector (including Financial Analysis) Local Authoriti'es, NHC, oIdentification of or l b | Housing Program for the - 4 Consultants Medium Term (5/6yr) Housing Needs, l Supply Analysis J for Given Urban Area - Direct Responsibility - - - Feedback Responsibility World Bank-23730 -36- proposals on the local authority's overall financial position. This could be accomplished by a strengthening of the financial review procedures of the Local Government Loans Authority (LGLA), a topic which is discussed in more detail in Chapter IV. 3.18 Following the financial review by MLG, the investment proposals would be reviewed by a Joint Consultative Committee (JCC). Members of the JCC would include MLG in its oversight role for local authorities, technical ministries such as MWHPP and MWD, plus representatives of the particular local authority whose investment program is being reviewed. The JCC would be responsible for reviewing the internal consistency of the proposed investment program as well as the expected impact on the local authority's financial position, thereby satisfying the MICRO objective of the planning process discussed earlier. 3.19 Given the constraints on urban investment funds in Kenya, the crucial part of the investment planning process is to relate the proposed programs to the overall availability of funds. It is suggested that this should be carried out by an Inter-ministerial Committee at the level of the Permanent and Deputy Secretaries, which would work with MFP in the Forward Budget process. The Inter-ministerial Committee would be expected to play an advocacy role for urban development projects and would ensure that the funds necessary to support an integrated program of urban investments are provided in the sectoral budgets of the different ministries. Conversely, the Inter-ministerial Committee would also provide a forum to ensure that the scaling back of projects would be reflected on a consistent basis in the budgets of the various ministries. To give an example, the Inter-ministerial Committee would ensure that a decision to postpone a water supply project is reflected in the scaling back of low-income housing projects in that town until water becomes available. This step of the investment planning process would achieve the MACRO objective which was discussed above. 3.20 Until now, the description of the planning process has emphasized the planning for new investments. However, this framework is also applicable and perhaps even more useful for monitoring the progress of existing investments. Used in this way, the process would be initiated by the regular submission of progress reports on projects undertaken by the local authorities.47/ The JCC would review these reports and arrange for action to be taken on any matter which required coordination between a central government agency and the local authority. An additional task of the JCC would be to monitor progress on resolving past issues which had been raised in previous project reports. This would help overcome one of the current problems with central-local relationships, which is a lack of 47/ This procedure need not be applied for small investments managed wholly by the local authority, such as small pieces of road construction. However, it could be usefully applied for larger projects such as low-cost housing schemes which require a significant amount of liaison between the local authorities and the central government agencies. It would also apply to donor-financed projects with their greater requirements for regular reporting. -37- consistent follow-up on issues which have been previously raised. By reviewing project performance across several towns, the JCC would be in an excellent position to spot problems which are consistently hampering implementation. These issues could be raised to the Inter-ministerial Committee for discussion and resolution at the senior level of decision-making. The Inter-ministerial Committee would also provide a forum for monitoring the cash flow implications of the urban investment program and could take steps to ensure that anticipated disbursements on the investments are within the targets established in the Development Estimates. Finally, the project monitoring and evaluation function which has been included in the investment process would provide feedback on the experience with various projects. This would constitute a valuable "learning by doing" resource which would improve the efficiency of urban investments over time. 3.21 One of the main virtues of the proposed process is that it is relatively simple and internally consistent with the existing institutional framework which operates in Kenya. Because the urban investment program is expanding and reaching out to more local authorities on a spatially exten- sive basis, it is necessary to formalize these procedures to ensure that consistent information is provided to key decision-makers on a timely basis. This will require a careful detailing of the various steps in the process, including the format and information content of forms to be sub- mitted by the local authorities, the timing of the review process and the composition of the various committees, and the procedures for summarizing information at each step and presenting it for decision at the next level of review. The next two sections provide back-up discussion of this process, with section C focusing on the current and proposed task responsibilities of each of the major central Government agencies, while section D discusses in general terms some of the implementation issues involved in the planning process. C. Agency Tasks 3.22 Using the investment planning process described in the previous section, the tasks required of the various central Government agencies can be defined in more detail. While particular task items may be shifted between agencies, it is felt that the overall institutional strategy presented here is the one most compatible with and supportive of a decentralized focus for urban development. 3.23 The Ministry of Local Government. In the Kenyan framework, the Ministry of Local Government is virtually the "guardian" of local authorities and wields considerable authority over them. In recent years, MLG has steadily moved from a regulatory position to one of actively supporting and strengthening the local authorities in their development programs. However, by definition, urban development is multisectoral, and implementation of Kenya's urban development strategy will involve working with many other ministries. Close collaboration is called for between MLG and ministries such as MWHPP, MWD, and various parastatal organizations. Given the present and anticipated resource constraints, it is important that priorities for public sector investments in urban areas are sustained through the Forward Budget process. Considered in this light, the major tasks of MLG in the coordination of urban development projects are: -38- (a) formulate a program indicating the type of investments, public and private, required in urban areas over the medium-term period of 5 to 10 years; (b) canvass support from other ministries and agencies to promote economic development in the selected urban areas; and (c) establish and service the Inter-ministerial Committee of agencies directly concerned with urban infrastructure provision to provide guidance to the Forward Budget process. 3.24 At the same time, MLG has an important job in overseeing the financial position of individual local authorities. MLG also provides the capital financing for most local authority schemes (other than housing) through the Local Government Loans Authority (LGLA). At present, the LGLA is more of a conduit for the MLG development budget rather than an active financing agency. Within the existing framework of enabling legislation, it is certainly feasible to refashion LGLA's work so that it can take a more active role in the financial appraisal of local authority investment proposals. This would facilitate an integrated view of the capital and recurrent financing needs of local authorities. Within this framework, the tasks of MLG in working with local authorities should be: (a) provide guidelines for local authorities to prepare proposals for infrastructure and community facilities; (b) encourage local authorities to prepare city-wide capital development programs; (c) assist local authorities to evaluate the impact of the capital development program on their financial position, both for capital and recurrent budgets; (d) integrate the various local authority investment proposals into the Ministry's development budget; (e) organize technical assistance to local authorities; and (f) formulate and implement training programs for MLG and local authority staff. 3.25 Ministry of Works, Housing and Physical Planning. Housing is an important sector of urban development. In recent years, there has been a shift in emphasis from traditional high cost housing to affordable and realistic shelter programs for low-income families. However, an assessment of the housing program points to the need for new initiatives (see Chapter V). The role of MWHPP in formulating housing policy is therefore very important. Arising from this, the Ministry's tasks relating to housing are: (a) provide housing policy guidelines to NHC, local authorities, and other organizations concerned with housing; -39- (b) formulate housing programs indicating the nature and size of public investment over the medium-term and integrate these investment proposals into the Forward Budget process; (c) promote mechanisms for the effective participation of the private sector in providing housing; (d) encourage NHC and HFCK to establish forward planning procedures and assist these agencies to obtain funds both from government sources and capital markets; (e) monitor and evaluate housing programs to assess the usefulness of different approaches; and (f) promote research and training in housing. 3.26 Because of the Government's emphasis on shelter projects as a means of providing basic services to the low-income population, there has been a tendency in discussions of urban policy in Kenya to equate shelter projects with urban development planning. In practice, this can promote an .enclave" approach to investment planning, where residents in a particular project area may receive services (schools, health centers, sewerage facilities, etc.) far in excess of the standards prevailing in much of the community. While it is sound practice to make provision for the eventual extension of various urban services in all areas, a city-wide approach to urban development planning may provide a broader overview of service deficiencies and perhaps a more equitable distribution of scarce investment resources across the entire community. 3.27 Within MWHPP, the Physical Planning Department (PPD) has the responsibility for preparing and updating spatial development plans for almost all urban areas in Kenya.48/ To do this, the PPD has assembled a sizeable group of experienced professional staff who have established good working relationships with local authorities in preparing town plans and advising local authorities and other government agencies on issues of land use. Over the years, the PPD has attempted to broaden its approach from the purely physical aspects of land use to include a concern with the economic and socio-demographic aspects of urban development. Nevertheless, the frequent reshuffling of PPD during cabinet realignments has hindered the work of the Department in integrating land-use plans with broader aspects of urban development. Therefore, consistent with the mandate of MWHPP for urban development policy, the PPD has a responsibility to work within the JCC to ensure that its expertise in spatial planning is integrated with other concerns of urban development planning. 3.28 Ministry of Water Development. Extensions to water supply and sewerage systems are one of the key elements in determining the spatial pattern, intensity, and types of urban land use. Many aspects of urban 48/ Nairobi and Mombasa have their own Town Planning Departments to do physical planning. -40- development planning must be coordinated with the planning of water and sewerage systems, from the development of low-income housing estates to the location of new industrial areas. In addition, water and sewerage investments can be a large element of a town's capital program, and the financial planning of these investments must be integrated into the overall financial position of the local authority. On this basis, the role of MWD in urban development planning is: (a) maintain and operate water systems in urban areas falling under the direct control of the Ministry; (b) provide technical assistance and training to the staff of local authorities involved in operating their own water systems; (c) provide technical advice to MLG and the JCC on the planning of water and sewerage investments; (d) liase with NHC and PPD on the land-use planning aspects of extensions in urban water and sewerage networks; (e) plan strategic investments in national water supplies to provide for the anticipated growth in urban and rural populations; (f) develop pricing and financial policies for water investments and be an advocate for sound financial management in the sector; and (g) integrate proposed water and sewerage investments into overall urban investment profiles through the Inter-ministerial Committee on urban development. 3.29 Coordination with Other Agencies. This discussion has concentrated on local and central Government agencies which are involved in the provision of urban infrastructure. However, as Chapter II brought out, urban development requires more than investments in housing and other services. In particular, urban employment is likely to remain a major problem throughout the next two decades and beyond. Because of differences in outlook, it is unlikely that agencies such as KIE will ever be closely involved in the planning of urban development. Still, urban employment concerns are too important an aspect of urban development to permit institutional alignments to allow them to be neglected. A logical alternative is for MLG to develop an expertise in urban employment concerns, first by compiling data and undertaking studies and later in the articulation of policy directions and legislation to assist local authorities in efforts to support urban employment concerns. This role could help local authorities in meeting their responsibilities for economic development planning which have been established through the district planning focus. D. Implementation 3.30 The institutional structure which has been presented in this chapter relies as much as possible on the present responsibilities of existing agencies in Kenya. The tasks of the various sectoral ministries -41- and parastatals and the local authorities remain much as they are in the current system. What is new is the addition of an explicit mechanism for the coordination of urban investments, a change which is designed to give substance to the recommendations of the Working Party on Government Expenditures. 3.31 By necessity, responsibility for urban investment coordination will call for some institutional changes. The Development Department of MLG was established in 1979, and it has played a key role in working with MWHPP and the local authorities in the implementation of various projects. An important constraint on the expansion of this Department will be the difficulty of retaining experienced personnel in Government service, as well as the overall constraint on the expansion of staffing due to the Government's very tight financial situation. For this reason, MLG will have to work with other Government agencies in establishing particular functions. For example, MWHPP has recently established a Program Management Unit to coordinate the implementation of low-income shelter projects, and it will be necessary for both of these agencies to work together to ensure efficient monitoring of urban investment programs. Finally, the important function of the financial review of proposed urban investments can probably best be accomplished within the structure of LGLA, an option that is discussed in more detail in the next chapter. 3.32 Finally, it goes without saying that organizational structures are at best enabling mechanisms and depend wholly on the staff comprising them. Staff training and upgrading of skills is a theme recurring in all discussions about administration in Kenya. Compared to most other African countries, Kenya's experience in the provision of urban services and in the running of local authorities has been significant. A considerable pool of senior administrators is also available in the Government. Various donors have recently been involved in an assessment of training needs and the formulation of training programs for staff in government, parastatals and local authorities in the field of housing and urban development. Such efforts need to be encouraged and supported if the Government's urban development strategy is to succeed. IV. LOCAL AUTHORITY SERVICES AND FINANCING 4.01 Local authorities are the foundation for local participation in urban development in Kenya. Until now the larger local authorities have been able to maintain a reasonable level of services to their residents, at least to those living in authorized housing developments. However, because of the continuing decline of their financial base in relation to their obligations, the ability of local authorities to operate as independent entities is becoming increasingly constrained. It is the purpose of this chapter to set out the major factors contributing to this situation, as well as to make suggestions for possible improvements. 4.02 Various Government policy statements, including the report of the Working Party on Government Expenditures, have affirmed the positive role of local authorities in providing for urban development. For example, the -42- 1979-83 National Development Plan set the following objectives for local government during the plan period: (a) Improvement in the finances of local authorities so that they can finance most of their own requirements for shelter and infrastructure as well as meet maintenance and other recurrent expenses. This objective will be achieved through fiscal and management reforms with appropriate training of staff. The goal will be to restore fiscal autonomy similar to that which existed prior to 1970, within the municipalities and town councils. (b) To develop a grant-in-aid system to assist local authorities with inadequate resources. (c) To promote democratic local government and public participation through elected representatives. (d) To continue to train elected representatives for democratic leadership, and to train staff members in all categories of skills required to administer local authorities efficiently.49/ 4.03 In reality, these policy statements have been honored more in the breach than in practice. A number of Government committees and other agencies over the past decade have examined the problems of local authorities, with all of them coming to the conclusion that local authorities need to be strengthened.50/ Despite these recommendations, the Waruhiu Committee in 1980 found so little progress had been made in implementing reforms that it merely restated its agreement with the recommendations on local authorities made by the Ndegwa Commission and the Nyaga Committee almost ten years earlier. While this ambiguity about the role of local authorities is perhaps understandable during a time when the central Government has been expanding its role in the economy, the long-term implications of continued Government inaction on needed reforms will be for local authorities to become increasingly marginalized in their support role for the Government's economic development strategy. 4.04 What is at issue is not whether the functions of local authorities should be significantly expanded, but whether they can maintain 49/ "1979-83 National Development Plan", p. 455. 5°/ See, for example: (i) "The Report of the Public Service Structure and Remuneration Commission, 1970-71" (Ndegwa Commission); (ii) "The Report of the Nyaga Committee," 1973; (iii) "An Analysis of Local Government Finances in Kenya with Proposals for Reform," IMF Report, 1976; (iv) "Report of the Civil Service Review Committee, 1979-80" (Waruhiu Committee). (v) "Report and Recommendations of the Working Party on Government Expenditures," 1982, op. cit. -43- their current responsibilities in the face of a growing imbalance between their obligations and their resources. It is easy to cite deficiencies in the services provided by local authorities and their increasing dependence on the central Government. It is less easy to propose reforms necessary to reverse this process which are also politically acceptable. For example, there is no question that the revenue base of local authorities was eroded by the Government's takeover of the Graduated Personal Tax (GPT) in 1973.51/ As a result, the councils which were financially the most sound in the 1970s now register at best a marginal performance in their recurrent budget. Where previously the local authorities were able to generate substantial development funds on their own, they are now almost totally dependent on the central Government development budget. Nevertheless, no consensus has developed on reforms necessary to improve the revenue performance of local authorities, as indicated by the decision of the central Government in 1984 to terminate several local authority grants that were introduced to compensate for the loss of the GPT. Due to continuing Government indecisiveness on the issue of local authority revenues, it is fair to say that the financial position of local authorities currently represents the most important constraint on urban development activities in Kenya. 4.05 Section A of this chapter provides an overview of past trends in the revenues and expenditures of local authorites in order to put their performance into perspective. Section B provides a more detailed analysis of the recurrent financing of the larger local authorities in order to highlight individual differences in adaptation to an increasingly constrained environment and to evaluate various possibilities for improvements. Section C provides suggestions for improvements in local authority capital financing which would assist local authorities in fulfilling the important role assigned to them in Kenya's urban development strategy. A. Overview of Local Authority Revenues and Expenditures 4.06 In the evolution of central-local relationships in Kenya, the years 1969, 1973, 1978 and 1984 mark critical changes. In 1969, the Government passed the Transfer of Functions Act which transferred responsibility for primary education, health services, and road maintenance, as well as revenues from the GPT, from all local authorities except the seven largest municipalities to the central Government. In 1973, the Government completed the abolition of the GPT for the largest municipalities (with effect from January 1, 1974) and replaced it with a compensating grant which was gradually reduced over time. The loss of this important revenue source for municipalities without an off-setting reduction in their responsibilities forced the Government in 1978 to introduce several specific grants to cover the cost of services mandated by the central Government. Of these, the most important were the grant to cover teachers' salaries and the drugs and dressings grant to cover the cost of medical supplies which the local authorities are required to 51/ The GPT was payroll tax which was easy to collect (as it was levied on employers) and provided significant elasticity in revenues in response to economic growth. -44- provide free of charge. Finally, in 1984 in response to continuing budgetary pressures, the central Government eliminated all grants to local authorities except for the grant for teachers' salaries. As a result of these changes, the local authorities have been caught between rapidly escalating costs of services due to inflation and rapid urban growth, while the revenue base to support this growth has been steadily eroded. 4.07 Expenditures. The main activities and services which local authorities are involved in can be grouped as follows: (a) Control functions - such as land-use control, building by-laws, trading and other licenses, public health, public nuisances, traffic control, town planning, etc. (b) Provision of facilities and services - water, sewerage, roads, drainage, street lighting, health centers, education, solid waste removal, recreation facilities, cemeteries, housing, markets, commercial and industrial premises, slaughter houses, etc. As was discussed above, municipalities are involved in most or all of these activities. The smaller town and urban councils have diminishing responsibilities depending on their relative strengths and the particular needs of their jurisdiction. 4.08 There has been a fairly substantial growth since 1972 in the level of expenditures by local authorities. Table 4.1 provides a breakdown of the trend in total expenditures over this period by type of local authority. We see that total expenditures by local authorities increased from K£ 22.42 million in 1972 to K£ 78.15 million in 1981, a percentage increase of 14.9% p.a.52/ While significant, this rise was smaller than the growth of total GDP(16.6% for the period) and much smaller than the growth in central Government expenditures (21.4%).53/ Given that prices increased during this period in excess of 14% p.a.57/ and urban population growth was on the order of 7%, it is clear that reaT expenditures per capita by local authorities have declined sharply over the past decade. 4.09 Table 4.1 also shows that while Nairobi City Council has dominated local authority expenditures, accounting for over half of total expenditures throughout the period, its relative share fell slightly from 52/ Growth trends in this section are computed for the periods 1972-81 or 1975-81, because the provisional revenue data for 1982 and 1983 are not considered to be reliable. 53/ Growth in central Government expenditures is based on FY 72/73 to FY 80/81. To the extent that transfers to local authorities have been incorporated into central Government expenditures, this comparison is slightly inflated, although not enough to change the conclusion. 54/ Based on the change in the relative consumer price index in Kenya, International Financial Statistics and Direction of Trade, IMF. -45- 59% of the total in 1972 to 56% in 1981. The smaller town and county councils showed the highest rate of growth of expenditures, which may reflect the diversification of urban growth into secondary urban centers. 4.10 Table 4.2 provides the breakdown of expenditures by type of services for municipal councils over the period 1975-83.55/ From this, we see that the most important categories of expenditure in 1981 were primary education (14% of total expenditure), housing estates (10%), sanitary services (10%), administration (11%), water undertakings (16%), and health (9%). Interestingly enough, this percentage breakdown of expenditures in 1979 is very similar to the pattern which existed in 1964, just after independence, when the percentage distribution for municipalities was education (14%), housing (11%), sanitary services (13%), administration (10%), water undertakings (10%), and health (8%). The most significant change in expenditure was for roads, which went from 13% in 1964 to 7% in 1981. 4.11 Table 4.3 provides an analysis of recurrent and capital expenditures by municipal councils by input category. From this, we see that labor costs dominated the expenditures of municipal councils, accounting for 63% of recurrent costs in 1981. Purchases of goods and services accounted for another 34% of recurrent expenditures, and various types of transfers for the remaining 3%. Labor costs increased at less than the rate of inflation during the period 1975-81, despite the growth in the establishment sizes, due to Government policies which restrained the rate of wage increases in the public sector. Turning to capital expenditures; new capital investments accounted for approximately three-quarters of the capital expenditures for municipalities, while loan repayments and transfers accounted for the other quarter. 4.12 Table 4.4 presents a similar analysis of expenditures for town and county councils by type of input. From this, we see that the break-down of recurrent expenditures in town and county councils is roughly similar to municipal councils. Recurrent costs have grown at a slightly faster rate than for municipalities, while the rate of growth of capital expenditures has been slower than for municipalities. 4.13 A final way of gauging the growth of expenditures by local authorities is by comparison with the central Government. In 1964 and 1969, the total expenditures of local authorities including both recurrent and capital expenditures were about 25% of the size of central Government recurrent and development expenditures. By 1975, the ratio of expenditures by local authorities had fallen to 8% of central Government expenditures, and by 1979, it had been further reduced to 7.4%. These figures confirm that the major growth in size and responsibilities of the public sector has been with the central Government rather than the local authorities. As a result of these trends, the share of the local authorities in gross capital formation in the public sector fell from 15% in 1969 to approximately 5% in 1975 and 1979. 55/ A change in the definition of categories in the official statistics prevents extending the series back to 1972. -46- 4.14 Revenues. In order to finance their expenditures, local authorities have had access to a variety of revenues. The recurrent expenditures of the councils are met from three principal sources: (a) General revenues - primarily the property tax (rates) which is levied mainly by municipalities and town councils, plus agricultural cesses imposed by county councils; (b) Specific fees and charges - revenues collected for specific services provided by the local authorities (such as water tariffs and housing rents) and charges arising from local authority control functions (business licenses, etc.); and (c) Government grants - which until recently consisted of: i. the GPT Compensating grant, introduced in 1974 to compensate municipalities for the loss of the Graduated Personal Tax; ii. the Teachers' Salaries grant, introduced in 1978 to cover the cost of teachers' salaries on the grounds that the salaries are negotiated directly with the central Government; iii. the School Fees grant, which had been increasing over time to compensate local authorities for the loss of revenues as schools fees were progressively abolished for primary education; and iv. the Drugs and Dressings grant, to compensate local authorities for health supplies which they are required to provide free of charge. As noted above, these grants have now been fully eliminated except for the grant for teachers' salaries. 4.15 Capital expenditures of the local authorities are met from internal cash generation and by loan funds, with the two principal sources for loans being LGLA and NHC. In addition, Nairobi City Council is authorized to issue its own stock (debt obligations) with the permission of the Ministry of Finance and Planning, and other local authorities may obtain bank loans with permission of the Ministry of Local Government. 4.16 Table 4.5 presents the trends in recurrent and capital revenues of municipalities for the period 1972-1983. Overall, revenues under the control of the local authorities (taxes and sales of goods and services) have grown at a rate of 13% p.a. over this period, less than the rate of inflation but reasonable considering the relatively inelastic tax base available to the local authorities. Government grants decreased at a rate of 9% p.a. between 1974-81, and this decline would be even greater if it was measured against the potential level of GPT which would have been collected by the municipalities. Finally, while loans raised by local authorities grew at a rate of 13% p.a. during the period 1972-81, virtually all of this increase took place in 1981 due to large disbursements on -47- donor-financed projects. Overall, it is clear that the resources which the central Government has made available to local authorities have grown at a rate significantly below the rate of inflation and have provided no margin for urban population growth. 4.17 Table 4.6 presents the revenue trends for town and county councils for the period 1975-83. The primary difference as compared to municipal councils is the greater extent to which town and county councils rely on their own direct and indirect taxes for their revenues (54% of recurrent revenues versus 43% for municipalities in 1981). This is partially explained by the fact that functions which were financed by grants for the larger municipalities are operated directly by the central Government for town and county councils. Table 4.6 also reflects the heavy reliance of county councils on agricultural cesses for their revenues. A few councils, particularly those in coffee growing areas, do very well from the collections of cesses and probably collect more revenues than are actually needed, given the limited rage of functions carried out by the county councils. On the other hand, many county councils which do not have access to a good base of agricultural cesses are chronically deficient in revenues. The often chaotic financial affairs of the county councils is sometimes used as an argument against the current system of local authorities in Kenya, even though the roles and responsibilities of the county councils and the more urban-based municipal and town councils are drastically different. The role of the country councils is expected to become even more limited as the District focus strategy is implemented, giving a stronger role in development matters to the District Development Committees. 4.18 Staffing Situation of Local Authorities. Besides their financial problems, several other factors which are often cited as contributing to the difficulties facing local authorities include their management shortcomings and their inability to retain qualified and competent staff. Local authority salaries, although generally competitive with the central Government, are not so relative to the private sector. This causes severe shortages for skill groups heavily in demand by the private sector, particularly for technical staff such as engineers, surveyors, land valuers, and financial staff. While these problems also affect the central Government, to some extent the ministries have been able to compensate by the use of limited term expatriate advisors, an option which is generally not feasible for the local authorities. The result is that project implementation capacity of local authorities is severely constrained, a problem which has caused serious delays in the implementation of various donor-financed projects. 4.19 Another factor which has been cited as contributing to the difficulties experienced by the local authorities in retaining qualified staff has been the influence exercised by elected councilors on the engagement, advancement, and dismissal of staff. This factor has led to many councils being over-staffed at the lower levels, with Nairobi City Council being most often cited in this regard. Because of these problems, legislation was recently enacted to bring all senior and professional staff of the local authorities, including the town clerks, under the jurisdiction of the Public Service Commission, which oversees the employment of central Government personnel. -48- 4.20 Conclusion. In summary, the ability of local authorities to maintain their financial position within an increasingly constrained economic environment has been increasingly hampered by decisions of the central Government which have increased their responsibilities without providing an adequate increase in their revenue base. As a result, the expenditure patterns of the local authorities have not been able to keep up with inflation and provide for the rapid population growth which has occurred within most urban areas. It is clear that local authorities will not be able to sustain their potential role in providing for urban growth unless they can be assured of an adequate resource base. For this reason, the next two sections of this chapter will focus on issues of local authority financing. The next section provides a discussion of the recurrent financing position of the seven largest municipalities in Kenya. B. Recurrent Financing of Local Authorities 4.21 The financial system used by local authorities is based upon the U.K. model of local government accounting. This system provides for a separation of recurrent and capital revenues and expenditures by the maintenance of separate funds to handle major activities of the council. Most recurrent expenditures and income are handled through the General Rate Fund (GRF), the only statutory revenue fund specified in the Local Government Act. In addition, most of the local authorities maintain separate funds to cover activities such as water supply, public housing, trading services, abattoirs, and renewal of equipment. 4.22 Financial Performance. The financial performance of most local authorities has been poor. Table 4.7 provides a summary of the position of the GRF the years 1980-82 for the seven largest municipalities, prior to the abolition of most Government grants.56/ The most interesting feature of this table, aside from the overwhelming dominance of Nairobi in all categories of revenue and expenditures, was the consistently weak position of the GRF in most of the municipalities. Only two of the councils (Nakuru and Thika) expected to maintain a marginal surplus position in their GRF over the three-year period. The remaining local authorities expected to sustain deficits, and the trend was generally for these deficits to grow. 4.23 In order to maintain themselves financially in the face of persistent deficits, most of the local authorities have abandoned the pretense of separate fund management. A few of the local authorities had built up surpluses in their general reserve funds, which were being drawn down to cover the deficits in the GRF (see Table 4.8). The remaining local authorities covered their recurrent deficits by tapping surpluses in funds such as water supply and housing. 56/ Because of a lack of uniformity in the presentation of financial accounts, the various categories of revenues and expenditures are not necessarily fully consistent across local authorities. Also, estimated budget figures should be treated with caution, as experience has shown that they can be unreliable as a guide to actual budget performance. -49- 4.24 While looked upon as a necessity by the financially-strapped local authorities, the practice of using surplus balances from funds such as water and housing amounts to little more than "begger your next generation" in an economic sense. The reason for the establishment of separate capital funds is to permit the generation of surplus balances to cover replacement and expansion of the existing capital stock.57/ In fact, the basic rationale of the Local Government Loans Authority, which is discussed in the next section, was to permit local authorities to hold their financial surpluses in an interest bearing account, while LGLA would loan out the funds deposited with it to other local authorities in need of capital funds. This arrangement, which previously allowed the local local authorities as a group to generate most of their required capital funds, no longer operates in this fashion because the local authorities have diverted any surpluses to cover deficits in the GRF's. To the extent that LGLA has funds to lend, they are taken from the Government's overall development budget at the expense of other needed investments. In effect, the local authorities are covering their recurrent expenditures by running down their capital stock, leaving a capital deficit which will eventually have to be covered from other sources. 4.25 Even the nominal positions of the various funds shown in Table 4.8 overstate the true financial situation of the municipalities. The system of local authority accounting used in Kenya realizes revenues as they are accrued, rather than collected. Most local authorities have substantial arrears in their collection of rates and other fees. In order for the budgeted accounts to have any relation to their nominal cash position, it would be necessary to include a provision in the estimates for "contributions to working capital in anticipation of collection of revenues." No local authority includes this provision in its estimates, with the result that most of the surplus balances of the various funds actually consist of accounts receivable of dubious merit. The MLG has refused requests by local authorities to write-off portions of the receivables which the local authorities deem to be uncollectable.58/ As a result, the budget estimates submitted annually by the local authorities have little relation to the actual cash position of the various funds. 57/ This principle, if fully applied, would require the local authorities to price services at their long-run marginal cost. Most local authority services are priced to cover average historical capital and operating costs and perhaps to generate a small surplus. Only Nairobi has priced its water supply at a level close to long-run marginal cost, and this permitted Nairobi's Water Fund to generate large surpluses which were eventually used to cover deficits in the GRF. 58/ A prominent example of this is the fact that most local authorities carry large arrears on their books for rates owned by Kenya Railways. These accumulated balances can be very large, sometimes amounting to half a year's total rate bill for the local authority. Given the very weak financial position of Kenya Railways, there is little likelihood that these arrears will be paid; moreover, Kenya Railways is disputing the claims. Despite this, most local authorities continue to include rates due from Kenya Railways in their budget estimates and fund balances. -50- 4.26 The large amount of arrears in rates is partially caused by a lack of political commitment within the local authorities to enforce revenue collection. Another important reason, however, is that the legal provisions available to enforce collections are outmoded and difficult to implement. The steps required to process a case of arrears are cumbersome and time consuming. In addition, the penalty for arrears, which is fixed in the Rating Act, is below the current interest rate for commercial borrowing, making it advantageous for rate payers to delay paying their rates as long as possible. Some of the local authorities have engaged private solicitors to process rate cases, but legislative reform of the Rating Act would be a more efficient way of improving collections. Finally, various central Government agencies, including parastatals, have been among the slowest rate payers, over which the local authorities have very limited control. 4.27 The local authorities are required to submit their budget estimates annually to MLG for approval. In practice, there is a great deal of dissatisfaction with this process. The focus of the budget review process has often emphasized individual items of expenditure rather than the overall financial position of the local authority. The local authorities complain about arbitrary changes to estimated expenditures and delays in the approval of estimates, which can cause significant delays in the issuing of rate notices. At the same time, the Ministry often fails to take effective action about serious financial problems, for example by neglecting persistent arrears run by many councils. Even after the estimates are approved, there is often poor expenditure control and cost overruns by the local authorities are common, despite the fact that expenditures in excess of budgeted amounts are supposed to be approved by the Minister of Local Government. Given the short staffing in MLG, it would be more effective for MLG to concentrate its efforts on assisting the local authorities prepare viable financial plans, while letting expenditure control be covered by the audit process. Unfortunately, local authority audits are also significantly behind schedule. 4.28 Recurrent Revenue Sources. Prior to the recent elimination of most Government grants, the major local authorities relied upon rates, license fees and user charges, and Government grants in roughly equal proportions for their recurrent revenues. Table 4.9 presents the percentage distribution of the different sources of recurrent financing available to the seven largest local authorities in 1980. Although the trends are not uniform, there was a tendency for rates to constitute a greater share of revenue for the larger local authorities, with correspondingly less emphasis on licenses, user charges, and Government grants. The smaller reliance on grants by Nairobi and Mombasa reflected the fact that these municipalities provided a wide range of services which were not covered by Government cost-reimbursement grants. 4.29 Property rates are the most important source of general revenues for the local authorities. Although the Valuation for Rating Act allows wide flexibility in the basis for valuing property, in practice all local authorities in Kenya rely on unimproved site value rating, which is based on the market value of the land only and not on the structures on the land. The Valuation for Rating Act provides for revaluation of property at -51- least every five years, but this has proven to be difficult because of a general shortage of valuers in the country (only Nairobi and Mombasa maintain their own staff of valuers). As a result, the valuation rolls for most local authorities fell substantially out of date during the 1970s. As a condition of the Bank-financed Second Urban Project, Mombasa introduced its first updated property roll since 1959 at the end of 1981, and Nairobi introduced its first new roll since 1971 at the beginning of 1982. Substantial public outcry resulted from the introduction of the new rolls, but the actions helped establish the principle of revaluation, which will assist other local authorities in improving their valuation base. 4.30 Until fairly recently, many local authorities waited until their periodic five-year revaluation in order to increase their revenue from rates. While local authorities are legally entitled to increase their rate levies (the percentage tax rate which is applied to the valuation base) between revaluations, political considerations have often restrained many of the councils from taking this action. As a result, the revenue yield from rates could remain essentially static for a five-year period, except for small additions to the rate base from supplementary valuation rolls. Financial pressures due to the effect of inflation on expenditures have now forced most local authorities to be more flexible, with annual or biannual increases in rate levies becoming standard practice. 4.31 A number of suggestions have been made to introduce more elasticity into the revenue yield from rates. One of the most frequent suggestions has been to change to improved site value rating, which would include both structures and land in the valuation base. The administrative problems of changing to this valuation base are substantial, particularly given the shortage of qualified valuers in Kenya. The local authorities have been hard pressed to maintain accurate land registration rolls; introducing the valuation of individual structures would be more than the system could handle. There are also theoretical reasons for preferring the unimproved site value basis, on the grounds that it is neutral with respect to land development and hence discourages holding land in an unproductive use. As an alternative measure, many local authorities in developed countries have indexed valuation ratings between revaluation periods, using an index of property sale prices during the year. This change provides automatic buoyancy in property tax collections geared to changes in land values, without requiring the politically difficult step of increasing tax rates. As an additional measure, many local governments have found that building permit fees can be a lucrative source of revenue. This latter change would require improvement in the building approval and inspection process to be effective, but these administrative changes may be easier to implement than extension of the valuation base to include improved property valuations. 4.32 Another possible compromise position on improved site value rating would be to introduce a tax on the rental value of land and structures in prime areas of the city, such as the commercial and industrial zones. This tax would primarily benefit Nairobi and Mombasa, but towns such as Eldoret and Thika with a significant manufacturing base could also benefit from the tax. On the other hand, because the tax could reduce incentives for firms to locate in secondary towns, many of the -52- smaller local authorities may be reluctant to introduce such a tax. An alternative version of this tax has been introduced by Nairobi, which imposed differential rate levies on the unimproved site value of land in the commercial and industrial areas of the city as compared to the levy in residential areas. 4.33 Another major source of recurrent revenues for the local authorities consists of various license fees and service user charges. These cover a wide range of activities from hawkers' licenses to ambulance fees, from cesspit cleaning charges to building approval permits. In general, local authorities have not made maximum use of these charges, and they represent a source of revenue which could be exploited further. Referring to Table 4.9, it is interesting to note that both Thika and Eldoret were able to produce budgets with either a small deficit or a surplus by relying heavily on fees, despite the fact that their revenues from rates were relatively low. On the other hand, Nairobi, with a very strong rate base, ran a large deficit because the council had not raised fees for a number of years. To the extent that user charges represent payments for services which directly benefit residents (such as solid waste collection), there is a strong case for ensuring that the fees are adequate to cover the full costs of providing such services. On the other hand, license fees and other indirect charges may not be an equitable way of raising general municipal revenues, particularly if they act to discourage small-scale entrepreneurs, and certainly they cannot substitute for an adequate property tax base. 4.34 From this discussion, it is clear that local authorities are not making adequate use of the revenue instruments which are already available to them. In a time of scarce budgetary resources in general, the local authorities must be encouraged to make greater use of existing taxes and charges, and the central Government should seek to encourage greater reliance on and enforcement of local revenue measures. Nevertheless, it is also true that the current revenue instruments of the local authorities do not provide the elasticity required to cope with inflation and high urban growth rates. Revaluation of property every five years and increases in license fees and charges will not provide the buoyancy of revenues required to allow the local authorities to maintain their present services and expand them to provide for new urban residents. The local authorities can not expect to be unaffected by financial cutbacks or be allowed to leave untapped any revenue base which is open to them. On the other hand, in imposing financial discipline throughout the economy, the Government should seek to avoid jeopardizing its long-term economic development objectives, which should include the maintenance of strong urban local authorities. For this reason, it is necessary to review carefully the logic and implications of the Government's decision in 1984 to eliminate most revenue grants to the local authorities. 4.35 The IMF Report in Perspective. The last comprehensive study of local authority finances in Kenya was carried out by the International Monetary Fund, in fulfillment of a condition of the credit agreement for -53- the Bank-financed First Urban Project.59/ This report has never been officially released by the Government, although it has been influential in the discussions on local authority finances in Kenya. As the financial situation which generated the IMF report (i.e., the abolishment in 1973 of the GPT for local authorities) is similar to the existing situation (i.e., the recent abolishment of most central Government grants to local authorities), it is helpful to reexamine the recommendations of this report as they relate to the current financial circumstances of the local authorities. 4.36 In point of fact, many of the reforms recommended by the IMF report were implemented by the Government. At the time the DMF report was written, the Government had not instituted the grants to cover the cost of teachers' salaries and drugs and dressings. With the GPT compensating grant being reduced, the municipalities were in a very difficult financial position, with every expectation of a serious deterioration in their ability to provide services. Therefore, it is not surprising that the IMF report focused largely on the problems of recurrent revenue sources and that the remedies suggested were in fact very similar to the actions which the Government ultimately introduced. In particular, the main recommendation of the IMF report called for the establishment of grants for teachers' salaries and for drugs and dressings, actions which were inevitable in view of the Government's mandate that these services had to be maintained. 4.37 Other recommendations in the IMF report and their rationale and current status are: (a) Gradual elimination of the GPT compensating grant. In general, the IMF report favored changing Government grants from a general purpose to a cost reimbursement basis in order to improve financial discipline. While the GPT compensating grant has now been abolished, the recent elimination of the grants for drugs and dressing and school fees means that the Government has not been consistent in applying the principle that centrally mandated services should be matched by an adequate and defined revenue source. (b) Regularization and centralization of the revaluation process. This was agreed as part of the Second Urban Project and has been partially implemented. An increase in the number of Government valuers and better coordination with the local authorities would be desirable to fulfill this objective. (c) No recommendation for the introduction of improved site value rating, except for a tax on the rental value of commercial property in Nairobi and Mombasa. The IMF report rejected improved site value rating on administrative grounds. The tax on rental values in Nairobi and Mombasa has not been implemented, 59, "An Analysis of Local Government Finances in Kenya with Proposals for Reform," IMF, December 8, 1976. -54- but differential rate levies have been introduced in particular areas of the cities. (d) Other taxes. The IMF report had little to say about other possible revenue sources and only briefly mentioned the entertainment tax (small yield), increases in fees, licenses, and other charges (not investigated), and cross-subsidies from the water and housing funds to the GRF (no firm position). (e) Budget preparation and approval. The IMF report noted the large discrepancies which exist between the accrual and the cash budgets of the local authorities, the lack of uniformity in accounting practices, and the difficulties faced by local authorities in maintaining current budgetary procedures. However, the recommendation of the IMF report to put all the accounts and estimates of the local authorities on a cash basis has been rejected by many financial analysts familiar with local government accounting.60/ The problems discussed in the IMF report can be addressed by much less drastic changes than a switch to cash budgeting. (f) Creation of a Local Government Service Commission and secondment of chief officers between councils. Chief officers were brought under the jurisdiction of the Public Service Commission in 1984. In addition, the IMF report made similar suggestions for the smaller town and county councils and suggested a more equitable revenue sharing system for agricultural cesses. 4.38 In general, therefore, we see that the major thrust of the IMF recommendations was to establish a clearer definition of local authority service requirements and to ensure that each service was matched by an appropriate and adequate revenue source. Where a service was purely local in nature (e.g., refuse collection, markets, water supply), the local authority was expected to fund the service either from local revenue sources (e.g., rates) or from specific user charges on consumers of the services (e.g., market fees or water charges). However, when the local authorities provide services mandated by the central Government or where the central Government chooses to restrict revenue sources on social grounds (e.g., school fees), then the DIF report recommended as a matter of equity that the central Government should provide funding for the service through a cost-reimbursal grant to the local authorities. The clearest example of this principle is in the case of primary education. The central Government extended the provision of primary education nationwide as a matter of social policy, and it establishes teachers' salaries in all jurisdictions, including those where the local authority administers the school system. In this case, providing a grant to cover teachers' salaries 60/ While there are good financial reasons for maintaining an accrual accounting system for municipalities and town councils, there may be a valid case for introducing a simple cash accounting system for urban and county councils because of their lack of qualified staff. -55- is a matter of equity (i.e., as the local authority is required to provide a service funded by the central Government outside of the major urban areas) and as a principle of fair taxation (i.e., as urban residents pay taxes to support primary education outside of urban areas, using local revenue sources to pay teachers' salaries in urban areas would amount to double taxation). While the central Government continued the principle of cost-reimbursement for teachers' salaries during the latest revision of local authority grants, this principle was not maintained for other centrally mandated services such as drugs and dressings (required to be provided free through local authority health centers) and school fees (which were used by the local authorities to fund recurrent expenditures other than teachers' salaries). In general, therefore, the central Government has not followed a consistent approach to local authority revenues, particularly with regard to centrally mandated services, and this constitutes an important impediment to establishing an equitable sharing of revenues and responsibilities between the central and local levels. 4.39 Possibilities for Improvements. There are no magic recipes for improving local authority finances in Kenya. This report has made several for suggestions for improving the revenue yield of existing taxes. In addition, it may be worthwhile at this point to update the DMF report in order to establish a renewed basis of agreement for the sharing of revenues and responsibilities by the local authorities. The following suggestions for topics could provide an agenda for a revived discussion of local authority finances in Kenya. 4.40 Virtually all the previous studies of local authority finances have neglected questions of cost-efficiency and coverage in the provision of local authority services. Local authority officials claim that services are being cut back and maintenance neglected. However, it is difficult to document these trends in a way which would allow public discussion of whether these cuts can be justified in the face of financial restraint at all levels of Government. In the absence of reliable indicators of output, local authorities are open to charges of waste and inefficiency which they have no basis for disputing. It would seem clear that any discussion of local authority financing in Kenya should begin with a consideration of output, both over time and across different local authorities. This would have to be more than an extrapolation of expenditure trends, and should include the calculation of physical performance measures (such as students per teacher, gardeners per park, average response time per fire, administrators per staff member, etc.). Only with this factual basis can a reasonable discussion proceed about the future course of local authority services in Kenya. 4.41 The second area which needs investigation concerns the existing revenue sources, including user fees and license charges. Simplified methodologies are available to ensure that user fees cover at least the full cost-recovery price for the services, and the various license fees and other charges need to be classified and summarized to see if additional revenues can be generated from them. While it does not appear feasible to introduce improved site value rating for property taxation at this point, serious consideration should be given to various indexing systems to adjust property values on an interim basis between the five-year revaluation -56- periods. Finally, particular effort should go into proposals to improve the collection efficiency for existing revenues. 4.42 The area of local authority cash management, financial controls, and accounting procedures needs to be improved. This does not require the radical changes proposed in the IMF report, but rather could rely on modifications and codifications of existing procedures to establish consistency across local authorities. Preparation of an updated accounting manual for local authorities would appear to have high priority, and this effort should be coordinated with the Kenya Institute of Administration which would be responsible for instructing local authority staff in its use. Finally, the budget estimates for local authorities should be modified to ensure a better correspondence between cash and accrual budgets, in particular by requiring adequate provisions for working capital balances in anticipation of revenues and by formulating procedures for allowing the local authorities to write-off uncollectible debts. 4.43 On the basis of these studies, it should be possible to make a reasonable assessment of the options for improving the performance of the revenue and expenditure systems currently available to the local authorities. This will raise the related issues of: (i) what level of services should the local authorities be expected to maintain, and (ii) what additional revenue sources will they need to sustain this level of activity. No matter what technical improvements are made to the existing revenue base for local authorities, it is unlikely that they will be adequate to provide the long-term elasticity in municipal revenues required to cover the effects of inflation and continued urban growth. This will mean that additional revenue sources will be needed, and this can only be accomplished with central Government support. In particular, a revised agreement on centrally mandated services and the designation of adequate revenue sources to support them, whether through Government grants or improved sources of taxation, would appear to be a prerequisite for a satisfactory resolution of the problem of local authority finances. 4.44 In addition to the subjects discussed above, it is also suggested that the proposed study should include the capital financing of local authorities, a topic virtually neglected in the IMF report. The next section of this chapter provides an overview of local authority capital financing in Kenya, as well as suggestions for improvement. C. Capital Financing of Local Authorities 4.45 The two major sources of capital financing available to local authorities are the LGLA for general purposes and NHC for housing programs.60/ Reference was made in the previous section to the original purpose o7fthe LGLA, which was to serve as a repooling fund for surplus capital funds generated by the local authorities. At the present time, the 60/ While there are good financial reasons for maintaining an accrual accounting system for municipalities and town councils, there may be a valid case for introducing a simple cash accounting system for urban and county councils because of their lack of qualified staff. -57- LGLA exists as little more than a conduit for channeling funds from the central Government development budget to the local authorities. Nevertheless, the LGLA has wide legal powers under the Local Government Loans Act61/, and indeed the LGLA could be constituted as a municipal development bank under the existing legislation. 4.46 As indicated in Table 4.5, the loans raised by municipal councils for capital investments grew at a rate of 13% p.a. between 1972 and 1981. Considering the effects of inflation and urban population growth, this rate of growth in capital financing is clearly inadequate, especially as urban development is taking place in a larger number of urban centers throughout the country. Because of the macroeconomic constraints on the Government's development budget, any improvement in the amount of urban investment financing will probably take some time. When it does, the LGLA could play an important role in allocating the available funds and ensuring better standards of financial appraisal for proposed investments. 4.47 Various proposals have been made that LGLA should use its powers more broadly to raise capital funds for urban investments from domestic capital markets and international lending agencies.62/ Given the financial weakness of the local authorities, however, it is doubtful that private financial institutions would lend to LGLA without a Government guarantee. As for quasi-Governmental agencies such as the National Social Security Fund (NSSF), the LGLA would be competing with the Government for development funds, which are currently allocated directly through the Government's development budget. Finally, it is doubtful that any significant surplus capital funds can be generated internally by the local authorities unless major changes are made in the structure of local authority recurrent financing. In essense, it may be difficult for LGLA to expand its penetration of domestic capital markets, although this can be kept open as an option for the future when the economic position of Kenya improves. 4.48 Where the LGLA can play a potentially important role is in the financial review of proposed investments, especially their relation to the overall financing plan for the local authority. Referring back to Charts 3.1 and 3.2 in Chapter III, an important step in the proposed investment planning process is the financial review of proposed investments and the granting of a formal loan sanction. In present practice in Kenya, the LGLA is responsible for granting loan sanctions for proposed local authority investments, and local authorities must submit an application specifying, inter alia, that the project satisfies the following criteria: - Council approval of the project and authorization of the loan request; - Compliance with town planning and land use requirements; 61/ Laws of Kenya, Chapter 270, revised edition 1962. 62/ See, for example, "Finance for Housing Infrastructure," by J.D. Hagger, HRDU research paper, University of Nairobi, 1978. -58- Conformity with priorities set by the National Development Plan, the District Development Committee, and the Ministry of Local Government; Ownership and availability of land where necessary; The local authority concerned is not in default with respect to outstanding loans to LGLA; The scheme is-technically sound, cost estimates are realistic, and proposed implementation arrangements are appropriate; The scheme is financially viable in that income generated from the investment in the form of rents, fees, or charges will be adequate to cover loan charges (principal and interest), maintenance, insurance, administration, rates, rents, and other direct costs; or - If the scheme will not generate adequate income (such as a road project which will not recover its costs from frontage charges), the local authority has adequate general revenues to cover the capital and recurrent costs of the project. 4.49 These criteria appear to be adequate to allow a technical evaluation of the proposed investments. However, most loan sanctions have been granted almost routinely by LGLA, with the crucial decision on project approval being the actual allocation of funds. Given the general scarcity of urban investment funds, it would be desirable to improve the screening of urban investment proposals as to their technical and economic merit in order to ensure the most efficient distribution of the available funds. LGLA also has paid little or no attention to identifying the impact of an investment project on the overall financial position of a local authority. There is clearly a need for LGLA to issue guidelines and develop procedures which would require an analysis of the required increases in rates and/or in the level of fees and charges which would be required for the local authority to maintain a proposed investment and cover the debt service charges. This type of impact analysis is a crucial step toward ensuring that the investment planning process described in Chapter III takes adequate account of the financial capacity of individual local authorities. It would also provide an incentive for local authorities to improve their recurrent revenue position, as the financially stronger local authorities would be given higher priority in the allocation of additional capital funds. 4.50 There are various ways in which the review of local authority investment proposals can be handled within MLG. The Development Department with its core of technical staff is the logical place to screen individual investments on their technical merit and provide input to LGLA. LGLA, because it has the responsibility for granting loan sanctions, could be upgraded to provide a financial appraisal of the local authority's overall investment capacity. This will require a strengthening of the staff of LGLA and technical assistance in the development of financial forecasting -59- procedures. Nevertheless, it is a necessary step for MLG to move toward a capacity for allocating and overseeing urban investment funds on a spatially extensive basis. V. HOUSING POLICY: INSTITUTIONS AND FINANCE 5.01 Housing policy has been an area of particular concern to the Government of Kenya, and Kenya has been a leader in articulating a housing policy appropriate for developing countries. This attention has been demonstrated by events such as the establishment of the Housing Research and Development Unit (HRDU) at the University of Nairobi and the movement of the United Nations Center for Human Settlements (HABITAT) to Nairobi during the late 1970s. Kenya has also been the recipient of substantial financial assistance from multilateral and bilateral donor agencies for shelter-related projects, including the World Bank, USAID, the European Development Fund, and the Commonwealth Development Corporation. 5.02 Despite the attention which has been paid to housing policy, there is a general feeling in Kenya that the performance of the sector has been much less than satisfactory. The public housing sector has produced only a fraction of the units called for in the last two National Development Plans, even allowing for the fact that the original targets were overly ambitious. Illegal, unplanned, and over-crowded slum settlements continue to increase around the cities and towns in Kenya. There are unresolved differences on the question of appropriate standards, which has contributed to such problems as the inability of Nairobi City Council to deal effectively with the slum settlement of Mathare Valley. These difficulties have been analyzed in a number of reports, and the technical deficiencies of current procedures are well-known.63/ On the other hand, the remedies which have been proposed often do not take adequate account of the institutional environment of the housing sector and the incentives which exist to promote change. For this reason, this chapter will focus primarily on the financial and institutional structure of the housing sector to determine if it is consistent with the needs of housing policy in the economic climate of the 1980s. A. Performance of the Housing Sector 5.03 Indicators of Current Performance. The 1978-83 Development Plan identified the shortfall in housing units located in urban areas as 140,000 units, with an annual addition of 30,000 units needed to provide for population growth, thus giving a total five-year requirement of 290,000 63/ Kenya Shelter Sector Study and AID's Experience, Office of Housing, U.S. Agency for International Development, August 1979. -60- units.64/ This level of investment in housing is acknowledged to be infeasible, as the current 1984-88 Development Plan identifies a total housing program for the five-year planning period of only 59,500 units, which includes rural public housing, upgrading of the existing housing stock, and private sector construction. Thus the public and the formal private sectors were expected to provide only one-fifth of the total required units, and less than half of the units needed just to provide for new urban growth. Implicitly, it is assumed that the remainder of the urban population will have to rely on the informal sector for their shelter needs. 5.04 In order to evaluate any proposed targets for the housing sector, it is necessary to have some information on the sources of supply for the housing stock. In this regard, the available data are fairly limited. Most official statistics cover only a portion of the housing which is constructed, such as publicly-owned housing. The Central Bureau of Statistics (CBS) is planning to conduct an expanded survey of the housing stock which would do a great deal to fill this information gap. In the meantime, the data from a 1979 CBS Rent Survey in Urban Areas provide probably the most accurate coverage of the housing stock that is currently available. 65/ Table 5.1 provides the percentage breakdown of the existing housing stock in urban areas by ownership pattern. We see that over 80% of the urban housing stock is owned by the private sector. The remaining housing is made up of local authority housing, which consists of 64/ Even this figure may be an underestimate of the additional housing required for urban growth. Using an average urban growth rate of 7% and an average number of persons per dwelling unit of 4.16 (from the 1979 Rent Survey of Urban Areas), the annual requirement for new housing in 1980 would be close to 39,000 units in order to provide for urban growth. 65/ See "Report on the 1979 Rent Survey in Urban Areas of Kenya," Central Bureau of Statistics, January 1981, and "Experience of Undertaking Rent Survey in Urban Areas of Kenya," by P.O. Tindi, reproduced in "Report of the Seminar on the Role of the Private Sector in Housing Development in Kenya," Ministry of Urban Development and Housing, September 1-3, 1980. The 1979 rent survey covered ten major urban areas including Nairobi and Mombasa. A distinctive feature of the survey was that it was the first attempt by CBS to include coverage of informal settlements on a fairly comprehensive basis. While the results of the survey are broadly consistent with other data, there are large unexplained discrepancies between the city-specific population projections of the rent survey and the population counts of the 1979 census for individual municipalities, particularly for Nairobi and Mombasa. It is presumed that these discrepancies reflect problems in obtaining adequate survey coverage of urban squatter areas. Since the published results of the rent survey are in aggregated form, some of the results presented here have had to be extrapolated from the tables. -61- staff housing and rental units mainly constructed prior to 1975, and central government staff housing, with Kenya Railways Corporation being one of the larger holders. The rent survey also estimated that 18% of the housing stock (23% of the private stock) is shanty-type housing in informal developments. This is probably an understatement of the amount of sub-standard housing in urban areas, which other surveys have estimated to comprise 25-30% of the total housing stock. 5.05 Table 5.2 provides information on the type of structure and tenancy status of the private housing stock. One of the most striking results is that the implicit goal of the Government housing policy, single families occupying their own separate house, is achieved by only 4% of the occupants of private sector housing (3% when public sector housing is included). In fact, less than one-fourth of private sector housing is owner-occupied, most of it in conjunction with other households under the classification "other housing." Over three-quarters of urban households are renters. This is consistent with experience in the Dandora site and service project in Nairobi, where two years after allocation of plots, two-thirds of the residents were tenants and only one-fifth of the plots were occupied by a single household. Table 5.2 also indicates that most Kenyans have not been able to achieve the goal of having a two-room dwelling unit, as the average number of rooms occupied by the residents of "'other housing" (82% of all residents) is 1.4 rooms per dwelling unit. These data imply that Government planning guidelines which call for single-family occupancy are unrealistic in the present very tight housing situtation. 5.06 The fact that most urban housing in Kenya is rented is now generally accepted. Planners of site and service projects design the infrastructure to handle much higher plot densities than would be needed for single family occupancy. The proto-type house designs recommended for self-help construction are adapted for subletting by the inclusion of a wing of single rooms. It has also been found that lowering costs enough to make a site and service plot affordable to the lowest third of the income distribution requires standards that are generally unacceptable to the local authorities. In order to maintain accessibility for the lowest income group without compromising financial viability, the expected income from subletting is often included with other income in defining the threshold of "affordability." All of these measures are pragmatic adjustments to make the projects conform more realistically to the current housing market. Moreover, they represent a substantial change in attitude on the part of the local authorities, as compared to an example in the late 1970s when Nairobi City Council refused to approve an NHC project at Kibera on the grounds that the proto-type house designs made explicit provision for subletting. 5.07 The 1979 rent survey also estimated that the number of dwelling units in urban areas has been increasing at a rate of 8.1% p.a. over the period 1975-79, reflecting an average increase of 9.9% for the private -62- sector and 1.4% for the public sector.66/ This is compatible with data on the value of residential building plans approved by the different municipalities (shown in Table 5.3), which show a sharp increase over this period in approvals of private residential building plans, particularly by Nairobi City Council.67/ At the same time, the value of approved public residential building plans was no more than a fraction of the value of private building plans and showed little growth over the period 1974-82. This result confirms that the main deficiency in performance in the shelter market has been with the public sector, rather than the private sector. 5.08 The final indicator of the housing market is the behavior of rents. Table 5.4 presents rental indices for Nairobi stratified by three income groups. From these data, we see that increases in rental prices for upper income residents have generally matched increases in other prices over the period 1974 to 1982, while rentals for the middle and lower income groups have outpaced other price changes. The lower-income group in particular has faced increases in rents which significantly overran increases in other prices. These price changes indicate that the housing sector has been able to maintain an adequate provision for the upper-income group, but that the middle and particularly the lower-income groups have suffered a deterioration in housing conditions over the past several years. This reflects the relative dominance of the private sector in serving the upper-income group and to some extent the middle-income group, while the lower-income group must depend on the public sector or the informal private sector to obtain its housing. 5.09 The picture which emerges from these data is that of a well-functioning private sector providing housing for the upper part of the urban income distribution, while the lower-income group must rely either on subletting of existing housing, informal housing development, or public sector agencies. The poor performance of the public sector is not unexpected, as the period which these data represent was a time of rapid change and experimentation in public housing policy. Nevertheless, the data do support the dissatisfaction of many Kenyans with the urban housing market and the need to obtain improved performance to meet the need for urban housing resulting from rapid urban growth. 66/ These estimates should be considered as order of magnitude only. The focus of the study on dwelling units rather than structures may also conceal an increase in crowding of the existing stock. 67/ The value of building plans is considered to be a more reliable indicator of activity than the number of approvals, as a single building plan can cover a number of houses, such as for an estate. On the other hand, many approved building plans are never constructed or only after a significant time lag. The CBS estimates that only 56% of the plans approved between 1975-79 had been completed by the end of 1980. -63- 5.10 Lessons from the Bank's Involvement in Kenya's Housing Sector. The Bank has been involved in Kenya's housing sector since 1971, starting with technical reviews of the recommendations of the Nairobi Urban Study Group, a UNDP-funded project which was commissioned to provide a long-range development plan for the city of Nairobi. Because of suggestions by several Bank missions, the study was broadened from a concern with land-use and transportation planning aspects of urban development to include consideration of affordable policies for urbanization of land and housing development in Nairobi. The result of this effort was one of the earlier site and service projects to be financed by the World Bank, the Dandora project in Nairobi which was presented to the Board of Directors of the Bank in 1975.68/ The objective of this project was a demonstration of the feasibility of large-scale involvement in site and service development. Planning was made for 6,000 plots to be developed at a site on the eastern side of Nairobi, with the inclusion of materials loans to individual allottees, on-site construction of community facilities such as schools, health centers, and markets, and innovative approaches to physical design in order to reduce the cost of plot development.69/ 5.11 Despite official endorsement of the site and service concept starting in the 1974-78 Development Plan, there was by no means universal acceptance of the project concept in 1975. A number of public officials stated reservations about the project, arguing that it would degenerate into a slum or be taken over by higher-income groups. Moreover, the project represented a radical change in direction for NCC, which had previously emphasized the construction of higher-income tenant-purchase housing and rental units combined with wide-scale demolition of squatter settlements.70/ 5.12 In retrospect, the Dandora project has succeeded in accomplishing almost all of its original objectives. Experience with the project has demonstrated the validity of most of the assumptions associated with the 68/ "Kenya: Appraisal of a Site and Service Project," World Bank Staff Appraisal Report, April 14, 1975. 69/ The objectives and features of the Dandora project have been documented in "Sites and Services: A Strategy for Kenyan Urban Development" by R.M. Beardmore, master's thesis in city planning, Massachusetts Institute of Technology, 1978. 70/ The demolition of squatter settlements reached a peak in the early 1970s and had virtually ceased by the time of the Dandora project. As part of the loan agreement with the World Bank, NCC agreed to refrain from further demolition unless a viable alternative was made available for the displaced residents. -64- site and service approach. In particular, the monitoring studies of the project confirmed that:71/ (a) Consolidation (completion of individual housing units) can take place fairly rapidly after sites are allocated. For Phase I at Dandora, 70% of plot holders had completed the minimum required housing construction within 18 months after construction was permitted, and the sites on which construction had not started were held mostly by members of self-help contruction groups which had not been able to begin on the sites. (b) If enforcement measures are rigorously applied, arrears on plot charges can be minimal. In March 1981, less than 4% of allottees for Phase I in Dandora were more than six months in arrears.72/ (c) Site and service projects can be effective in reaching lower- income groups. The target group for the Dandora project was households in the 20th to 50th percentiles of the urban income distribution, and the monitoring studies have shown that most allottees did in fact fall within this range. 5.13 Based on the experience of the Dandora project, the Bank approved the Second Urban Project in Kenya in 1978. This project introduced very ambitious objectives for the housing sector. On the physical side, the project included both site and service developments and upgrading projects in three cities: Nairobi, Mombasa, and Kisumu. As compared to the Dandora project which was located on city-owned land, the Second Urban Project included a variety of sites with land tenure ranging from government ownership to private freehold land to communal trust land. Innovative features of physical design were specified, including very low standard settlement plots for the poorest households, and a program for family health and nutrition was included in all three project towns. Finally, the Second Urban Project committed the Government to a far-reaching program of institutional reforms, including basic changes in the structure of municipal finance, strengthening of the MLG and the NHC, revisions of housing bylaws to reflect the needs of low-income housing projects, and preparation of an urban investment program which would reach out to the secondary towns of Kenya. 5.14 Unlike the initial phases of the Dandora project, the Second Urban Project has suffered from significant delays in implemention which demonstrate weaknesses in the administrative structure for site and service 71/ "Monitoring and Evaluation Study of Dandora Community Development Project," Senga, Ndeti and Associates, various reports dated December, 1977 to February, 1980. 72/ By May 1984, the percentage of allottees in arrears by over six months had risen to over 25% because of subsequent problems of enforcement. NCC started a phased program of evictions in late 1984 which was expected to reduce arrears to managable proportions. -65- projects. The first lesson is the importance of sustained political support within the local authorities for the proposed investment program.73/ For this reason, Chapter III emphasized the necessity of involving the local authorities at an early stage in the urban planning process in order to build political support for the projects. A second lesson is that the process of land acquisition in Kenya consitutes a major barrier to current procedures for implementing public sector housing projects. The appraisal team for the Second Urban Project examined the process of land acquisition for the project sites in depth and clearly documented the steps which would have to be taken to resolve the land issues. Nevertheless, four years after project appraisal, land issues continued to be a constraint on some of the project sites. A third lesson from the Second Urban Project is that the process of obtaining approvals within the local authorities can be very time consuming, and the process is further hampered by the persistent shortage of qualified techical staff. The Second Urban Project attempted to circumvent the problem by consolidating the required staff within a Housing Development Department74!, but the Departments have proved to have high overheads without significantly improving the speed of implementation. Finally, problems arising from the persistent shortage of recurrent revenues for the local authorities have proved to be a continuing problem in all of the towns. 5.15 Because of the delays in implementation, cost overruns due to inflation have forced the cities to scale back their intended projects. Recent experience with the Kayole site and service project in Nairobi indicates a cost in 1982 for construction of on-site infrastructure of approximately KSh 17,000 (US $1,350) per plot, as compared to an estimated cost during project preparation in 1978 of KSh 6,600 per plot (US $800 at the previous exchange rate), representing an increase in cost of almost 160%, primarily because of project delays and changes in standards. When the additional costs of implementation are added to the cost per plot (administrative overheads, materials loans, interest during construction, etc.), the total plot charge is beyond the affordability of the original intended target group. Moreover, investment costs of this magnitude represent a significant drain on Government funds at a time when the funds available for urban areas are becoming increasingly strained. It is clear that emphasis in future projects will have to be given to interventions with shorter gestation periods, faster recovery of costs, and possibilities for utilizing other sources and types of financing as much as possible. 73/ The Second Urban Project had unusual problems in this regard, as all of the project cities had their elected councilors dismissed at some point during the period of project implementation, with the councils being run temporarily by Government commissions. 74/ The concept of a separate Housing Development Department to coordinate and implement low-income housing projects was recommended in a study of NCC's housing operations funded by the Bank ("Nairobi's Housing Needs: Meeting the Challenge," Coopers & Lybrand Associates Ltd., 1976, three volumes). -66- 5.16 Summary. Despite the major innovations in housing policy which Kenya has introduced in the past decade, the performance of the sector is still unsatisfactory. Recent experience has demonstrated that Kenya's housing policy is at variance with current conditions in the sector in at least three areas: (a) housing policy in Kenya has emphasized public sector interventions, despite the fact that the private sector provides the bulk of shelter for most people in urban areas; (b) housing policy has been based on the promotion of home ownership, despite the fact that the great majority of urban residents are renters; and (c) Public housing projects have not demonstrated a capacity for efficient and sustained housing production at a cost generally affordable to the lowest income groups. In the next two sections, we provide an in-depth review of key factors which have inhibited the performance of Kenya's public housing sector, with section B focusing on urban land policy and section C on housing institutions and finance. Section D summarizes the major conclusions for changes in the sector. B. Urban Land Policy 5.17 The review of the Second Urban Project identified land issues as a major constraint on public housing policy. Up until now, most housing projects implemented by NHC and the local authorities have been on Government-owned land allocated by the Commissioner of Lands or on land held by the local authority. However, the reserve of Government and local authority owned land suitable for low-income housing projects is rapidly dwindling in many urban areas, and the expected increase in the urban population will put severe pressure on the stock of Government land which is remaining. Continued dependence on public land could relegate public housing policy to a minor position in the shelter market. 5.18 Urban Land Requirements. Land in Kenya can be held under three broad categories of rights or types of title: (a) Customary rights, which have always been enjoyed by tribal people under their own laws. These are normally based on communal ownership and sharing, and the land is not formally demarcated into plots. Such land is now known as Trust Land, since it is held in trust by the county councils for the respective tribes. Experience under the Second Urban Project with upgrading of trust land in the Migosi and Manyatta sites in Kisumu indicates that the process of regularizing land titles can be complex, slow, and expensive under present procedures. (b) Freehold land, being land to which an individual has obtained absolute title from the Government. There are some statutory restrictions on the use of such land, for example the Land -67- Planning Act, the Land Control Act, and the Land Acquisition Act, apart from the local building codes as applied by the local authorities. Although the transfer of agricultural land is restricted by the Land Control Act, there is no such restriction in respect of urban land. (c) Leasehold land, which is held for a fixed-term from either the Commissioner of Lands or the local authority. The normal term is 50-99 years for urban plots and 999 years for agricultural land, although other terms can be granted, including a temporary occupation license of 33 years. To the above formal classes one must add squatters' rights, which are de facto rather than de jure. 5.19 Low-income shelter projects in the past have relied on publicly owned land or the compulsory acquisition of private land for development by public sector agencies. This method follows well-established laws and procedures, but experience has shown it to be exceedingly time-consuming. Because the pace of urbanization cannot stand and wait on such a cumbersome process, it is not surprising that a variety of de facto procedures have evolved in its place. Chart 5.1 provides a description of the range of methods used for urbanizing land in Kenya. These procedures range from legal to informal (informal subdivision of company-held land) to illegal (squatting on private land). The main positive qualities of the non-public initiatives are: (a) Speed of execution. The land-owning company or the individual squatter does not take long to decide on the location for a building, to assemble the materials, and to commence construction. (b) Flexibility in the sequence of steps to be taken. For instance, one can build before laying the services, or one can occupy the structure before it is completed; transfers can take place before survey; planning may occur after structures are already up. (c) Flexibility in the standard of services provided. The quality of infrastructure can be improved gradually as resources permit rather than as the law will allow. (d) The involvement of the community in the choice of options they can afford and execute. When a private developer is servicing and subdividing the land, he can study the market and aim the product at a specific clientele. Perhaps more importantly, given the current shortage of Government investment funds, these methods have the advantage that they rely on the mobilization of private funds to make shelter available for urban residents. The primary disadvantages of these approaches are that they may not provide a minimum level of basic needs services, or that the quality of planning and basic layout may be so poor as to make later improvement of the land and installation of services very expensive. Chart 5.1 ALTEONATIOf LAND ____ OUN POIVAS IN IVW L P U g 1. T C A G e N C I E 5 | P R I V A T eG C H U P S t R I Y A T E I x DI V I D U A L S Pusni.C SUBDIVISIONSS GassPAFUANML ROUPOOLING SCIHOU ISOLATID SDiTTRV (INCLUINIlG S/S) SLUM UPGRAIING AGCoC SYUArTERICOLOSY CD1fmmI3Y COOPS & CO'S UROAN LIDI OCINS MIS PAITICIPANEI KIOSK GWNIR SUSURSAN LASUNRJK PRIlVATe 1.AJSLO Location On Kntosopsl lend Kol.iiog *qnstter Poor load by o1rtoe of lo trios... oa iota. UrIo E re. SoorhP btyategio oUSation- t- - Ida. of bigh-loo. e Vaf ot lead I. K.tb .t the nrban edgo sel_eloa ooe-tan,. ib.llty. fD. in the Higl-nd. P.p.olly *n the v_olthy oslaborheods. or.tablt-hd moe. to, of the t-rrtin, -eviaing. big lti.s otty _:ter. end neighborbood. tonvq eep..illy voter table or soil typo industrial area Uba.a. - llCill.plot. P-r *.allebldIgs ootetds Nalonbl ToDure SovetncnL orauqotrod Att,ps eade to t- .fl nd illg.1 999 Yet rgrtanl torl Fr ...d or sg. lo-lt1ral rost lond or fr. hold Public l.Dd ospeololly ehold or ab D ro.held r.atsied by Chooges IDd sItvetled to grant stNre to ho...- ettl_nsr on govores lt ... in corporos Iso ee; no rroofer to hld by abeet.e l1ad- to-d roev, public leasehold (99 yar)- Isadlord iIe eou 35-99 yar laee onesr- on preninoely or prineto od ea; f-r Inrea1ip ldlolduel plot Onnere/ lord; ..hb Ldilidoal opve spacee. rod so hbng. fter _ _b- baiId:r oq build- soner entae Truat or parabaced Ito. tshlte .bsra .110 llo * atqatrhe j:nhtt-ee sod oar dloietoo. eacpt for Lg only *ed pays P,toOse lnd attler fr. hold/ll.obld parbk; h abaogo In oLner-hip rAteud rt title 11t0* Pro IStseo 1. Prosr_lng 1. Plzanin/D g, 1. Settl.K Dt opn a eDDd S. Orfomsea oopprorde I. odiodole botid 1. Pore ooation of 1. fInd stabl I atIOr 1. PrItar" - bd,inino 1. Alloate butliang 2 A..bI/AqIi.r lasnd 2. Progr_/bdgtitng oegeIotd load s_r o. otaop 2. Occupr and/r oblet pInt/hoe. oNer 2. PIbllessoorts pl sIte 3. Prp are Pl -looute 3. Dlspoieeeeoo & eatablihbd neIghborhood 2. Prohase. a.D 3. Orga.ie aoop or .-P 2. Pr.par layot on 3. SiId 2. lnllt fr app. l 2. Ilottee .obeit. 4. Lay aerviaee envolosen1 t of or n_r gr.htb poit. 3. SlbdtVid into plo.e 4. Parobon lead fin reponlteg/rellet- 4 Occupy (strt baLnoa by Cgorll & Pl for appes vP 5: S-rv yley d-lo- I ledonetr 2. tolvc oods for env 21 i. op tn l:a ..er ea bode of 1100k) C<_I .onr of by l_dlerd ed S. Allottoase h. Seraalsa ssttlars e.g. r eee 55 o 6In 5. Gtop prepare. donulnp- 3. D-.rate.s/rvy S. Seek ceaaol ItosNe- L-ode C.-oIIl 7. BoIldiog parit. A . Cohty relstleas featpatbe 4. =Io ebre to cet ple oitb hlp ef ne plot 3 Srvy 3. Allott- haild aOs N-6. 3-tvey/ 3. Coneolidete *-ttl- so indlivdnol tOD-ultante or gonro- 4. lealloct so *t- 4 INrbet plsta soope . Settl Dt r.gl-tratlna 4. lselop politicl Idoatity plot esNt ble bJe 5. o-yer pply for dor *eblet 5. Develop 1 ekt 5. =hers batld vd 6. Mo-oy ouSht for blIding p.rite 4. C-tl lays eoacvv or *ubl-t oosrsdtg esaxirs0 Thre.holds - Pinoc - Co_olty *pptov-l - Offialal reoogoitito - Ideatifying a soIt- - Con-tn-ina o-or to seli - rlging o.ars - koqolrfog ealtoble Ies - Sttitteg -aity Rolala IBad * d Pisoeag approvaI - Iaod -aquit:tlso/ - SeinIng t.bility & able od -nalloble - leisstilg load aqolsi- toethr - Galsieg p0bll seirviee aVoldidg ctwplty - Allocettios dloltitloa pooonosd ftrP tinO - - .etesleia * b es acc-ptdoc. - Till ere y robose - gesistrallsn/Te.sore -. Cetting .r dit-bl ftor sharing co_te leaders sod r_lloc taltoI - Istro " tI -d Pinoriag Go-rsoet _ Sere ot - Prirate saloge _ Savins Savisge PtraDI otlgse - Ccrtl.l soon.. All.tte.. pep etuwd -oal authorIty - Locol sothority - Isfo L oelders B Seperasiv ok - Co-rl-ll rtob-a laurcee k tiosk ds-nlaPsrs prduo"I - NWC loao - SOC - Self help - C-.ert,l Saak. - Self help - Senor *ageeceC - iosaor ageaotse - .6sicaltorO1 CredIt - Self-bele0 Te Stal 2-5 yers. deteeDinod ni.i 3 year. gpeed . -rIticrl Patriy qook POIrly qotok =lo pene-e dapoadleg QOtLk pre.enco - -eetitI 1-2 y-rs; delay. Prea--s ltly Opet by h-r-ueretI deloys dtlas. aa eed by edjodi- soco os to *-tablieb credibitity cBead _ oly by plea- If Is tigt tedatl I atitr & .cqoisitLo iny ppra-l sod lItle Horv s Ieprdlesotr Settiog apprval so L.ok of sperlen. Offictnl oppneLtio. SabdivLeLo- Is illegal 1 aildioa Code Leak of epsf -bal- Rot1dtne aod- Sefrertora Sostillsy a_rd- standards, Land Pti itnl con.Idora- e gal disapprov l isj I- & ptoiedoro. P.binc b_ltih lsv nd rdo h e ltn dlerd.5 naqoistlos tlon- Poeslble vdsr edjadi- LoDd osoorabip steno- c tles & loca g oveas- tore. *vnt lao. E.ples NNC/iSS Prjats Kte opg-rdi.g ath.r. URUAN III sit.s It hth.r. Vlloy & Chge_e eponllg 11ks In ,is Irobi od 4 rinats sdabdii.oie Ki-soot -ad Cb_e- DiDder sits gib.r. Eldore. t. . "akDaji, reabk/Srloboagl Sch-, hoee heba ally itetee Is the lr-e to_. Is N1b . Al- Chae IMly-nd-l f bShatt YTet. *n SairobI sd sod..telel -r_ 'ilrlry ;ill layeas. sWretgt EosIOtin feede roasseesaHot Offere great probsitll Pa..le. .ctIon a big Sand potentlal far Sand potH btltl for Joint sa beo negleatd. 1 o-attl fwr paooetlel Se On1 hat neods to hsrsr . onrae that nssd soplagltne prinais loltietlne. gnonstoo Creet poteagial for orben grevib Urben dentiop- flard. eaptoislty tappIng ond rhannelinsg *titislIoe progrea shoold ha cooi{y hasod Pohita soctor nesde Fest tnol teaure. protrotton of In the rr1otrd enpended action. see to be toon *plarG era teOtff relevd nt dic tloGs doct pid. reiL d L- nbatlv 2 ye of bt eaeloatee sod and trid. I prote tarkly vith plale lnto_eDt of teahnlque and sater tn aahben seeolcta ororedorse. eclonal 1soa KaJon legol Soildlog Code and Pablit N_olth Los; Trespass and Noi..na Lad Control cAt; btse a. for Freing Se- lepodiest. Tra.p. --; .Pbic l lsIth; L nd Cootrl At sod Seldlg Cods; Unrdl .e ublgo H_ralIE egietr Lad Art L.e.; P r isoH th ol-; H ilentag Act; Local CooPers. Ve nd Buildieg Cods; W.I-sDte Land Plaeosg Act Loonl Ososesmot bol diig Coda G- -oerat Act Cooe lane ACt -69- 5.20 The relative magnitude of the various urbanization processes can be roughly estimated from existing information. Over the period 1981-85, it is estimated that approximately 30,000 residential "plots" were required annually in urban areas of Kenya. Table 5.5 shows the expected source of new residential plots, based on the current performance of the public sector in approving subdivisions and allocating land. From this, we see that public sector housing projects can be expected to provide only one-tenth of the required plots at current levels of output, and the total amount of formally approved residential land subdivisions (both public and private) will amount to less than one-fourth of the required number of plots. 5.21 Although various informal processes provide a large proportion of new residential plots, the formal processes still account for the majority of land which is urbanized every year. Table 5.6 presents the estimated annual land requirements for Nairobi and for all urban areas in Kenya, based on current settlement densities and the prevailing ratio of residential to non-residential land (estimated to be approximately 1:1 based on studies by the Department of Physical Planning). Urban growth will require approximately 2,000 additional hectares of urbanized land per year in Kenya, of which about one-fourth will be provided by informal land development processes.75/ 5.22 Besides the time consuming nature of the approval process, many developers choose to circumvent formal approval procedures because of the insistence by local authorities that services must be provided at "adoptive standards." Each local council has its own adoptive standards, most of them based on rigid physical specifications inherited from the colonial period. For example, the adoptive standards required by Thika Municipal Council include: 7.0 and 5.5m. paved carriageways built to the council's structural specifications, open lined drains and/or piped drains for stormwater drainage, 2m. wide footpaths on either side of the carriageways, and street lighting using the council's standard lanterns. Little opportunity is provided for the developer to gradually introduce services as the area fills up and demand justifies the investment. This means that the developer must invest a substantial amount of money up front in developing the area and recoup his investment gradually over time as the plots are sold. Given the high interest cost of investment funds and the limited availability of capital resources, it is not surprising that many developers prefer to forego the requirement of adoptive standards. 5.23 As a solution to the continuing spread of unplanned housing developments in the urban areas of Kenya, it has been proposed that the approval process be strengthened to prevent private developers from circumventing the building regulations and that the requirements of 75/ Another way to arrive at this estimate is to assume that the urban population will continue to increase at a rate of 7% p.a. and that each additional person will need an average of 140 m2 of urban space, which is somewhat smaller than the 1969 actual urban density of 160 m2 per person. -70- adoptive standards be modified to reduce the cost of compliance and encourage more developers to follow the formal procedures. The Kenya Low-Cost Housing By-laws Study has looked at the process of planning approvals and considered realistic alternatives to current building standards. Adoption of the recommendations of this report would be an essential step in ensuring that the formal planning and approval process does not become increasingly marginalized in the face of rapid population growth and the high demand for urban land. Nevertheless, technical modifications of building regulations are unlikely to be a sufficient mechanism for halting the continuing creation of slums in the face of other constraints in the economic and institutional framework for shelter provision in Kenya. 5.24 Even if building standards and planning approval procedures are substantially relaxed, private land owners will almost certainly continue to create substandard developments because of a lack of capital financing for installing infrastructure. Existing housing finance institutions are geared almost exclusively to high standard developments for the middle and upper income levels. Moreover, legal barriers, such as the problem of shared debt liability among members of land cooperatives (which hold much of the urban fringe land in Kenya) makes this market unattractive for commercial finance institutions. In the meantime, the local authorities are the most reasonable option for extending financing for urban infrastructure on a scale which is commensurate with the requirements in the sector. 5.25 Priorities for Public Sector Responses. Existing public sector housing programs are insufficient to ensure the continuing expansion of basic needs services to the low-income urban population in Kenya. Regardless of any improvements which may take place in the implementation of site and service projects, it is clear that the scale of intervention will fall far short of the level of activity which is required. Nor will an expansion of conventional housing finance be adequate to resolve these problems, as the funds would almost certainly be used by higher-income segments of the population. The solution to this problem is not a massive increase in investment funds devoted to public housing projects. These funds are just not available in the current ecomomic circumstances, and the existing public sector agencies do not have the absorptive capacity to utilize a significant increase in resources. The only other option, other than continuing the growth of slum settlements, is for public sector to service privately-held land at minimum standards and on a large scale. 5.26 The advantages of servicing privately-owned land without going through the acquisition process are obvious. The private land owner is required to surrender free of charge the land needed to install roads and services, often amounting to 20% of the total land area. The use of public funds in leveraging the provision of shelter is maximized, as the private owner is responsible for housing construction using his own funds or financing secured on the private market. Finally, quicker methods of recovering public investment costs can be utilized, such as frontage charges for road construction. Other investment costs may have to be recovered through more indirect means, such as water tariffs and increased property taxes arising from changes in the land use classification. -71- However, since better procedures and safeguards exist for collecting water fees and taxes than for the plot charges used in site and service projects, the lower administrative costs and better collection performance may more than compensate for any loss in revenue as compared to plot charges. 5.27 In practice, there are a number of impediments to the implementation of this approach. First of all, it will require the local authorities to enter into voluntary agreements with private land owners. There are well-established procedures for obtaining these agreements (filing of a sub-division plan, approval by the local authority, registration by the Commissioner of Lands, etc.), but the process has generally been used for upper and middle income housing only. The emphasis on minimum standards will also be difficult for many local authorities to accept, given the physical orientation of most urban planners and local politicians and the lack of a clear linkage between financial and physical considerations in the approval process. Nevertheless, given the economic conditions expected to prevail in Kenya over the next decade and the restricted outlook for real growth in the investment funds available to urban areas, the only choices available for the bulk of the urban population will inevitably be between minimal standards and no standards. 5.28 One possibility for addressing the planning concerns of local authorities and the needs of different constituencies would be to introduce various standards of land-use control according to a graduated zoning system. For example, a possible four-tier structure of zoning could provide for the following standards of land-use control: (a) Free zones, where there is very limited control apart from ensuring the minimum standards of public health, especially water supply and low-cost sanitation. This type of land use would include existing unplanned housing developments and would seek to ensure that a minimum level of basic services is provided and perhaps exert some control on gross densities. (b) Minimum control areas, applicable in areas such as urban finge land. Minimum control would seek to ensure a regular layout so that services can be installed at minimum cost and that road reserves and open space are retained for surrender to the local authority when basic services are provided. (c) Moderate control zones, of the level now being exercised under the law, would apply to areas where resources are available to install urban services to an adoptive standard. A distinction could be made between areas where more permissive by-laws would apply, versus areas where the local authorities might wish to retain the stricter standards of the current Building Code. (d) Tight control zones, in which the community would wish to invest large amounts of money for purposes of environmental protection. Such areas would include the city center, pollution-prone industrial areas, areas of special attention to tourists, and architectural or historic monuments. -72- This system would formally recognize the prevailing standards of urban land development in Kenya, and as such it would provide a more realistic basis for the local authorities to work with private land owners in the extension of urban services. C. Housing Finance and The Role of Public Sector Agencies 5.29 Compelling as the preceding argument is on financial grounds, it is up to the political process in Kenya, most especially at the level of the local authorities, to recognize the constraints on funding and make provision for the servicing of privately held land. The central Government can assist by making funds available for minimum service projects while maintaining a limitation on the financing provided for higher quality housing projects. However, as the central Government and the local authorities have a long tradition of direct support for housing projects, it is unrealistic to expect that the Government will abandon its commitment to the direct provision of housing. 5.30 As noted in Chapter II, the Government provides financial support to housing development in Kenya in two ways: (i) by encouraging the expansion of commercial housing finance; and (ii) providing housing funds directly to public sector agencies to implement shelter projects for the low-income population. In this section, we discuss both types of housing finance, considering first the provision of commercial financing and secondly the role of public sector agencies in providing housing finance for the low-income population. 5.31 Commercial Housing Finance Resources. Kenya's housing finance sector is reasonably well-developed, with three large institutions and several smaller ones specializing in the sector.76/ HFCK was incorporated in 1965 as a partnership between the Government 7f Kenya and the U.K.-based Commonwealth Development Corporation (CDC). An interesting feature of both HFCK and EABS is that they are associated with real estate development companies. HFCK participated with CDC in the development of the Bura Bura Estates in Nairobi, a highly successful middle-income housing development, and it has also participated with NHC in some of its middle-income housing development. EABS is associated with Akiba Loans and Finance Ltd, a private company which carries out development of middle-income housing. 5.32 The housing finance companies experienced very good growth in their deposits over the period 1975-80. EABS reported a growth rate of 24% p.a. in the value of deposits and 12% p.a. in the number of depositors during this period, while HFCK showed a growth rate of 20% p.a. in the value of deposits. During the period around 1980, however, all three of the large institutions experienced an excess demand for mortgage finance coupled with a slowing in the growth of deposits, reportedly because interest rates being paid on deposits were not competitive with other institutions. For this reason, the Government introduced tax-free Home 76/ The large institutions, as measured by total deposits, are HFCK, Savings and Loan Kenya Ltd. (SLK), and the East African Building Society (EABS). -73- Mortgage Bonds in mid-1981 to stimulate deposits in HFCK and SLK. Other measures which have been taken in recent years include loans from the National Social Security Fund (NSSF) to HFCK and SLK, direct Government loans to HFCK in the amount of KSh 20 million in FY79/80 and KSh. 25 million in FY80/81, and increases in the interest rates paid to depositors to make the institutions more competitive in attracting funds. 5.33 A number of analysts have shown that a significant gap occurs in the provision of housing finance between the higher income market served by the housing finance institutions and the housing loans provided with site and service plots through Kenya's low-cost public housing programs. In 1983 costs, this financing gap covered housing units in the range of KSh 40,000 (US$3,200), which was the cost of a fully-serviced site and service plot with a materials loan to build a bare two-room house77/, to a housing unit costing in the range of KSh 100,000-150,000 (US$8,000-12,000), which covered approximately the lowest cost unit being financed by HFCK. This bimodal distribution of housing finance helps explain the significant pressure by middle-income groups on the available site and service plots. 5.34 The existence of a substantial gap in the housing finance market is unsatisfactory from a development standpoint and indicates a failure to develop sectoral policies which service the housing needs of the full spectrum of Kenyan households. As a result, middle income households are largely frustrated in their desire to improve their housing and gain access to a significant form of asset accumulation. The Government should support innovative programs to extend the coverage of housing finance to lower and middle income households. These might include special programs involving private developers and housing finance companies in providing low-cost housing using reduced building standards, as well as site and service type programs combining housing finance companies with the local authorities or NHC in promoting self-help development. Such programs are underway in Zimbabwe and Malawi, as well as efforts in Kenya, and these concepts need to be developed and extended to provide a full range of housing options for urban households in Kenya. 5.35 Local Authorities. Local authorities have come relatively recently to their role as developers of owner-occupied housing. The original entry of local authorities into the housing market was in colonial times in order to provide rental accommodations to African workers. This has left a tradition of local authority housing as a social service that still influences the approach of the councils. For example, a Government ruling in 1976 on the rentals charged by local authorities confirmed the principle that local authorities should make "no profit-no loss" on their rental housing stock, citing as justification "the Government's social 77/ Based on cost estimates developed for the Bank-financed Secondary Towns Project. -74- responsibility towards housing the low-income group, which must not be left to market forces."78/ 5.36 Even after more than a decade of official policy emphasizing owner-occupied housing, rental housing still dominates the stock of housing held by the local authorities. Table 5.7 shows the composition of the local authority housing stock in 1969 and 1980. In 1969, rental housing comprised 90% of the housing stock provided by local authorities, with the remainder consisting of a small amount of tenant-purchase housing and site and service plots. By 1980, the share of rental housing had shrunk to 69% of the housing stock. If we look at the net addition to the stock between 1969 and 1980, however, we find that new construction of rental housing accounted for 37% of the increase. Over half of local authority rental housing is owned by Nairobi City Council, with Nakuru and Kisumu Municipal Councils holding the second and third largest shares respectively. 5.37 As discussed in Chapter IV, local authorities have a clear mandate to provide basic urban infrastructure. What has been added in site and service projects is that the local authority is responsible for providing long-term mortgage finance to the plotholder for the capital cost of the plot and for the cost of housing construction through loans for building materials. While some of the local authorities have accepted this responsibility, it is a significant administrative burden, particularly for smaller local authorities, and a responsibility that the local authorities are not necessarily well disposed to carry out. Given the "social service" orientation of local authority housing, it is not surprising that a number of local authorities have substantial arrears on their site and service developments. For example, a USAID audit of NHC-funded housing projects in five secondary towns found extremely unsatisfactory loan management procedures and arrears averaging 27% of amounts billed to plotholders. Given this situation, it is not surprising that a number of the local authorities are in arrears to NHC on their loan repayments. 5.38 There would be a number of advantages from greater participation of commercial housing finance institutions with local authorities in site and service type programs. For the local authorities, it would allow them to concentrate on the provision of urban infrastructure without the corresponding responsibility of providing long-term mortgage finance. Because the cost of infrastructure could be refinanced by the housing finance companies as part of the total mortgage loan, the local authorities would be able to recover their investment cost on a much shorter term basis and use the money to extend urban services to new 78/ "Report by an Inter-Ministerial Committee on Rent Increases of Local Authority Housing in Kenya," Ministry of Housing and Social Services, August 31, 1977. The charges for rental housing are based on historical cost. If the councils would agree to sell part of their rental housing stock at a price approaching the current market price, or alternatively to revalue the houses at market prices in setting rentals, they could generate substantial sums to invest in building new housing without having to rely on loan funds from the central Government. -75- housing areas. By relying on private commercial funds for mortgage financing rather than the Government development budget, scarce public funds could be leveraged to provide substantially more housing to lower income households. Regarding the commercial housing finance companies, they would gain access to a larger market for mortgage lending while relying on the local authorities for the details of infrastructure construction, housing inspection, and technical assistance to plot allottees. In addition, the wider availability of mortgage financing is likely to increase deposit mobilization among lower and middle income households, thus directing additional resources into the sector. Additional incentives may also be necessary in order to ensure that this market is attractive to commercial lending agencies. These could include a guarantee by the local authority on stage payments made to plot allottees during the process of housing construction, assistance to the housing finance agencies for enforcement of penalties on arrears and reallocation of plots after eviction proceedings, and possible mixing of public funds at attractive interest rates with private funds in covering the mortgage finance requirements. Nevertheless, given the large premium which currently exists between the cost of site and service plots and their market price, this market should entail relatively low risk to all participants, while expanding greatly the availability of housing funds to a broader range of Kenyan households. 5.39 National Housing Corporation. NHC was established in 1966 out of the former Central Housing Board, following the recommendations of a 1964 UN Mission to Kenya on Housing.79/ The 1979-83 National Development Plan lists the following activities as the role of NHC: (a) transmission of government funds to local authorities for the development of low-cost housing; (b) provision of technical assistance in housing to local authorities; (c) development and management of housing estates; (d) support and encouragement for housing research; and (e) stimulation of private sector participation in mortgage housing estates. Though this description implies that considerable initiative is left to local authorities, in practice NHC has taken over responsibility as the major implementing agency for public sector housing programs in the country, with the exception of the largest municipalities such as Nairobi, Mombasa, and Kisumu. 79/ See "Housing Policy for Kenya," Sessional Paper No. 5 of 1966/67, Republic of Kenya; and "United Nations' Mission to Kenya on Housing," by L.N. Bloomberg and C. Abrams, 1965. -76- 5.40 A consultant's report on the operations of NHC has concluded that the past performance of NHC has been poor.80/ Table 5.8 presents information on the housing units and site and service plots completed by NHC over the period 1976-83. NHC's major area of activity during this period was in the construction of completed housing units rather than site and service plots, despite the emphasis in the various National Development Plans on site and service development. Since the average price of NRC-completed housing units has generally been beyond the affordable range of the lower income group, the amount of public sector housing funds directed to the lower income group by NHC has been in fact relatively small over this period. 5.41 The consultant's report on NHC recommended a number of organizational changes intended to improve NHC's operations. However, it is more likely that the problems faced by NHC arise because its mandate is overly broad and not conducive to the promotion of financial discipline and cost-efficiency. One can identify at least two separate roles which are carried out by NHC. The first role is as a housing developer, both as an agent for local authorities and by developing a limited number of schemes (tenant purchase, etc.) which it holds on its own account. However, the housing development role can be logically separated from NHC's second major function, which is to be a secondary housing bank which on-lends housing funds to local authorities and implicitly retains the responsibility to review and appraise projects for their technical and financial feasibility. In effect, NHC is in a virtual monopoly position on public funding that reduces its incentives for cost control and innovative project design. Although this tendency is supposed to be moderated by ministerial direction through housing policy guidelines, this leverage has not been effective. An alternative approach would be to have NHC concentrate solely on low-income housing development, with housing finance channeled through other institutions such as LGLA or commercial housing finance companies. This would promote greater competition in the use of public funds and could stimulate innovative efforts in servicing the housing needs of lower income households in a more effective manner. D. Conclusions 5.42 This chapter has examined a number of factors relating to the unsatisfactory performance of the housing sector in Kenya. Because of the constrained economic situation in the country in general and the priority needs of other sectors in responding to Kenya's adjustment policies, the Government is unlikely to sanction any significant redistribution of public resources to the housing sector. This means that new initiatives in the housing sector should emphasize greater efficiency and improved use of resources rather than expanded public investment requirements. At the same time, economic constraint does not justify complaisance; there are areas where Kenya's housing policy is inconsistent with the current situation and can be changed by Government action. 80/ "Study of the Operations of the National Housing Corporation," Final Report, January, 1981, Coopers and Lybrand Associates. -77- 5.43 First of all, in making the choice between quantity and quality in the servicing of urban land, all of the evidence favors quantity as the higher priority in Kenya. Public sector housing projects, including site and service projects, are too expensive and too slow to meet the problem adequately. Private sector resources are meeting the problem, although at very low standards. The priority of the Government should be to supplement these private initiatives by encouraging the provision of basic services and minimum planning controls. Only in this way can the public sector expect to promote a scale of production adequate to service the rapid urban growth that is occurring in Kenya. 5.44 In undertaking to service private land, the Government and the local authorities will need to forego the tight planning controls which have characterized public sector housing projects up to now. This means that politically attractive policies such as implicit subsidies through artificially low plot charges, allocation of land to individual plot holders, and building restrictions on the quality of housing construction will not be possible in most projects. Indeed, it is often these bureaucratic controls which have limited the speed of implementation of public housing projects. In effect, the Government's low-income housing policy should be expanded to include provision for renters on private land, rather than focusing solely on homeowners (and tenants) occupying publicly provided plots. This direction is consistent with conditions prevailing in the housing sector, but it would imply a significant redirection in the Government's housing program. 5.45 In undertaking this change in housing policy, the local authorities will play a major role in the identification and implementation of projects. However, it is unrealistic to expect the local authorities to take a leadership role in pursuing this policy without advice and encouragement from central Government agencies. The Secondary Towns Project includes a number of components involving the servicing of private land. Experience with these components can play an important role in demonstrating the concepts to the local authorities and establishing procedures and priorities for expanding the program on a larger scale. 5.46 The appropriate role of NHC in Kenya's housing policy needs better definition. NHC has an experienced technical staff and could be of substantial assistance to the smaller local authorities in identifying and implementing urban projects. On the other hand, NHC has not demonstrated commitment in the past to its low-income housing program. There is therefore little justification for continuing to maintain NHC in a monopoly position on the allocation of public housing funds. These funds could be made available through other institutions, including LGLA and the commercial housing finance companies, on the basis of the competitive merit of proposals to service the target income group, rather than relying solely on NHC to provide this initiative. As for NHC's housing development activities, it would need to define a position in the shelter market which would justify its conti'nuing involvement. One possibility is the planning and implementation of urban servicing and shelter projects on a contract basis for the smaller local authorities, and another possibility is innovative site and service type developments in partnership with commercial housing finance institutions. The Government should consider -78- providing technical assistance to NHC for a limited period to allow it to establish a position in the shelter market, after which NHC would be responsible for obtaining development funds on a competitive basis rather than through the Government budget. 5.47 Expanding the role and participation of the commercial housing finance companies in Kenya's low-income housing progam represents the most important challenge for future policy development. Many of the local authorities are not equipped to operate efficiently as long-term mortgage lenders. Moreover, given the constraints on the Government development budget, the public sector programs will become increasingly marginalized unless they can leverage public resources with private funds through the housing finance institutions. Innovative efforts to combine the successful elements of existing site and service shelter programs with the greater financial discipline and resource mobilization potential of the private sector will ultimately help bridge the gap in the housing finance market and make a wider range of housing options available to urban households. 5.48 In considering new priorities in the housing sector, it is important to bear in mind that Kenya has made major advances in its housing policy during the past decade. Questions of physical standards, cost recovery, and affordability which were on the forefront of the housing debate ten years ago are now accepted and incorporated into the housing policy which Kenya has been pursuing. If these concepts and the projects which introduced them have yet to provide a fully satisfactory model for sustained housing production, the issues which impede the implementation of housing projects are much better understood. Certainly the economic constraints which Kenya faces will limit the resources that the Government can provide to the housing sector. Nevertheless, sustained momentum in the provision of basic urban services and access to land remains an important priority in terms of maintaining Kenya's ability to manage urban growth. VI. SUMMARY OF RECOMMENDATIONS 6.01 The major findings of this report are the following: (a) Role of Local Authorities: Urban local authorities provide the basis for the development of local capabilities, initiatives, and resources, all of which are necessary to provide for urban growth in Kenya. The changing partnership of the local and central agencies in urban development needs redefinition, with a clear designation of responsibilities and identification of the financial and human resources necessary to carry them out. (b) Local Authority Finances: The deteriorating state of local authority finances is the major constraint on the continued growth of urban services. The structure of recurrent revenues needs improvement, with particular emphasis being given to improving the buoyancy of general revenue sources. -79- The system of central Government grants-in-aid to local authorities requires careful scrutiny, with a more consistent basis established between the obligation on local authorities to extend services and the revenues available to to support them. Finally capital financing decisions for local authority investments need to pay greater attention to the financial ability of the local authority to maintain the investment and cover the debt-service obligation within its revenue structure. (c) Urban Investment Planning: The planning of urban investments should focus on the local authorities as the "prime client" for urban development. The role of the central Government agencies should be to assist in the screening of urban investments within the District Focus on development, provide technical assistance and training to local authorities in developing their own capabilities, and assume an advocacy role for urban development through the identification of resources and cost-efficient programs to provide for urban growth. (d) Spatial Investment Criteria: Urban investments should be chosen according to their individual rates of return and the capacity of the implementing agencies, rather than to influence a particular spatial pattern of urban growth. This implies a policy of "evenhandedness" in choosing urban investments which would primarily support growth in areas where it is already occurring. The secondary towns can be expected to be major centers of urban growth and will therefore require access to a greater share of investment funds. (e) Servicing Urban Land: Major priority should be given to the extension of urban services on a widespread basis, such as through the servicing of privately held land. To reduce costs and mobilize private resources as much as possible, consideration should be given to the phased introduction of basic infrastructure, voluntary agreements with private land-owners to obtain land rather than public acquisition of land, faster approval procedures by local authorities, greater willingness to compromise on adoptive standards, and minimum controls consistent with public health objectives. 6.02 In support of these findings, the following actions are proposed in the report: (a) Strengthening of local authority finances, with emphasis on measures to increase the elasticity of local authority revenues, improvements in revenue collection procedures, methods of measuring the productivity and demand for local authority services, and clearer designation of the responsibility for funding centrally-mandated services; -80- (b) Establishment of an inter-ministerial urban development planning framework, which would facilitate the screening of urban investments in the Forward Budget process, provide a forum for ensuring inter-project consistency, and undertake monitoring and evaluation of urban investments to improve the speed of implementation; (c) Improvement of the financial review procedures of LGLA, with emphasis on the analysis of the impact of proposed investments on the overall financial position of the local authorities; (d) Development of criteria for pricing urban services, such as water, housing, and trading undertakings, which would cover recurrent costs and provide adequate funds for capital replacement and expansion; (e) Adoption of the Kenya Low-Cost Housing By-laws Study; (f) Establishment of NHC on a self-financing basis, with public housing funds being redirected through other financial agencies; and (g) Innovative programs to extend coverage of commercial housing finance to lower and middle households through joint projects with local authorities, NHC, and/or private developers. 6.03 In addition, the following actions are recommended for serious consideration: (a) Clarification of the policy framework for urban transportion, including support for urban road maintenance, allocation of a reasonable amount of investment funds for road expansion and traffic engineering improvements, and clarification of ministerial responsibilities for public transport policy; (b) Adoption of a uniform pricing policy for water supply; (c) Adoption of a graduated system of land-use zoning with varying degrees of control on physical development; (d) More attention to employment concerns in urban development planning; and (e) Review of training requirements for local authorities and other personnel involved in urban development. Conclusion 6.04 Urbanization in Kenya has neither top priority in the Government's list of current problems nor any special claim over other -81- sectors to increased development resources. On the other hand, urbanization is too important as a long-term issue to neglect the institutional and financial framework required to support this transition in Kenya's development. Given the heavy demands on the central Government, local authorities will bear the major responsibility of providing for urbanization, both through the development of local resources and by the identification of innovative and cost-efficient ways of extending urban services. However, local authorities cannot do it on their own. The central Government must provide financial legislation and urban planning procedures which support the local authorities. This report has presented a number of recommendations for improving the cost incentives and financial discipline of urban development programs. As many of the recommendations would require changes in current methods of operation, it is not expected that they will be accepted without additional study and careful consideration by the Government. It is intended that the analysis presented here should provide a framework for a continuing dialogue on the role of urbanization during the difficult economic times being experienced in Kenya. This is an important task in preparing for the adjustments in Kenya's economy which will be required during this decade. -82- STATISTICAL ANNEX cc I -84- Table 1.1: POPULATION OF KENYA, 1962-79 Urban, as Total Urban Percentage Number of Population Population of Total Urban Centers a/ 1979 15,322,000 2,307,000 15.0% 90 1969 10,943,000 1,080,000 9.9% 47 1962 8,636,000 748,000 8.7% 34 Intercensal Growth Rates 1969-79 3.4% 7.9% 1962-69 3.3% 5.4% Source: Population Census 1962, 1969, 1979. a/ Urban centers are defined in the census as towns with a population of more than 2,000 inhabitants. -85- Table 1.2: URBAN CENTERS WITH POPULATION EXCEEDING 20,000 (1979) Population Area (km2) Population Growth Rate (II) (VII) 1969 (VI) 1969-79 (I) (adjusted (III) 1969-79 (adjusted for (VIII) 1979 to 1979 1969 (IV) (V) (adjusted for urban/rural 1962-69 Town (census) boundaries) (census) 1979 1969 boundary changes) population) (census) Nairobi 827,775 509,286 509,286 684 684 5.0 5.0 5.6 Mombasa 341,148 247,073 - 247,073 210 210 3.3 3.3 4.7 Kisumu 152,643 111,700 32,431 270 19 3.2 3.5 4.6 Nakuru 92,851 50,700 47,151 78 32 6.2 6.2 3.1 Machakos 84,320 - 6,312 324 19 -- -- -- Meru 70,439 52,900 4,475 128 2 2.9 10.4 -- Eldoret 50,503 24,900 18,196 58 25 7.3 7.3 -1.1 Thika 41,324 23,400 18,387 92 15 5.9 5.9 4.0 Nyeri 35,753 22,600 10,004 72 7 4.7 7.3 3.5 Kakamega 32,025 18,300 6,244 47 8 5.8 10.5 -- Kisii 29,661 -- 6,080 35 3 -- -- Kericho 29,603 - 10,144 68 10 -- -- -- Kitale 28,327 13,300 11,573 80 18 7.9 7.9 3.1 Bungoma 25,161 -- 4,401 60 2 -- -- -- Busia 24,857 -- 1,100 45 -- -- -- -- Malindi 23,275 10,800 10,757 21 18 8.0 8.0 9.2 Source: "Population Census," 1969 and 1979 Central Bureau of Statistics Definitions Column I 1979 population by town, based on 1979 town boundaries. Column II : 1969 population by town, adjusted to correspond to 1979 town boundaries. Column III : 1969 population by town, based on 1969 town boundaries. Column IV : Area of 1979 boundaries. Column V : Area of 1969 boundaries. Column VI : Population growth rate from census data, adjusted for changes in municipal boundaries (Cola. I and II). Column VII : Population growth rate for the urban core, determined by deducting estimates for rural population residing within town boundaries.a/ Column VIII: Population growth rate, 1962-69, adjusted for changes in town boundaries prior to 1969. a/ To compute a growth rate for the urban core, the rural population included within urban boundaries was subtracted from the total population using estimates of the rural-urban split provided in "Human Settlements in Kenya," Physical Planning Department, Ministry of Lands and Settlement, 1978. -86- Table 1.3: DISTRIBUTION OF TOWNS AND THEIR POPULATION BY SIZE, 1969-79 1969 1979 Percentage Percentage No.of Total of Urban No.of Total of Urban Population Size Towns Population Population Towns Population Population (in 'OOOs) (in 'OOOs) 100,000 and over 2 756 70% 3 1,322 57% 20,000 - 100,000 2 80 7% 13 568 25% 10,000 - 20,000 7 91 8% 10 140 6% 5,000 - 10,000 11 71 7% 22 154 7% 2,000 - 6,000 25 82 8% 42 123 5% All Towns 47 1,080 100% 90 2,307 100% Source: Population Census, 1969 and 1979 -87- Table 1.4: URBAN GROWTH PROJECTIONS TO YEAR 2000 (000's) 2000 2000 1979 (Based on 1969-79 (Based on 1962-69 (Actual) Growth Rates) a/ Growth Rates) b/ Popu- % Share of Popu- % Share of Popu- % Share of lation Urban Pop. lation Urban Pop. lation Urban Pop. Nairobi 828 36% 2,300 27% 3,500 41% Mombasa 341 15% 700 8% 1,100 13% Other Existing Centers 1,138 49% 5,100 59% 3,500 41% New Urban Centers -- _ 500 6% 500 6% Total 2,307 100% 8,600 100% 8,600 100% Percentage of Population Living in Urban Areas 15% 28% 28% a/ Assumed growth rates: Nairobi, 5.0%; Mombasa, 3.3%; other existing centers, 7.4%. b/ As given in "Human Settlements in Kenya," pp. 88-95. The estimates shown are an average of the high and low projections for the major cities and generally assume some diminuation of the 1962-69 growth rates. -88- Table 1.5: ESTIMATED RURAL-URBAN INCOME DISTRIBUTION, KENYA, 1976 Population Income Avg. Hshld.Inc. (%) (%) (KShs/yr.) A. Urban 11.9 30.3 9,179 Private Sector (Modern) 7.0 19.6 16,100 Public Sector 1.9 5.8 10,830 Informal Sector 3.0 4.9 5,900 B. Rural 88.1 69.7 5,014 Private Sector (Modern)a/ 5.1 16.4 19,358 Public Sector 3.4 6.2 10,830 Agriculturea/ 79.6 47.1 3,776 Source: This table is adapted from the ILO report, Planning for Basic Needs in Kenya, 1979, p. 19. a/ Unskilled and informal workers in the rural sector have been included under agriculture, while they are broken out separately for the urban sector. If low-income unskilled workers in the rural sector were included in the rural modern private sector, the apparent urban-rural wage gap in the private sector would essentially disappear. -89- Table 1.6: WAGE EMPLOYMENT AND EARNINGS IN 22 TOWNS, 1982 Wage Employment Earnings Population (1979)a/ Town Number Z Rank Amount % Rank Number % Rank (K£ 000) Nairobi 291,327 56 1 382,609 66 1 827,775 42 1 Mombasa 95,026 18 2 93,813 16 2 341,148 18 2 Kisumu 17,454 3 4 14,729 3 4 152,643 8 3 Nakuru 20,080 4 3 18,895 3 3 92,851 5 4 Machakos 4,285 1 16 2,390 1 17 84,320 4 5 Meru 5,479 1 11 3,633 1 11 70,439 4 6 Eldoret 15,689 3 5 10,032 2 6 56,503 2 7 Thika 15,097 3 6 11,467 2 5 41,321 2 8 Nyeri 9,417 2 7 7,367 1 7 35,753 2 9 13 others 49,335 9 - 37,659 5 - 251,041 13 - TOTAL 523,189 100Z - 582,594 100 - 1,947,797 100 - Source: Statistical Abstract, 1983, Central Bureau of Statistics. Coverage includes 22 towns with 1,000 or more persons engaged in wage employment since 1972. a/ Population Statistics from 1979 Census. -90- Table 1.7: PERCENTAGE DISTRIBUTION OF WAGE EMPLOYMENT BY TOWN AND INDUSTRY, 1982 Industry Next Employment as Four Towns ina/ % of Total Terms of Remaining Wage Industry/Town Nairobi Mombasa Employment 16 Towns Employment 1. Community, Social, and Personal Serviceb/ 54 13 12 21 37 2. Manufacturing 51 18 24 7 21 3. Wholesale and Retail Trade, Restaurants and Hotels 57 20 10 13 12 4. Transport and Communications 41 49 6 4 9 5. Construction 70 11 6 4 9 6. Finance, Insurance, Real Estate, and Business Services 74 13 7 6 8 7. Other Industries c/ 59 15 7 19 4 Total Wage Employment 56 18 13 13 100 Population Percentage (1979) 42 18 17 33 Source: Statistical Abstract, 1983, Central Bureau of Statistics. a/ Nakuru, Kisumu, Thika, Eldoret. b/ Includes Government and Parastatal Bodies. c/ Agriculture, Forestry, Mining, Quarrying, Electricity, and Water. -91- Table 1.8: GROWTH RATES IN WAGE EMPLOYMENT AND POPULATION BY MAJOR TOWNS Growth Rates Urban Town Wage Employment Population (1969-79) (1970-82) Nairobi 4.9 5.0 Mombasa 4.3 3.3 Nakuru 2.9 6.2 Kisumu 2.5 3.5 Thika 7.8 5.9 Eldoret 4.4 a/ 7.3 Average for 42 Towns 5.1 6.9 Source: Statistical Abstract, 1983 and 1979, Central Bureau of Statistics. a/ Eldoret suffered a large decline in wage employment in 1973 due to the loss of a major industry. Wage employment subsequently grew at the rate of 9.7% for the period 1973-82. -92- Table 1.9: NAIROBI: HOUSEHOLD INCOME DISTRIBUTION, 1974 Household Mean Per Capita Income Group Income Population Income (KSh/mo.) (KSh/mo.) % 7 0-99 15.3 1.0 0.05 100-199 71.6 1.0 0.21 200-299 91.2 6.5 1.80 300-399 100.6 8.0 2.45 400-499 147.7 6.0 2.73 500-699 143.9 14.1 6.19 700-999 179.2 18.1 9.88 1,000-1,499 288.0 17.6 15.49 1,500-1,999 291.0 8.4 7.44 2,000-2,499 386.0 3.0 3.54 above 2,499 1,003.5 16.4 50.22 Source: "Household Budget Survey," Central Bureau of Statistics. This table is reproduced from "Poverty and Growth in Kenya," Annex I of "Growth and Structural Change in Kenya," IBRD, 1982. Table 2.1; URRAR INVESTMENTS IN CENTRAL GOVERNMENT ACCOUNTS, 1980/81 (Khb 'UUU) 1979 Type of Total2i Large 6 Urban vestment by Small Popu- Urban Infra- Land Low-Cost Urban Industry Provincial lation Area Water Sewerage structuret Acquisition Housing Roads Area (KIE) Hospitals Airports Share Year: 1980/81 Municipalities: Nairobi 2,250 _ 800 100 3,650 _ 6,800(39%) 212(8Z) 2,342(51%) 1,510(66%) 36% Mombasa 3,250 100 450 40 440 200 4,480(26%) 170(6%) 800(17%) 745(32%) 15% Kisumu - - 500 - 760 - 1,260(7%) 184(7%) 300(6%) 42(2%) 7% Other 1,350 300 - - 709 - 2,359(14%) 1,709(65%) 1,185(26%) _ 16% Town Councils 250 550 _ - 650 - 1,450(8%) _ _ _ 11% Urban and County Councils 130 350 - - 500 - 980(6%) 370(14%) _ _ 15% Not Allocated 2,030 631 370 1,475 125 644 5,275(-) - - - Total by Type of Investment 9,260 1,931 2,120 1,615 6,834 844 22,604 2,645 4,627 2,297 (41%) (9%) (9%) (7%) (30%) (4 . . . Source: "Development Estimates for the Year 1980/81," Republic of Kenya a/ As percentage of total funds, excluding those in "not allocated" category. b/ This category consists mostly of off-site infrastructure and community facilities financed as part of low-cost housing projects. Table 2.2: SEWERAGE COVERAGE IN MAJOR TOWNS IN KENYA, 1979 TOWN EXISTING SERVICE SITUATION (1979 Septic/Conservancy Population) % Population Served Type of Treatment Works Tanks Pit Latrines Nairobi 70% - sewers Full conventional & Both permitted. Septic Exist in the (Population 20% - septic and oxidation ponds tanks with radial arms outskirts and in 828,000) conservancy Outfall to watercourse or Soakaways unauthorized tanks development 10% - pit latrines Mombasa 17% - sewers Partial (screening & Septic tanks form the Exist mostly (Population 24% - septic tanks sedimentation) and commonest method, with in mainland 341,000) 59% - pit latrines oxidation ditches soakage pits down to residential Outfalls to the sea water level areas, and in Majengo parts of the island Kisumu 60% - sewers Full conventional and Septic tanks with Exist in out- 'O (Population 35% - septic tanks lagoons soakaways permitted skirts and 153,000) Outfall to the lake squatter areas Nakuru 50% - sewers Full conventional and Septic tanks and Exist in the (Population 45% - septic tanks lagoons soakage pits outskirts and 93,000) 5% - pit latrines Outfall to the lake permitted old parts of the town Thika 99% - sewers Lagoons Permitted in the Exist in un- (Population 1% - septic tanks Outfall to watercourse remaining 1% of the authorized 41,000) & pit latrines town area due to development topographical obstacles Source: "Kenya Low Cost Housing By-law Study," Consultant's Report to the Ministry of Urban Development and Housing, 1980; Volume Two; Discussion Paper No. 3, pp. 19-20. -95- Table 4.1: EXPENDITURE OF LOCAL AUTHORITlES, 1972-1983 KE Million Nairobi Other Town and City Municipal Sub- County Council Councils Total Councils Total 1972 13.13 5.77 18.90 3.52 22.42 1973 13.05 6.00 19.05 3.86 22.91 1974 13.80 7.92 21.72 3.37 25.09 1975 16.13 8.50 24.63 5.10 29.73 1976 18.74 9.71 28.34 5.58 34.03 1977 21.56 11.05 32.61 7.32 39.93 1978 24.24 11.39 35.63 9.44 45.07 1979 26.69 12.72 39.14 11.64 51.05 1980 35.36 17.93 53.29 11.05 64.34 1981 44.05 20.78 64.83 13.32 78.15 1982 * 59.03 27.15 86.18 17.57 103.75 1983 * 65.22 39.57 104.79 20.23 125.02 Percentage Distribution, 1981 56% 27% 83% 17% 100% Annual Percentage Increase, 1972-81 14.4% 15.3% 14.7% 15.9% 14.9% * Provisional Source: Economic Survey, Central Bureau of Statistics. Table 4.2: MUNICIPAL COUNCILS: RECURRENT AND CAPITAL EXPENDITURE ON MAIN SERVICES 1975-1983 (KE million) Annual Percentage Percentage Distribution, Increase, 1975 1976 1977 1978 1979 1980 1981 1982* 1983* 1981 1975-81 Administration 3.1 3.4 3.6 4.3 5.2 9.4 8.8 8.9 10.0 11% 19% Community Services: Roads 1.4 1.7 1.9 2.3 2.3 3.1 4.3 4.5 9.3 7% 21% Sanitary Services 2.7 3.3 5.0 5.1 5.4 6.4 6.8 9.3 10.9 10% 17% Other 1.0 1.0 1.1 1.3 1.5 2.4 2.7 2.7 4.8 4% 18% SUB-TOTAL 5.1 6.0 8.1 8.7 9.2 11.9 13.8 16.4 25.0 21% 18% Social Services: Health 2.8 3.2 3.3 3.9 4.3 5.1 5.9 7.9 7.9 9% 13% Education 4.5 4.9 5.3 6.1 6.7 7.8 9.3 13.0 12.9 14% 13% Other 0.3 0.3 0.4 0.4 0.5 0.5 1.9 1.9 3.8 3% 36% SUB-TOTAL 7.6 8.4 9.0 10.4 11.5 13.4 17.1 22.7 24.6 26% 14% Economic Services: General Administration 0.9 1.1 1.0 1.1 1.0 1.2 1.5 1.6 2.9 3% 9% Water Undertakings 2.5 3.6 4.2 4.3 4.7 5.2 10.5 21.6 24.7 16% 27% lousing Estates (including Staff Housing) 4.4 4.8 5.5 5.3 5.9 6.0 6.6 7.3 8.7 10% 7% Other 1.0 1.1 1.2 1.5 1.9 6.2 6.6 7.6 8.9 10% 37% SUB-TOTAL 8.8 10.6 11.9 12.2 13.5 18.6 25.2 38.1 45.2 39% 19% TOTAL EXPENDITURE 24.6 28.4 32.6 35.6 39.4 53.3 64.8 86.2 104.8 100% 18% * Provisional Source: Economic Survey, Central Bureau of Statistics. Table 4.3: MUNICIPAL COUNCILS: ECONOMIC ANALYSIS OF EXPENDITURE, 1975-1983 (KE million) Annual Percentage Percentage Distribution, Increase 1975 1976 1977 1978 1979 1980 1981 1982* 1983* 1981 1975-81 Recurrent Expenditure Labor Costs 10.92 11.97 12.66 14.31 15.28 18.99 24.43 30.26 32.13 63% 14% Other Goods and Services 5.37 5.73 6.66 7.23 9.06 10.82 13.30 16.92 23.07 34% 16% Transfer to Households and Enterprises .45 .69 .42 .63 .55 .69 .95 .93 .79 2% 13% Transfer to Funds (current) .19 .30 .31 .58 .30 .32 .36 .49 .45 1% 11% TOTAL RECURRENT 16.93 18.69 20.05 22.75 25.19 30.82 39.03 48.60 56.43 100% 15% Capital Expenditure Gross Fixed Capital Formation 4.15 5.57 7.81 7.98 9.24 16.59 19.65 29.35 40.19 76% 30% Loan Repayment** 3.43 3.92 4.50 4.49 4.93 5.38 5.61 7.29 7.21 22% 9% Transfer to Funds (capital) .12 .27 .25 .40 .05 .50 .55 .94 .97 2% 29% TOTAL CAPITAL 7.70 9.76 12.56 12.87 14.22 22.47 25.81 37.58 48.36 100% 22% TOTAL EXPENDITURE 24.63 28.45 32.61 35.63 39.41 53.29 64.84 86.18 104.79 18% * Provisional ** Includes interest. Source: Economic Survery, Central Bureau of Statistics. Table 4.4: TOWN AND COUNTY COUNCILS: ECONOMIC ANALYSIS OF EXPENDITURE, 1975-1983 (KE million) Annual Percentage Percentage Distribution Increase 1975 1976 1977 1978 1979 1980 1981 1982* 1983* 1981 1975-81 Recurrent Expenditure Labor Costs 2.36 2.95 3.50 4.60 5.40 5.56 6.54 8.67 10.08 71% 19% Other Goods and Services .98 1.28 1.83 1.87 1.79 1.99 2.04 3.47 4.26 23% 13% Transfers to Households and Enterprises .45 .18 .36 .32 .47 .25 .41 .45 .56 4% -2% Transfer to Funds (current) .04 .05 .06 .10 .14 .18 .19 .32 .17 2% 30% TOTAL RECURREsNT 3.83 4.46 5.75 6.89 7.80 7.97 9.17 12.90 15.08 100% 16% t Co Capital Expenditure Gross Fixed Capital Formation .95 .77 1.23 2.12 1.46 2.62 3.76 3.73 4.44 91% 26% Loan Repayments** .28 .32 .31 .32 .34 .34 .26 .63 .47 6% -1% Transfer to Funds (capital) .04 .03 .03 .11 .03 .11 .13 .31 .26 3% 22% TOTAL CAPITAL 1.27 1.12 1.57 2.55 1.83 3.08 4.15 4.67 5.16 100% 22% TOTAL EXPENDITURE 5.10 5.58 7.32 9.44 9.63 11.05 13.32 17.57 20.23 17% * Provisional ** Includes interest. Source: Economic Survey, Central Bureau of Statistics. Table 4.5: MUNICIPAL COUNCILS: ECONOMIC ANALYSIS OF REVENUE, 1972-1983 (Kt million) Annual Percentage Percentage Distribution Increase 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982* 1983* 1981 1972-81** Recurrent Revenue Graduated Personal Tax 3.90 3.23 .02 -- -- -- -- -- -- -- -- -- -- -- Direct and Indirect Taxes (Rates, Licenses, Cesses) 3.86 4.15 5.00 5.87 7.08 7.59 7.76 8.47 9.39 16.20 16.88 23.67 43% 17% Sales ot Goods and Services and Other Income 7.45 8.03 9.06 10.10 12.46 13.95 15.57 16.31 20.15 19.06 25.08 33.88 51% 11% Government Grants .14 1.43 4.34 5.40 5.65 5.33 6.42 6.95 2.68 2.28 5.34 4.25 6% -9% TOTAL RECURRENT 15.35 16.84 18.42 21.37 25.19 26.87 29.75 31.73 32.22 37.54 47.30 61.80 100% 10% Capital Revenue Loans 4.19 3.01 3.89 3.02 5.69 5.73 5.79 5.87 7.72 12.29 13.52 17.84 -- 13% * Provisional ** Percentage annual increase for Government grants computed for period 1974-79. Source: Economic Survey, 1982, Central Bureau of Statistics. Table 4.6: TOWN AND COUNTY COUNCILS: ECONOMIC ANALYSIS OF REVENUE, 1975-83 (KE million) Annual Percentage Percentage Distribution Increase 1975 1976 1977 1978 1979 1980 1981 1982* 1983* 1979 1975-79 Recurrent Revenue Direct Taxes (Rates**) .57 .46 .98 .91 1.00 .91 1.32 1.44 1.43 13% 15% Indirect Taxes (Licenses and Cesses) 1.73 3.22 5.43 4.12 3.77 4.57 4.36 5.56 6.12 41% 17% Income from Property .63 .72 .83 1.12 1.03 .83 .98 .82 1.48 9% 8% Current Transfers .07 .18 .26 .45 .45 .71 .74 .88 1.38 7% 48% Sale of Goods and Services 1.62 2.02 2.23 2.65 2.79 3.43 3.15 5.09 4.62 30% 12% I TOTAL RECURRENT 4.62 6.70 9.73 9.25 9.04 10.45 10.55 13.79 15.03 100% 15% ° Capital Revenue Loans Raised .40 .36 .31 .46 .63 .78 1.54 3.23 2.86 25% * Provisional ** Paid by households and enterprises. Source: Economic Survey, 1982, Central Bureau of Statistics. Table 4.7: IE IlI REUENES AND EXPE1rruREs OF IARL ER lUNICIPALrrIES (KE ODO) Peveies 51qenditures Ucenses and T Ihers' Service GFPr Salaries/ Other Total Total dxlget Surplus User Goensatirg School Grants Pecurrent Public Social blic Finarce and Pecurrent (lWficit) an Council Buiget Year Rates (aig- Grant Fees Grant and Income Revemjes Education Health Services Works Mministration Other Expenditures General Fund 1. Nairobi 1980 Actual 6,380 3,127 1,000 3,229 804 14,540 4,450 6,344 1,127 2,284 1,757 943 16,905 (2,365) City 1981 Prdbable 10,240 3,670 617 4,177 804 19,508 5,838 7,722 1,440 2,778 2,461 1,731 21,970 (2,462) Co,Idil 1982 Estinate 11,320 4,892 562 4,268 804 21,846 6,463 8,503 1,590 2,925 2,690 2,223 24,394 (2,548) 2. Mbra 1980 Actual 1,115 1,231 300 986 119 3,751 1,379 1,135 125 1,238 295 39 4,211 (460) lMiicipal 1981 Probable 2,609 1,604 199 1,472 125 6,009 1,685 1,679 204 1,563 358 84 5,573 436 Ocxicil 1982 Estimate 2,609 1,821 185 1,334 125 6,074 2,111 2,077 289 2,141 494 112 7,224 (1,150) 3. Kisumu 1980 Actual 169 401 67 339 40 1,016 193 332 103 382 213 - 1,223 (207) lxnicipal 1981 Probable 320 490 67 181 50 1,108 396 417 141 320 275 - 1,549 (441) 0hncil 1982 Estinate 350 841 67 896 50 2,204 1,046 604 152 367 363 - 2,532 (328) 4. Nakuru 1979 Actual 353 423 67 535 45 1,423 616 301 124 256 31 - 1,328 95 ?Micipal 1980 Actual 452 439 54 624 45 1,614 726 374 145 275 38 - 1,558 56 (xincil 1981 Probable 423 468 59 868 75 1,893 900 448 158 333 38 - 1,877 16 ° 5. Eldoret 1979 Actual 113 306 39 154 14 626 149 68 43 203 52 45 560 66 1xnicipal 1980 Probabe 131 346 27 197 14 715 209 97 57 315 79 62 819 (104) Couil 1981 Estimate 131 385 27 197 14 754 244 111 65 341 93 82 936 (182) 6. lhika 1979 Actual 100 239 37 163 13 552 144 80 23 135 155 29 566 (14) tmnicipal 1980 Prbable 30(a/ 318 31 138 18 805 168 140 40 169 216 45 778 27 Coumcil 198l Fstieste 201 359 30 140 47 777 154 96 32 198 220 54 754 23 7. Kitale 1979 Actual 96 127 31 139 11 404 125 58 32 122 91 - 428 (24) midipal 1980 Prdbale 98 153 23 173 18 465 204 75 46 141 119 - 585 (120) Coamcil 1981 Estimte 98 150 31 218 25 522 214 83 51 148 109 - 605 (83) a/ Includes collection of arrears. -102- Table 4.8: SUMMARY OF FUND BALANCES, 1981 a/ (K£ '000) Reserve Water Housing Other Council Fund Fund Fund Funds Total Nairobi City Council (4,105) 12,977 (1,736) _ 7,136 Mombasa Municipal Council 936 - 276 - 1,212 Kisumu Municipal Council (331) 393 420 130 612 Nakuru Municipal Council (16) 249 464 - 697 Eldoret Municipal Council 412 409 141 19 981 Thika Municipal Council (168) 700 161 - 693 Kitale Municipal Council (297) (42) 114 - (225) Source: Budget Estimates for various towns; all balances are estimated as of end 1981. a/ Excluding Capital and Renewal Funds. -103- Table 4.9: SOURCE OF FINANCING FOR RECURRENT EXPENDITURES a/ Licenses and User Surplus (-) Rates Charges Grants Deficit (+) Nairobi 38% 18% 30% + 14% Mombasa 27% 29% 33% + 11% Kisumu 14% 33% 36% + 17% Nakuru 26% 32% 49% - 7% Eldoret 20% 55% 37% - 12% Thika 18% 42% 38% + 2% Kitale 22% 30% 42% + 6% a/ Nairobi, Mombasa, Kisumu (1980). Nakuru, Eldoret, Thika, Kitale (1979). -104- Table 5.1: PERCENTAGE DISTRIBUTION OF URBAN DWELLING UNITS BY SECTOR, 1979 a/ Private Sector Formal b/ 64% Informal c/ 18% Sub-total 82% Public Sector Local Authorities 10% Central Government d/ 8% Sub-total 18% Notes: a/ A dwelling unit is defined as a place of residence for a family, an individual, or a group of persons eating together and sharing the budget for common provisions. A single housing structure may contain multiple dwelling units. b/ Includes site and service plots provided by the public sector. c/ Informal housing includes both squatter settlements and housing developed by land-owners who have not completed the required legal procedures of sub-division and registration. Much of the latter type of housing is owned by land companies, such as those holding large tracts in Mathare Valley in Nairobi. d/ Includes public boards and corporations. Source: "Report on the 1979 Rent Survey in Urban Areas of Kenya," Central Bureau of Statistics, 1981. -105- Table 5.2: PERCENTAGE DISTRIBUTION OF PRIVATE HOUSING BY TYPE AND TENANCY STATUS, 1979 Tenancy/Type a! House Block Other Sub-total Owner-occupied 4% 2% 18% 24% Rental 5% 7% 64% 76% Sub-total 9% 9% 82% 100% Average Number of Rooms per Dwelling Unit 4.1 3.3 1.4 a/ Definitions: House: A self-contained dwelling unit on its own compound and occupied by a single family or group of people sharing a common budget for basic provisions. Block: Includes both flats (a self-contained dwelling unit joined to others in a single multi-storied building) and maisonettes (a semi-detached or terraced self-contained dwelling unit on two floors). Other: All other dwelling units, such as structures occupied by several families in dwelling units that are not self-contained, for example by having toilet facilities outside the main dwelling unit. Source: "Report on the 1979 Rent Survey in Urban Areas of Kenya," Central Bureau of Statistics 1981 -106- Table 5.3: VALUE OF RESIDENTIAL BUILDING PLANS APPROVED BY NAIROBI CITY COUNCIL 1974-82 (KE '000) Private Public 1974 11,204 1,324 1975 7,961 1,415 1976 6,927 598 1977 15,177 2,964 1978 26,202 4,115 1979 33,641 1,765 i980 53,238 1,322 1981 56,269 1,952 1982 34,013 6,091 Source: Statistical Abstract, Central Bureau of Statistics Table 5.4: RENT INDEX FOR NAIROBI, 1974-82, BY INCOME GROUPS a/ Average Average Annual Annual Change in Change in Price Index Year b/ Rent for All Other Income Group 1974 1975 1976 1977 1978 1979 1980 1981 1982 Index Commodities Upper Income 96.9 109.3 115.6 135.1 145.7 164.1 186.1 221.0 259.4 20.3 20.9 Annual Change 12.4 6.3 19.5 10.6 18.4 22.0 34.9 38.4 Middle Income 96.2 111.1 120.5 146.3 158.7 175.3 195.2 243.5 288.3 24.0 20.0 , Annual Change 14.9 9.4 25.8 12.4 16.6 19.9 48.3 44.8 Lower Income 95.3 114.0 124.3 161.0 187.7 205.1 232.0 277.6 314.6 27.4 21.2 Annual Change 18.7 10.3 39.7 23.7 17.4 26.9 45.6 37.0 Residential Bldg n.a. n.a. 181.0 198.9 217.9 250.1 n.a. n.a. n.a. Cost Index c/ 17.9 19.0 32.2 a/ (Base: January-June, 1975 - 100) b/ Computed at December of each year. c/ (Base: December, 1972 - 100) n.a. - not available. Source: Statistical Abstract, Central Bureau of Statistics. -108- Table 5.5: SOURCES OF NEW RESIDENTIAL PLOTS ( ANNUAL AVERAGE 1981-85) Plots Allocated by Commissioner of Lands 1,000 Private - legal subdivisions 4,000 squatting and informal subdivisions 23,000 Allocated by NHC and Local Authorities 2,000 Source: Mission estimates, based on Commissioner of Lands monthly reports, approved subdivision statistics, NCC monthly minutes, and NHC statistics. The definition of a "plot" in an informal subdivision is a loose concept and may include a single room in a multi-family rental structure. -109- Table 5.6: ANNUAL URBAQN LAND REQUIREMENTS (1981-85) Nairobi Kenya DU's a/ Land (ha) DU's Land (ha) NEW DWELLINGS Government/Parastatal Local Authority @ 250 m2/DU 2,000 50 3,000 75 Private Authorized @ 10 DU's/ha 2,500 250 4,000 400 Private Unauthorized @ 50 DU's/ha 10,000 200 23,000 460 Total Residential 14,500 500 30,000 935 NON RESIDENTIAL 500 Commercial and Other Uses (based on estimated 1:1 ratio of residential land to other land in urban areas) 500 935 TOTAL 1,000 1,870 Source: Mission estimates based on prevailing densities for residential construction and published statistics of approved non-residential development for 1976-1978. a/ DU = dwelling units. Table 5.7: LOCAL AUTHORITY HOUSING STOCK, (1969-80) 1969 1976 1980 1969-80 Type Number Percent Number Percent Number Percent Percentage Annual Change Rental Housing 27,092 90% 30,674 80% 34,820 69% 2.3% Tenant Purchase 1,852 6% 4,185 11% 6,243 12% 11.7% Site-and-Service Plots 1,321 4% 3,593 9% 9,861 19% 20.1% Total 30,265 100% 38,452 100% 50,924 100% 4.8% Source: Up to 1974, "Local Authority Housing in Kenya," HRDU, 1978. Thereafter, NHC Annual Reports. -111- Table 5.8: HOUSING UNITS COMPLETED BY THE NATIONAL HOUSING CORPORATION (1976-83) HOUSING UNITS NAIROBI ELSEWHERE SITE AND Average Average SERVICE No. of Value of No. of Value of PLOTS a/ Units Units (KE) Units Units (KE) No. of Plots 1976 1,031 596 752 1,164 1,128 1977 586 2,399 1,226 874 355 1978 862 3,042 682 1,861 1,077 1979 3,056 1,694 1,029 1,264 2,389 1980 616 2,604 2,911 1,395 2,454 1981 1,072 898 1,683 1,192 2,719 1982 2,311 878 617 3,538 2,550 1983 -- -- 680 2,878 598 TOTAL 6,767 6,695 10,405 a/ Includes site and service plots completed by local authorities with NHC financing. Source: Economic Survey, Central Bureau of Statistics I IBRD 15860R S U D A N ._-.7 I // E T H I 0 P I A I \\.--. EASTERNr U G A N D A - \i < / K 0~~~~~~~~~~~~~~~~~~~~ IFT VALLEY NORTH EASTERN 5- UGA DA ) / ) N 1F 1 NY'SZAS A/ v CENT 24 i W .~~~~NAIROBI(Kos 9 KENYA i * RIF VALLEY URBAN AREAS 1979 COASTRA T.-n a,h p.plbt- wer 10,000 \ \ . / 95.f0-1 ; "Om, Sooo ,b A F R I c A :wrA,2~ S N.t,on.l -cp-:l K-nyo/ \K- P-ovnc cait l 0 40 ro W2 Wh K o~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ . --Provi... b-udarifs . J >hoW4 20 40 WO BO 1~0. MLES URBAN AR S 1 9AUGUST 1985