Housing priorities and demand of lower income households in Mexico City

DRAFT R R-27605 THE WORLD BANK DEVELOPMENI ECONOMICS DEPARTMENT URBAN AND REGIONAL ECONCMICS DIVISION URBAN AND REGIONAL REPORT No. 76-5 HOUSING PRIORITIES AND DEMAND OF LOWER INCGIE HOUSEHOLrS IN MEXICO CITY Toäasz Sudra and John F. C. Tuimer June 1976 These materials are for internal use only and are circulated to stimulate discussion and critical camment. Views are those of the author and should not be interpreted as reflecting the views of the World Bank. References in publications to the Reports should be cleared with the 'author to protect the tentative character of these papers. kl6 HOUSING PRIORITIES AND PREFERENCES OF LOWER INCOME HOUSEHOLDS IN MEXICO CITY TABLE OF CONTENTS INTRODUCTION 1 I. OPERATIONAL PROBLEMS AND ISSUES IN HOUSING 4 II. THE HOUSING PROCESS AND HOUSEHOLD CHARACTERISTICS 17 III. HOUSING PRIORITIES, EXPENDITURES AND RESOURCES 25 IV. PATTERNS OF HOUSING PREFERENCES 31 V. SUMMARY 37 ANNEX: METHODOLOGICAL NOTE ON THE SELECTION AND EXECUTION OF THE CASE STUDIES I. Selection 1 II. Case Study Proceedings 2 HOUSING PRIORITIES AND PREFERENCES OF LOWER INCOME HOUSEHOLDS IN MEXICO CITY INTRODUCTION Basic Characteristics of the Lower-Income Housing System in Mexico City In the 1940's and 1950ts the housing system in Mexico City functioned reasonably well for the rapidly growing lower income sectors of the populations.1/ Those with very little money could rent or share the rent of a room in a vecindad 2/ which satisfied their high priority for low living costs and immediate access to jobs and opportunities to climb out of poverty. Subsequently, they could either remain in the central city in a larger rental vecindad or become owner-builders on the periphery. This latter group was composed of those who, although not able to afford commercially supplied new dwellings, had a high priority for owner-occupancy and could buy reasonably cheap unserviced plots in the colonias populares 3/ where they were free to build incrementally with neither mortgages nor threats of foreclosure. Eventually, public utilities and community services would be installed in these settlements. Within this system the great majority of the lower income popu- lation was able to find appropriate residential locations, appropriate forms of tenure and shelter of fair quality for the price paid. By the end of the 1950's, however, the government curtailed the construction of vecindades leaving the existing ones occupied,mainly by moderate to middle income permanent residents, who were attracted to the low rents enforced under rent control. As a corollary, these same households were discouraged from participating in peripheral low.ihcome owner occupancy developments because of high land prices. Consequently, the poorest groups were denied access to the inner city while many of the less poor, who were willing to 1/ Low income sectors are defined in this study as the group of households that cannot afford unsubsidized housing of the minimum standard satisfying the existing building codes and regulations. In 1973, approximately 70 percent of the population of the metropolitan area was estimated to fall in this group. 2/ Vecindades - low income tenements usually organiz-ed around a central courtyard with communal services. They were located mostly in the center or in the intermediate ring. 3/ Colonias populares - low income subdivisions, that are also called fraccionamientos populares and colonias proletarias. The oldest are located in the intermediate ring and the more recent on the periphery. The levels of legality of the subdivisions can vary between the com- pletely legal to the squatter. -2- move and build on the periphery, were prevented from doing so either by the high cost of land or the poor prospects for improvement in their housing situations. The colonias populares, therefore, were settled to an increasing extent by those excluded from the inner city vecindades who were almost always too poor to build or pay for utilities and essential community services. Thus, the incremental development nature of the original colonias was largely eliminated and this transformation further discouraged house- holds that could invest from moving to these settlements. Between 1955 and 1974 the Mexico City metropolitan area grew from four to over nine million people. The low income population reached, according to different estimates, between 69 percent to 75 percent 1/ of the total population and the area covered by its housing increased to 63 to 70 percent of the total urban area. The majority of the newcomers were migrants from rural areas and from smaller cities, living in one of the housing sub- systems and following highly diversified geographic itineraries within the metropolitan area. The original itinerary--province to vecindades in the center to low income subdivisions on the periphery--that once was the most common, now applies only to a very small percentage of the population. Currently, the most common itineraries include a direct move from province to periphery, the same move via the intermediate ring, and continued resi- dence in the center. Many more moves between different localities at the periphery and from the periphery to the center can now be observed, the latter being made by people who had already lived in the center at one point or another and are returning. One should also add to this list the most recent type of center-to-periphery move, a result of the removal of populations from the inner city shanty towns and some vecindades to govern- ment built housing projects in peripheral locations. During the last 20 years, rental housing became more common in some of the older peripheral colonias, which now constitute part of the intermediate ring. The population coming to these areas from the province tends, increasingly, to move to the periphery and settle in illegal subdivisions or squatter areas. Government prohibition of construction of new vecindades, which were declared a sub-standard type of housing in the mid-fifties and the freezing of rents on about half of the existing stock dramatically cut the available supply of this type of housing during the period of accelerated migration and growth. The new migrants increased the already high densities in the vecindades and increased the prices of the unchanged supply of- uncontrolled units. Furthermore, the occupiers of rent-controlled vecindades gained a strong incentive to stay in their dwellings, while in an open market situation they might have moved to the periphery. Only the vecindades with unfrozen rents are still available to newcomers, but at rapidly rising prices. I/ Precise data on migration to the metropolitan area as a whole are not available, but in the 1960's 2.4 million people migrated to the Federal District and an additional 1.1 million to the state of Mexico, the latter number includes parts of the state other than the metropolitan area. -3- The mutually exacerbating breakdowns of these basic and comple- mentary sub-systems of colonias and vecindades have stimulated the growth of three compensatory sub-systems, each of which -creates acute problems for their residents, for the landlords, or for the government and society as a whole. The private sector, unable to profit from the legal vecindades, has provided the worst and, for the inhabitants, most costly slums of all: the ciudades perdidas. 1/ The .lower income groups, unable to rent tolerable accomodations in central areas, or unable to afford land on the periphery, created the colonias paracaidistas, 2/ undermining public authority and often creating serious problems for themselves and their neighbors. The public sector, under increasing pressure to "do something" about deteriorating housing conditions, has increased the supply of conjuntos subsidiados. 3/ Most subsidized housing is enjoyed by moderate and middle income families at the expense of the poor. If used by lower income groups, such housing, due to its distant location from the central city area and its high cost, even if subsidized, worsens rather than improves socio-economic conditions for the poor. Housing construction and financing by centralized agencies and large corporations generally costs more than housing built by small firms, artisans and owner-builders. The breakdown of the dual vecindad-colonia popular system is due to the legislative and administrative failure to cope with the huge growth of the metropolitan area, and to counter-productive responses to the conse- quent symptoms. Attempts to proscribe further growth within the Federal " District, to impose high income housing standards, to freeze the rents of a shrinking supply of vecindades, and to substitute publicly sponsored for privately built housing, have compounded the problem and exaggerated the distortions of the basic system. The case studies analyzed in this report illustrate the consequences of these developments to the users of housing. 1/ Ciudad perdida - shanty town of rented land or shacks. They are located mostly in the intermediate ring. 2/ Colonias paracaidistas - squatter areas, the majority of which are presently in the process of legalization or awaiting it and can be found almost exclusively on the periphery. If legalized they would be classi- fied as fraccionamientos populares. 3/ Conjuntos subsidiados - subsidized public housing projects. The oldest are located in the intermediate ring, the newest at the periphery. I. OPERATIONAL PROBLEMS AND ISSUES IN HOUSING The 25 case studies of lower-income households in metropolitan Mexico, which provide the evidence for this report, 1/ present a number of apparent paradoxes. Some of the least desirable dwellings in conventional terms are among the most satisfactory for their occupants; some of the poorest households have made the greatest investments, and, in conventional terms, have some of the best dwellings; and some of the most materially deprived families and individuals have shown a great capacity to make use of scarce resources. The central government appears to be the least efficient provider of appropriate housing services. The following sections discu!s the three major issues raised by these contradictions. First, the issue of value in housing: does it lie in the material nature of goods and services or in their usefulness as measured by the dwellers? Secondly, the issue of housing economy: are the production, distribution and use of housing dependent upon the productivity of the construction industry or is it possible to use alternative sources of human and material resources'? Thirdly, the issue of authority in housing: should decisions regarding housing be taken at the central level of govern- ment or at the local level--who implements the decisions regarding the pro- vision of housing services--how should the decision making process be allocated among the decision makers? Finally, this chapter discusses the operational policy problems of goal definition, instrument selection and strategy design as basic issues determining the ways in which practical problems are stated and approached. Different interpretations of housing value imply alternative statements of housing problems and policy goals. If housing is valued for its material standards, then conventional definitions, which relate housing problems to deficits of standardized dwelling units are appropriate. If -housing value is defined as the usefulness of housing goods and services to the users, quantitative production targets can be highly misleading and often counterproductive, particularly in economies with low per capita incomes. Policy strategies reflect alternative interpretations of authority and assessments of locAl and central capacity to influence housing policy. If it is assumed that central agencies are best able to decide who should be the recipients of what housing goods and services, and that such agencies and/or corporate enterprises are best prepared to respond to housing demand, then conventional modern housing policies will be supported. If central institutions and corporate bodies are less competent suppliers than local organizations and enterprises, emphasis will be focused on support for local action through the equalization of access to resources. 1/ The methodology used in selecting and executing the case studies is presented in the Annex. The following examples taken from the case study data illustrate the proposition that some of the poorest dwellings, materially speaking, are best suited to the needs of the households using them. Two of the dwellings with the highest material standards appear to present the greatest problems for their occupants. (a) The Supportive Shack (Case 22) The poorest of the 25 dwellings observed first-hand, is the shack made of waste materials by the young car painter. Due to a series of unusually wet seasons which increased the necessary drying time, the turnover in cars was too small, and the young man was temporarily unable to make a living at his trade. In order to survive, the car painter's family had moved in with a comadre 1/ who gave them the use of her back- yard and the service facilities of her own house for as long as they needed it. The comadre was a pepenadora (rag-picker) who lived in a new house allocated to her when the shantytown near the dump--which had previously housed her and her neigh- bors and fellow rag-pickers--was cleared by public authorities. As a leader among the pepenadores, the comadre was able to give the car painter access to the dump, from which he was able to make about 900 pesos, or approximately 20 percent more than the absolute minimum for subsistence per month. 2/ FIYom this income, the family of two adults and six children, pays its share of the utilities, including running water and toilet facilities in the main house and electricity supplied to the shack. They have a small amount for saving toward their ant'cipated move because of the rent-free accomodation. In the relatively mild climate of Mexico City, the poorly insulated shack is not too great a hardship; the roof, made from plastic and other materials culled from the dump, keeps the rain out. The family has plenty of space and privacy for its domestic life in the shack. They use the enclosed and private backyard. They are well located for work and for social activities. There are shops and schools for the three school- age children in the immediate vicinity, so that little or no money must be spent on transportation. Their security of tenure is invested in the compadrazgo relationship between the car painter (but not his wife or children) and the owner- occupier of the property. As long as this relationship exists, they are secure. In fact, the car painter declined an offer of a room in the comadre's house to avoid damaging their rela- tionship. 1/ A male compadre or a female comadre is an unrelated person who assumes the role of a blood-relative. 2/ Income levels in this report are based on a "subsistence income" scale (S) which will adequately feed a household if it spends at least 85 percent of its income on food and cooking fuel,. at. 1973 prices of the second semester in Mexico City. -6- The family of the car painter is young, self-confident and optimistic about its future prospects. Barring unfore- seen mishaps on a personal or national level, it is very likely that they will greatly improve their social and eco- nomic conditions during the course of the car painter's working life. Being young, healthy and motivated by expecta- tions of future achievements, the family's priorities are well matched to the housing services they have. The highest general priority is to maximize opportunities for the reali- zation of their hopes and expectations. The household budget strategy is based on such optimization: to save as much as possible in order to take advantage of opportunities as they arise. While the family would undoubtedly enjoy a higher standard dwelling, the physical quality of the shelter is less a priority than security of tenure or minimization of household expenditures. The materially poor dwelling is extremely well located for the family's needs. The form of tenancy, based on the compadrazgo relationship, is ideal, giving them security without commitment and freedom to move at short notice. For the car painterts family, the shack provides a means of coping with their present situation while planning for their future. (b) The Oppressive House (Case 24) In sharp contrast to the car painter's shack is the disastrously unsatisfactory situation created by the modern, standard house. This family--with father aged 55; mother, 54; and son, 15--had previously lived in a shanty-town not far from their present peri-urban site, where it had supported itself by selling homemade carnitas on Sundays at the entrance of a tourist attraction and by the irregular employment of the father as a semi-skilled mason. The family income before the move was about three times the subsistence minimum (3S), but with low housing and transportation expenditures, they were able to eat reasonably well and maintain a fair level of health. The family was instrumental in pressing for rehousing when the shanty-town was threatened with destruc- tion by the government, and they were relocated to a vastly improved dwelling in the San Vicente Guerrero public housing project. This "improvement" has resulted in a tremendous deterioration in the lives of the family members, both in economic and human terms. The family is now required to pay 55 percent of its total income to meet rent-purchase and utility payments, as .compared with five percent before the move. In addition, the working members, father and son, pay another five percent for public transportation to work-- a total of at least 60 percent of family income spent on direct and indirect costs of relocation. Besides the rise in housing-related expenditures, the family suffered a reduc- tion in income because of the ban against street vending at their new location. This double loss sometimes occurs in "housing improvement" programmes for low and very low income people, as Barney Rush's data on the effects of relocation in Rio de Janeiro confirm. 1/ This Rio de Janeiro study supports the common-sense expectation that people prefer the more comfortable dwellings, but it also indicates that the price paid for such comfort is often disproportionately high. Far more damage is done by dislocating people, by disrupting their life style, and by reducing their social and economic security, than by allowing them to remain in materially poor housing. Expectations for the future for Case 24 are low--at best they can hope to maintain themselves at a subsistence level. At the time of the interview, the husband was ill and unable to work. The older children, living outside the household, have few skills or resources, and are unable to support their parents, the youngest son, living at home, is likely to leave soon. The family's major priority, therefore, is security. At their previous location, the household was able to opti- mize its expenditures by minimizing expenditures on housing. In the present situation, however, their general budget strategy is being largely determined by externally imposed conditions which.create economic behavior contrary to the household's interests and priorities. The major disadvantage of their previous situation was the absence of secure tenure. They did, however, have access to sources of livelihood, as well as rent-free accommodations in their rudimentary but tolerable shack; they were able to feed and clothe themselves reasonably well. The present situation has caused the family to lose nearly all of these advantages; and, given the high level of housing expenditure, it is questionable that they can maintain secure tenancy. Their food and clothing needs are being sacrificed for the benefits of their improved dwelling. The usefulness of housing for the occupants, is not necessarily . related to the material standards of the goods and services provided, as the case studies indicate; therefore, conventional measures of housing value may be misleading. If such measures are used, the solution to "sub-standard" housing is to replace it with standard or above standard units. An increas- ingly common observation in cases of major slum clearance and rehousing schemes is that this solution often increases the problems faced by the intended beneficiaries. 1/ See Barney Sheppard Rush, "From Favela to Conjunto: The Experience of Squatters Removed to Low-Cost Housing in Rio de Janeiro, Brazil," Harvard College, March 1974, mimeo. -8- To elucidate the problems of the users, one must define "housing" and "value." Market values are different from human values in many cases. The word "housing" is used as a collective noun for a given stock of houses or residential properties but it is also used as the present participle of the verb "to house." It is important to dis- tinguish housing stock from the activities of housing. While it is entirely reasonable to speak of the market value of houses and of the human and social values of housing activities, these concepts must hot be confused. Social and institutional processes have many quantifiable aspects, but considered as understandable wholes, their effects and consequences for people and nature are only partly quantifiable. To a certain extent it is meaningless to place only monetary or market values on them or on the stock of houses. The evidence presented in this report shows that housing problems may be restated to reflect the human values of housing processes, and may point to feasible solutions that are desirable from the users' points of view. As essential as quantitative methods are for resource allocation and as aids in the identification of complex systems and their components, the users' housing problems do not lend themselves easily to such solutions. The 25 case studies show that the average equity owned by the 11 households with upper-lower and moderate incomes (between 5S and 10S) is approximately 14,000 pesos, and the average for the 14 low and very low income-households is approximately 22,000 pesos. This anomaly is illus- trated by the contrast between the case of the low income factory worker's owner-built housing (Case 20) and the government employee's subsidized project dwelling (Case 23). The total household income in Case 20 is half that of Case 23, but the former's equity is equivalent to four to five years' earnings, while family 23 has no equity at all, although it will accrue as he amortizes the mortgage on his new home. The material'standards of the project dwelling are higher than those of the owner-built house, but the matching of their respective dwellings with their priorities is good in both cases. Materially speaking, both households are satisfied with what they have. The government employee's subsidized project housing is reasonably economical for households which can afford the mortgage and utility payments and the factory worker's owner-built housing has provided the low income dweller with the opportunity to improve his shelter at the cost of inconven- ience to himself. Generally the poor are willing to undergo inconvenience to improve their housing but are not willing nor able to make monetary pay- ments because other subsistence necessities are more important. For the majority of households with incomes lower than that of Case 23, there are major distortions in spending patterns in the project dwelling alternative. Excerpts from the case studies demonstrate this argument. (a) The Economy of Owner-Built Housing (Case 20) The factory worker's house is an example of the most comnon contemporary housing system in most rapidly urbanizing cities of countries with low per capita incomes. In various forms, ranging from squatting to more or less legal purchase of semi-serviced plots, 15 of the 25 cases studied concern ow:,ar-built housing. Among the occupants of the owner-built house, five have similar incomes, from 3S to just over 4S, and share stable household economies with several children, mostly of school age. The head of each household is of provincial origin. In all five cases, the highest priority is for security of tenure, followed by improved housing standards. The factory worker's household represents a paradigm of the conditions and circumstances of the five owner-built households. The location, anenities and form of tenure obtained by this family match their present and expected situations very well. Like the great majority of wage- earners who have reached or are approaching the ceiling of their earning capacity, the household's greatest priority is for secure tenure. In contemporary Mexico, this means ownership or at least de facto possession of a plot bf land that is unlikely to be disturbed by public works or other developments. Beyond this, the household's major needs will be satisfied by a location which provides access to workplaces, schools and shopping facilities, and neigh- borhood proximity to friends or relatives. This may mean beginning life on a plot in an unserviced shack, but as long as the dwelling can be improved to moderate standards over a reasonable time period, the physical discomforts are likely to be preferable to either excessive expenditures for housing services elsewhere, or the insecurities of rented housing. The first conditions for housing economy--an affordable price and reasonable value for the money--are met in the case of the factory worker's family. The household spends 24 percent of its total income on housing, but this somewhat high proportion is temporary, and will be reduced once trans- fer and improvements charges for the plot are paid off. 1/ The current market value of the property is equivalent to at least four years' total earnings of the head of the household, or twice the usual ratio. The net price paid is lower by one third due to income from the rent of one room. If the house- hold continues to rent this room after the plot charges have been paid, their net housing expenditure will be reduced to a mere two percent of total household income, which would provide a substantial margin for investment in housing improvements or in other priority items. Because the market value of the property is due in part to the inflationary rise of land prices, which have increased at a rate about twice 1/ Households which have completed payments for these additional charges resulting from State government legalization and improvement schemes are spending less than ten percent of their income on housing, excluding the period during which they are building. -10- that of consumer goods during the previous few years, the household's financial security is remarkably high. Their equity is the full,market value of the property (between 45,000 and 50,000 pesos in 1973) less a few thousand pesos for land transfer payments and improvements. The worker's unfinished house is built of permanent materials of reasonable quality and, when completed, will be as good as comparable commercially or publicly built dwellings, with the advantage of more roofed and open space. At the time of the interview, there were no in-house connected utilities, although these were being installed by the State improvement programme. Water was available from nearby standpipes and illegally tapped electricity was installed and paid for through bribes to company inspectors. When properly installed and fully legalized, the household's expenditures for utilities will increase by a far.smaller amount than the decrease of its housing expenditures to the State for transfer payments and infrastructure. As indicated above, this and other similar housing budgets fluctuate. Except for gelierally short-term ;repayments foi- land, 1/ the greater part of the faily's expenditure is volun- tary and self-determined. Typically, an owner-builder will invest bet7,een one and two years' income in the purchase of land and on improvements made by the family. Only a small proportion of these costs go towards interest payments, however, as credit is generally limited to that for the land itself. In several cases, owner-builders have obtained informal, short- term building loans from their employers, but these are usually advances against their severance pay and do not appear to involve interest payments. Annual expenditures can vary, there- fore, from a maximum of around one-third of total household income--which is apparently quite common during construction periods--to a low of five to ten percent, according to income level, for only taxes and utilities. Many if not most households in the col,onias populares or owner- built subdivisions (fraccionamientos) and in the colonias paracaidistas (squatter settlements), are either uninsured or partially insured against loss of income through illness or accident. The possession of a rent-free home is therefore a major compensation in reducing expenditures to a minimum in time of need. The owner-occupation of a sizable plot (most are between 150 and 200 sq. meters) allows for rental income from extra rooms, as in Case 20, or for the commercial use of space or buildings as in Case 12. Both chase potential sources of income are prohibited in public housing projects and are often impractical in commercially rented dwellings. In some cases, the primary source of income of an owner-builder household is the rental of 1/ Paid for twice over in this and many other cases as a result of initially illegal sales. rooms built on their plots, a phenomenon observed frequently among Swahili house builders in the cities of East Africa, as well as in Latin America. (b) The Diseconomies of Subsidized Project Dwellings (Case 23) The government employee's contract-built, modern standard dwelling unit is an example of a publicly sponsored housing scheme in which governments build and sell single family units. This single-floor house has 48 sq. meters of roofed space and a backyard patio of 24 sq. meters. While the house has the same amount of roofed area as the unfinished house of the factory worker,.the latter has over five times as much private open space for his family of 11. While the project dwelling is well finished and fully serviced, it is extremely small for its 13 occupants--less than 4 sq. meters of living space per person--and there id no room for expansion, except by illegally roofing the patio or by construction of a second storey for which the structure is not designed. The government employee household made a deliberate choice favoring the offered allocation of the project dwelling over his earlier intention to build on an already purchased plot, although the pattern of motives and priorities influencing his decision are not clear. The head of this three generation family is a member of Mexico's ruling party, and plays an impor- tant leadership role in the locality. As a 56 year-old civil servant, he is due for a pension as well as severance pay and the continuation of fringe benefits after his impending retire- ment. The offer of subsidized dwelling with little or no risk of foreclosure may well have seemed too good to pass by. Although it is perhaps not strictly legal, acceptance of this benefit does not preclude the household from taking advantage of the plot purchased for an owner-built dwelling. With three teenaged children and one daughter of 29 living in the over- crowded home, a second property would obviously be of great value and would provide additional security for the family. The price of the house is well within the household's capacity, at 17 percent of their relatively high income. The unsubsidized or market price of a comparable house would be at least twice as much. A considerable part of the difference is absorbed by the government, through public overhead costs which are higher than those for commercial developers. Although the family owned little or no equity at the time of the interview, since transfer and payments for the house had just commenced, the equity will mount rapidly. Potentially, the occupiers possess the equity represented by the difference between their debt and the market value of the property, which could be realized when policing of illicit transfers is - 12 - relaxed, or as soon as the occupiers are free to let the property at its market rent. Although this particular project is being closely controlled at a high management cost, due to particular political significance, such scrutiny is unusual and not likely to last. It is probable therefore, that this family, like most beneficiaries of publicly allocated properties with below-market rentals, will opt for the cash value of the subsidies by renting or even selling the property rights. The profits made at public expense could then be con- verted into a larger dwelling on the previously purchased plot. Although the governmept employee's household benefits, the net social losses are considerable. Even if construction costs of the project dwelling are comparable to those of smaller and usually more economic building firms, and in the unlikely event that administrative overheads are similarly economic, a major portion of the real cost of this government-subsidized housing is incurred by society. This includes the direct subsidization of production and management costs for land, selection and transfers, carried out by public agencies, plus the costs of planning and designing services, contracting and inspecting construction, and at least some of the subsequent project management costs. The major subsidies arise from the absorption of the opportunity costs of land transfered at prices far below market values, aiid the less direct subsidization of interests. While it is difficult to say who pays for these subsidies, it appears in part t& be a function of the tax structure. The diseconomies and social dysfunctions go beyond subsidizing the better off at the expense of the poor. In some cases, it is possible that the costly publicly-provided dwelling will delay the construction of an owner-built house. In addition, such projects are frequently the justification for slum clearance schemes which disrupt the lives of less fortunate households. Further, publicly sponsored projects frequently support the development of costly and inappropriate technologies and organizations through which building industry contractors and producers can claim self-justifying "economies of scale" and "productivity." These capital intensive and white-collar inten- sive organizations demand large-scale operations and high ratios of managerial and professional time instead of using more appropriate labor-intensive systems. Housing diseconomies are due in part to machinery which cannot serve most people because it cannot be used by them. Thus, in addition to the resetting of policy goals, proper instruments must be found which are based on the resourcefulness of the residents and their local institutions and, enterprises. The case of the Mexican project is as typical as the massive bulldozing of the industrial cities of Europe and North America by machines that cannot deal with the delicate task of upgrading existing structures or the traditional cycles of city maintenance and community development. Basically sound existing structures and community relations are often torn -12- relaxed, or as soon as the occupiers are free to let the property at its market rent. Although this particular project is being closely controlled at a high management cost, due to particular political significance, such scrutiny is unusual and not likely to last. It is probable therefore, that this family, like most beneficiaries of publicly allocated properties with below-market rentals, will opt for the cash value of the subsidies by renting or even selling the property rights. The profits made at public expense cv,ld then be con- verted into a larger dwelling on the previously purchased plot. Although the governmept employee's household benefits, the net social losses are considerable. Even if construction costs of the project dwelling are comparable to those of smaller and usually more economic building firms, and in the unlikely event that administrative overheads are similarly economic, a major portion of the real cost of this government-subsidized housing is incurred by society. This includes the direct subsidization of production and management costs for land, selection and transfers, carried out by public agencies, plus the costs of planning and designing services, contracting and inspecting construction, and at least some of the subsequent project management costs. The major subsidies arise from the absorption of the opportunity costs of land transfered at prices far below market values, and the less direct subsidization of interests. While it is difficult to say who pays for these subsidies, it appears in part to be a function of the tax structure. The diseconomies and social dysfunctions go beyond subsidizing the better off at the expense of the poor. In some cases, it is possible that the costly publicly-provided dwelling will delay the construction of an owner-built house. In addition, such projects are frequently the justification for slum clearance schemes which disrupt the lives of less fortunate households. Further, publicly sponsored projects frequently support the development of costly and inappropriate technologies and organizations through whiTh building industry contractors and producers can claim self-justifying "economies of scale" and "productivity." These capital intensive and white-collar inten- sive organizations demand large-scale operations and high ratios of managerial and professional time instead of using more appropriate labor-intensive systems. Housing diseconcmies are due in part to machinery which cannot serve most people because it cannot be used by them. Thus, in addition to the resetting of policy goals, proper instruments must be found which are based on the resourcefulness of the residents and their local institutions and enterprises. The case of the Mexican project is as typical as the massive bulldozing of the industrial cities of Europe and North America by machines that cannot deal with the delicate task of upgrading existing structures or the traditional cycles of city maintenance and community development. Basically sound existing structures and community relations are often torn -13- down and disrupted by machines incapable of replacing them with econo- mically or socially viable substitutes. The case studies indicate that the essential tools and resources which people commonly lack are the elements of all housing: land, technology and finance. Most commonly absent is land at affordable prices. The prices of building equipment and materials are also increasing and local access to credit may well be shrinking in metropolitan Mexico, in spite of-- or perhaps as a result of--the creation of a new central institution for financing low-income housing. Underdeveloped land is bypassed, and credit is diverted from small users with interest in the development of interstitial land. Disproportionately rising prices limit the capability of small developers and construction firms and owner-builders to provide shelter. Policies directed toward the use of available resources need legis- lative powers over the private use and social distribution of land; over the prices and distribution of building materials, tools and highly paid skills; and over credit for local improvements. There are also many technical and institutional innovations and developments that would increase the capability of local producers. Important recent developments in local government and community enterprise include tenant management and takeover of publicly owned housing schemes as in Oslo, Norway and St. Louis, Missouri, and the modern use of traditional forms of community action as in China and Tanzania. Technical and managerial capabilities are not strictly coincident with political or institutional authurity. The case studies raise two impor- tant issues with regard to housing authority: first, the study questions the effectiveness of central agencies in making housing decisions at the house- hold level, and second, it questions the assumed coincidence between compe- tence, formal education and level of income of persons involved in the housing process. For instance, one of the most efficient investors in our sample is one of the poorest households, that of the illiterate fruit vendor in Case 12. Although in the poorest fifth of the 25 low-income households investigated, the fruit vendor's family ranks in the upper fifth with regard to equity and has one of the largest, most useful and well built houses of the dwellings observed. Issues related to authority in housing are illustrated by the following case study: A Squatter-Builder's Housing Strategy (Case 12) Dona Elena, the head of the 1 9-member extended family, arrived in Mexico City with her young child in 1947. She had left her husband in her home village in a nearby province and lived at first in the servant?s quarters of a middle-class family for whom she worked immediately after her arrival in Mexico City. She left to take up fruit vending and lived with her second husband in a tenement room--an inner city vecindad--until high rent drove the family to a squatter settlement on the northern periphery. After repeated removals by the police, the family moved in 1.952 to the present location in Santo Domingo de los Reyes, a squatter settlement in an abandoned lava quarry on the southern outskirts of the city. The settlement, near a market,and a middle-class residential area, suitable to Doa Elena's trade, also provided job opportunities for her hus- band, a lorry driver. Fearing further forced removal, the family at first lived in a provisional shack. After five years without police .interference, they decided to risk investing in a permanent building, which, they believed, would consolidate their squatter's claim. Despite their very low income,.Doia Elena and her third husband, a brick- layer, have continued this policy and have built a six-room house with a crude shop on the street front. They share the use of this substantial property with the families of two of Dooa Elena's children, and with a third unrelated household which she took under her wing for no apparent material reason. The 12,000 pesos invested in the dwelling evidently represent contributions from other members of the aggregate household, although the illegal property is held by Dona Elena. She estimated its market value at about 45,000 pesos in 1973, which is about five times her own and her husbandts combined annual incomes. If legalized, the property would be worth considerably more than that, but much of the incre- ment would be due to capital gains in the highly speculative free market for building land. As such gains are greatly reduced by the risk factor, the market value quoted is a more accurate reflection of undistorted increases of real monetary value through self-help enterprises of the kind practiced by hundreds of thousands of owner-bmilders in Mexico City. The family's rate of investment in improvements has fluctuated according to its varying income. The large plot of 260 sq. meters has enabled a large aggregate household of 19 very low income persons--including seven under five years of age-- to live in relative comfort and security. To the investigators who spent many hours with the family, it seemed a happy, pleasant social environment. As a vehicle for the achievement of social and economic security for her approaching old age, Dona Elena's housing strategy has been sound. Her housing investment has reinforced her role as an informal elder leader in the local community as well as providing a pleasant and relatively spacious home for her extended family. Should they all leave, Don"a Elena and her present husband could rent several rooms and live modestly on that income, as a supplement to income derived from her shop. The existence of DoVa Elena's street-front shop is an especially relevant illustration of the adaptability of this user-controlled housing system to varying circum- stances. Not long before the interviews, DofYa Elena suffered a broken leg which made it impossible for her to continue street vending. She now continues her trade from a shop built on to the property. The family was able to adjust to the major change in their situation, after the accident, through an arrangement which would not have been possible within the regulations governing officially recognized housing procedures and standards. The case described above, as well as the others, highlights the complexity and variability of household situations and housing needs. A central proposition of this report is that users have a far greater degree of effective control over housing than current policies recognize. The success of housing policies is limited unless this personal and local authority is taken into account. If the use of most resources for housing depends on the will of the users then they ultimately become the decision- makers. On the other hand some of these findings and issues allow for a variety of interpretations as, for example, the extent to which users, or households do in fact control housing resources, and what these resources are. A crucial and highly debatable question is that of the usertf tolerance for mismatches between their priorities and available housing and the degree of flexibility and adaptability that housing systems must have in order to ensure users' satisfaction and the investment of their resources be they human or material. Thus, housing satisfactions are not necessarily a function of housing standards and other priorities besides physical conditions help to determine the degree of satisfaction and, therefore, the extent of willingness to invest savings, skills and time in housing. The case studies show how sensitive housing demands are to personal and household circumstances, and how families in similar social and economic circumstances often have totally different needs for structures, housing services and standards. 1/ A know- ledge of the preference of different income groups aids in determining the impact of housing policies. If the general guidelines tentatively proposed here were concretely established to cover the great majority of personal situations in a wide range of contexts, it would be theoretically possible for a central data compiling system to program the provision of suitable housing for all. This would be difficult to implement, not only for financial and administrative reasons, but also because of the problems involved in "personalizing" insti- tutionally supplied housing. The value of any specification,for a person, 1/ The young car painter's family is satisfied with their tar paper shack and Dona Elena, with an even lower income, is happy with her fanily's spacious brick and concrete house. While both dwellings are admirably suited to their users' respective situations, both would be condemned in any conventional analysis of housing problems. If either were removed or substituted by housing services provided by central agencies, it is likely that nothing but harm and dissatisfaction could result for the households involved. -16- household or community is at least partly dependent on the autonomy of the decision itself. The authority of people in their own localities over their housing is critical, and ultimately must outweigh all other kinds and levels of authority. If authority is taken away it must be converted into police or propaganda power. Policing is administratively and financially impractical in countries with low per capita incomes. In these countries, and probably in all countries in the long run, household preferences should be taken into consideration. This proposition raises unanswered questions concerning the nature of the preferences and the impor- tance of choice. Further work must be done before any useful hypotheses or theories for testing them can be formulated. The data in this report support the principle of local control and central planning. It is evident here that the most economic and socially viable housing developments are those in which the investors and users had access to the essential resources that they lacked, and were free to combine these with those they possessed, and in their own ways. As mentioned earlier, this generalization applies only to housing action at the personal and local levels. It is also clear that local and personal housing action depends on access to tools and resources which only supra-local authorities can guarantee. Particularly, access to land is a critical factor, and increasingly so as this access is being rapidly reduced by rising prices. In addition to land, central government generally has the power and authority to maintain lower-income access to credit and the availability of construction materials, tools and manual or managerial skills. The problem of resource planning and allocation is quite different from the problem of specific housing design and supply. The decision to make a given piece or category of land available for housing, for example, does not tie the developer to one type or specific mode of construction. The investor in improvements will generally have a wide range of alternative combinations of products and procedures which may be employed on any one site which can therefore respond to a relatively wide range of demands. Although the variability of the demand identified in the following chapters precludes centralized housing design, construction and management, the analysis suggests the existence of common patterns from which certain demands can be assessed and predicted and upon which resource planning can be based. While the principle of central planning and local control may be clear and acceptable in any genuinely democratic context, its practice requires the identification of boundaries between local and central respon- sibilities and authority and the institution of complementary norms and procedures. A full discussion of this problem is beyond the terms of reference of this study, which focuses only on the micro-scale of action in the housing and urban development process. A strategy for the successful use of locally manageable instruments, and for their effective combination with legislative controls over basic resources, depends on the differentiation of these various levels of action. Such differentiation is a pre-requisite for the emergence of appropriate institutional structures and effective urban housing policy development. -17- II. THE HOUSING PROCESS AND HOUSEHOLD CHARACTERISTICS The housing sets and systems for metropolitan Mexico indicate the paradoxes of the housing process explained in the previous chapter. If housing is analyzed as a process involving many highly varied actors and activities including sponsorship and planning, design and construction, and management and maintenance, the role of material values and physical standards in the preference structures of different income groups may be studied. For instance, the tar paper shack erected by the car painter in his comadre's backyard provided a highly satisfactory match with the priorities of that particular household (Case 22), and the fruit-seller's even poorer household (Case 12) was equally satisfied with the spacious and solidly built permanent house the family managed to build over the years. When the different situations of the households are known, this complex set of relationships explains housing preferences. This chapter attempts to specify the various elements contributing to housing choices and to demonstrate how the study of the interrelationships of these elements can be helpful in explaining or anticipating particular housing preferences of the users. The identification and analysis of basic relationships affecting the housing process requires a model that is general but allows for the representation of any particular case. The simple general systems model introduced by Bertalanffy 1/ provides a satisfactory frame of reference for studying the housing process. (See Figure 1.) The model is composed of * five elements common to all cases: (1) the context in which the process occurs;'(2) the actors or people and the institutions making decisions and carrying out operations; (3) the activities of the actors or the organization of these operations and decisions ;(4) the achievements of the actors or the physical, economic, and social products of these activities; and (5) the expectations of the actors with regard to the nature and the consequences of their achievements in a particular context. In its most general form, the model makes no explicit reference to the economy of the household. Because this is an essential element in the analysis of housing needs it is treated separate from, although associated with, expectations and housing priorities. The expectations correspond to the motivational component. Housing priorities are closely associated with the expectations brought into the housing process by relevant actors. This study focuses on the users of housing services, leaving aside, for the most part, the analysis of the interests and priorities of other actors who participate in the process. This is why only marginal reference is made to the public and private sectorst involvement in the process. 1/ Ludwig von Bertalanffy, General Systems Theory, ed. George Braziller, (New York, N.Y., 1968). - 18- Figure 1: THE SIMPLE GENERAL SYSTEMS MODEL OF THE HOUSING PROCESS to w 0 u P P =9 P cu 0 3:4 02 02 Context ---Actors---Activities--- Achievements---Context2 Expectations 0 Source: Ludwig von. Bertalanff'y, ~.cit. -19- A. Analytical Components Household Characteristics The descriptive material collected in the 25 case studies shows that two household characteristics usually coincide with budget strategies and priorities. These characteristics are the present socio-economic status of the family group, as measured by total household net income, and the household's expectations for proximate changes in that status. 1/ Other variables influencing household behavior, but to a lesser extent, are: the geographic and cultural origins of the family; the demographic and social composition of the household (age, sex, and the structure of kinship relations); and the residential itinerary, housing experience and socio-economic and occupational trajectories of household members. It is conceivable that in the context of a larger sample, all these variables could be taken into account. However, in the present study, we concentrate on the first two variables--total household net income and current expecta- tions. 2/ Housing Components and Actors Housing, as distinct from household characteristics, describes a set of independently variable components or services. There are a great 1/ The explanatory power of total household net inccme and expectations increases considerably when combined with other variables such as access to resources and budget strategies, which will be incorporated into the conceptual framework in the following sections. The interactions between present income, expectations and utilization of accessible resources, for example, may be illustrated by the case of people without high expecta- tions for the future or without other forms of social security who are unwilling to borrow from employers or relatives so as not to tie them- selves down to particular jobs or localities. Access to resources and skills to utilize resources, such as building land, tools, or materials, may also vary independently of income and, in some cases, of expectations as well. Some important resources, such as political influence, might be accessible entirely per chance, having more to do with who happens to be one's neighbor than with any other factor. Information, imagination, and determination--all critical resources in the housing process--seem more closely associated with personal experience and time spent in the city than with socio-economic status. 2/ Total household net income and current expectations could be combined into a single factor and designated as present prospects of socio- economic mobility, This composite factor is implicitly utilized in the analysis but no attempt is made to formally account for it. -20- many components of housing, most or all of which fall under the. following general categories: (1) networks of components of infrastructure, such as piped water, drainage and sewers, access roads and paths, electricity, gas mains, telephones, and the like; (2) structures and spaces or the buildings and the internal and external spaces they provide; (3) commercial, cultural and civic facilities, necessary for the support of domestic life and acti- vities of local residents. Each component of a particular housing set may be the responsibility of a different actor, institution or sector. In aggregate terms, one of these actors generally dcminates resulting in the classification of sets of housing services according to the predominant actor. Public housing, for instance, is dominated by the public sector in most of its aspects and phases, unless or until it is transferred to the occupiers. Squatter settlements are controlled by the residents themselves so long as they are not incorporated into the physical and administrative structure of the city. The types of housing referred to in this study (vecindades, ciudades perdidas, colonias paracaidistas, fraccionamientos populares and conjuntos subsidiados), represent characteristic mixes of actors and roles in the housing process, as well as characteristic physical forms and set8 of component housing services. When variations occur, as in rent-controlled vecindades in which the users effectively control transfers, there are corresponding changes in the distribution of actors. The combi- nations of components and actors, and the common deviations from them, are suarized in Figure 2. The Association of Household and Housing Characteristics The analysis of housing processes using a disaggregated descrip- tion of housing sets and the mix of actors typically associated with each set, indicates the various elements that may not be interdependent. Thus, households with similar socio-economic characteristics and situations occupy different kinds of dwellings with equal satisfaction (Cases 12 and 22) because the same set of housing services may be of different value for households with generally similar situations. 1/ Likewise, similar sets of housing services can be sponsored, built, or managed by different casts of actors and these variations can greatly alter the value of the same services for the users. This brings us to the functions of housing and how these functions are linked to particular household characteristics. 1/ The blacksmith's household (Case 6) and the data-compiler's (Case 4), for example, have similar incomes, expectations, and family structures, and both occupy inner-city vecindad tenements. However, the data-compiler's household has a better balance between their needs and the housing set they occupy, compared with the blacksmith's family's situation. The observed imbalance in the latter case can be attributed to the fact that the family has virtually no security of tenure in its uncontrolled rent dwelling, despite their urgent need for it at this particular time. The data-compilerfs family, on the other hand, has a substantial equity-- through key money they can demand--which would assist the family in case of a transfer, thus adding to the security they already enjoy from living in a rent-controlled unit. 叫 22 The case studies indicate that perceived user needs vary widely. In order to begin disentangling this complex set of interactions, it is necessary to introduce the notion of housing functions. This study supports the Turner thesis 1/ that the basic housing needs are for location, shelter and security of tenure. Unless a dwelling is accessible to the occupants, unless it provides them with a minimally tolerable degree of services to protect against hostile climatic and social elements, and unless the occupants are sure of staying long enough to make the move worthwhile, the dw _gf little use to them. This hypothesis assumes that'there is not a linear correlation between these three basic fuh( tions. On the contrary., as several of the cases demonstrate, the relative importance of these functions and the levels of tolerance that people have for mismatches between their perceived needs or desires and real conditions, vary signi- ficantly. Examples such as the observed contrasts between Gases 12 and 22, which have one ef the best and one of the poorest quality houses despite similar incomes., and between Cases 5 and 13, two of the poorest of all cases who live at opposite poles of the metropolitan region, should alert the reader to the range of possibilities. 2/ These cases strengthen the argument in favor of a disaggregated functional analysis of the housing process and, specifically, support the proposition that a high priority for any one of these three basic functions of housing can induce equally high levels of tolerance for unsatisfactory matches in other areas. Consequently., each alternative set of relative prioritie8 rovides a widely different range of tolerable options. The alternatives offered to low-income housing users emphasizes the critical linkages between the users' access and utili- zation of resources on the one hand, and motives and expectations of users on the other hand. Expectations and motivations that determine the parti- cipation. of users in housing processes are built up of complex and highly variable sets of factors. I/ See John F.C. Turner, "A New View of Housing Deficit," Housing Poll- for a Developing Latin American EconZa ed. Charles Frankenhoff HS a n Juan: University of Puerto Rico, 1966); John F.C. Turner,, "Housing Priorities., Settlement Patterns, and Urban Development in Modernizing Countries." AID Journal,, November 1968, PP. 354-363; and John F.C. Turner and Robert Fichter, eds., Freedom to Build (New York: The Macmillan Company, 1972). 2/ Household is headed by a teen-age woman who works as a prostitute to support her invalid mother and, appropriately., lives in the central city. Household 13 is headed by a sunflower-seed seller of ill-health who lives on the extreme periphery, also appropriately,, because he works the bQs routes passing by the area and his family prefers the semi-rural environ- ment. Cases 2 and 19 are also illustrative. Both these households are headed by headed by foremen, have upper-lower incomes and are upward mobile,, but their priorities,for security of tenure are quite different. Household 2 is far more interested in getting a house in a middle-class suburb than in securing ownership of their present dwelling., while house- hold 13 is content with its incomplete and unserviced house in a working- class fraccionamiento popular, mainly because they own it and feel secure there. -23- Poor as they are, the ciudades perdidas and other shanty-towns in Mexico City serve useful purposes and their eradication can only make life more miserable, or erode hopes and expectations for the majority that is directly affected and in whose name the supposed improvements are usually made. The news-vendor (Case 7), the street-vendor (Case 8) and the prostitute (Case 5), all live in dwellings threatened with imminent demolition in the name of urban renewal and social improvement. It is most unlikely that any of these unfortunate people will benefit from the proposed public actions. It would not be surprising that the hopes which some, especially the young, may still have of relieving their poverty are destroyed along with their homes and the consequent break-up of their families. 1/ Housing conditions imposed by exogenous agents upon households affect household spending patterns as evidenced among families relocated in subsidized project dwellings with housing that they cannot afford. In other words, the Mexican and the Brazilian 2/ families relocated to dwellings of far higher standards than they could have aspired to in the open market, do what they can to keep them. Although it is unlikely that these families would have voluntarily attempted to obtain improved dwellings at such great costs, once living there they seem anxious to keep the new houses. It is possible, although somewhat in contradiction with the original Turner hypo- thesis, 3/ that priorities for modern housing standards among --ow and even very low?income households are important but often suppressed by more pressing needs. It is also plausible, however, that once relocated, families have few other acceptable alternatives and are not prepared to risk the loss of what they have. On the effects of distorted housing priorities on the consumption of other goods an illustration is given by a Mexican case in which the son of one of the poorest families is spending his wage on a record-player, i.e. on a good generally associated with houses of the kind he and his family were relocated to by exogenous decisions and in accidental circumstances. Another illustration is given by Barney Rush for Brazil, L/ where he found a large number of poor occupants of favelas resettled in public housing, who 1/ Optimistic expectations are not uncomnon. For instance, the 16 year old prostitute looks forward to getting a factory job as soon as she is legally eligible for one, i.e. when she is 18. Destructive effects bearing upon whole household economies are evident in the cases of two very low-income families (Cases 24 and 25) removed to subsidized projects. These families could only spend 7 or 8 percent of their income in housing withoU+ risking malnutrition but are obliged to spend 35 and 55 percent respectively in order to meet rents and utilities payments. An even higher distortion would become apparent if transportation costs were added to the account. 2/ See Barney,S. Rush, ap. cit. 3/ John F.C. Turner, op. cit., 1966. 4/ Barney S. Rush, op. cit. were investing in improvements in these new dwellings even though they could not or would not pay the rent-purchase monthly quotas. 1/ The point is not so much that everyone enjoys living with higher material standards, but rather that what people do when they have to make their own trade-offs and decisions between comforts and other priorities, differs quite signi- ficantly from their actions when decisions are imposed by outside agents. In fact, priorities for high standards in housing are rare at the lowest levels of income. By and large, it is only when the most critical priorities for present or future security are satisfied that people will invest in higher than essential housing standards. 2/ Although there are extremely wide variations in the proportions of income spent on housing, even among those least affected by exogenous incentives and constraints, the housing expenditures of the households studied indicate that the normal percentage of income willingly invested in housing services is about half that assumed by many official agencies, ranging from 12 to 13 percent rather than from 20 to 25 percent. 1/ The most camon investments were in new floors which are far from cheap. 2/ When in fact families voluntarily invest in housing, the results are surprisingly positive, as in the case of the large and well-built house constructed by the fruit-seller's family in Case 12. III. HOUSING PRIORITIES, EXPENDITURES AND RESOURCES One of the starting points of this study is that all:people share a number of deep and similar needs of the kind Abraham Maslow identified. 1/ Given this communality of basic needs, this chapter explores some proposi- tions regarding the priority ranking that people attach to them and the implications of that to practical spheres of decision making. First, it is argued that priorities among universal needs, such as those for security and continuity of life, for love and identity, for excitement and creativity, vary according to peoplets changing life situations. Second, whatever the priorities among universal needs are, these are expressed in the form of priorities among specific needs for things which are vehicles or means for the satisfaction of vital needs. To illustrate, this means that the way in which a family is housed is bound to affect their security, identityg and their opportunities to engage in creative activities. The effects bf housing may sometimes be marginal but can often be critical to the family as in the case of life under harsh climatic or social conditions. The implication of these behavioral linkages to this study is that the changes in relative priorities among vital needs that occur over time are attached to correspondingly different housing demands. 2/ Thus, when housing services are being used primarily to satisfy underlying needs for security more than for identity, for example, the inalienability of tenure is likely to take precedence over the appearance of the dwelling itself and the social status associated with that appearance. The data from the case studies indicate that the most vital need among low-income households is for security. To a lesser extent, a cluster of non-economic needs are also significant. Behavioral orientation is either present-oriented or future-oriented. Gases of future-oriented behavior were identified as those activity patterns implying both risk-taking in relation to peoplets present security and the intention to maximize chances for an improved situation in the future. Risk-taking behavior was minimized among those with present-oriented security patterns. 72 percent of the households 1/ Abraham Maslow has suggested that the needs of human beings are hier- archical and, as each level is filled, the subsequent level becomes salient. This order of needs is: (a) physiological requirements (food, habitat, etc.); (b) safety and security; (c) companionship and affection; (d) self- esteem and the esteem of others; and (e) self-actualization (being able to realize one's potential to the fullest). See Abraham Maslow, Motivation and Personality (New York: Harper and Row, 1954). 2/ It is conceivable that priorities with regard to the satisfaction of vital needs may be different for different socio-economic groups. As this study focuses only on low-income households, however, the issue of class variation is not explored. -26- interviewed showed activity patterns characterized by needs for security. In most cases (16 households) the activities tended to be present-oriented. Only two households appeared willing to take risks in order to achieve future improvements in their security level. 1/ Five other households, also among the poorest, attribute secondary priority to seeking and taking oppor- tunities. Given their low level of expectations for the future, they place the consolidation of their present security above future possibilities. The cluster of basic non-economic needs is more difficult to specify. These are generally needs for attaining a "middle-class life- style" and are rare or absent among households living close to subsistence level. Among the cases studied, only those with relatively high expecta- tions and at the same time incomes above the group mean were primarily motivated by non-economic needs. Only two households, one headed by a securely employed factory foreman and the other by a dental technician, clearly oriented their efforts towards achieving middle-class life-styles. Six other households, also among the relatively better-off, indicated a second-rank priority for non-economic needs. 2/ Budget Strategies The ways in which households use their material resources and budget their expenditures are associated with their general needs and priorities. However, these behavioral linkages are too general to explain all the observed variations in specific housing priorities and expenditures. We can go beyond an explanation of the relations between basic needs and specific housing needs for households in different socio-economic and life cycle groups 3/ by adding the "budget-strategy" variable. This variable refers to the overall earnings and expenditures of a household. The overall budget strategies which were identified for the 25 households fall into three main categories: (a) cost minimization, (b) budget optimization, and (c) other strategies. The first type, cost minimization, characterizes households whose budget strategy is one of spending as little as they can on whatever they are forced to spend money on. 4/ This budget strategy is frequently associated with very low income households having negative expectations for the future (Cases $ and 13 are 1/ These two households (Cases 22 and 14) have high expectations for the future and very low present income. Both, therefore, have little to lose and plenty to gain by seeking and taking advantage of opportunities to improve their present situation. 2/ For a discussion of the role of material and existential needs and priorities in shaping housing demands, see John F.C. Turner and Robert Fichter, eds., op. cit., pp. 164-69. 3/ Ibid. h/ A similar concept is introduced by Robert A. Hackenberg when he discusses "income conservation /as a7strategy of the lower class /EestinZ7 on under- consumption of all durable goods." See Hackenberg, "The Poverty Explosion: Population Increase and Income Decline in Davao City, 1972," Davao Action Information Center, mimeo, 1974, p. 39. -27- illustrative). The second and most common type of budget strategy is that of cost optimization, a label for the behavior pattern of 18 of the 25 cases, who make deliberate trade-offs between various expenditures, so as to maximize a particular household utility function. The third type is clearly a residual one, bringing together budgetary rationales that have in common only the fact that they are not preponderantly guided by economic criteria. Cases identified under this residual type form a mixed set, covering the budgetary effects of an alcoholic household member (Case 8), conspicuous consumption by households aspiring to middle-class life-styles (Cases 1 and 2), and relocated households whose budget strategies seem largely dictated by factors stemming from the relocation (Cases 24 and 25). 1/ While specific housing priorities and cost minimizing and opti- mizing are, in a general sense, a function of the household's priorities among basic needs and of their budget strategies, there are often significant differences in the outcome of the ccrbination of apparently similar factors. Most discrepancies from the general pattern occur when housing services are not the prime vehicle for the satisfaction of a high priority basic house- hold need. There are several cases, for instance, in which present and future employment opportunities are priority goals for households in which the users would probably trade off housing services for easy access to jobs. In general, the case studies show: (a) the coincidence between the need for security in general and a secure tenure of dwelling in particular; 2/ (b) the priority for future security is usually associated with priority for housing location; 3/ (c) a recurrent coincidence between general non-economic priorities and priorities for dwellings of high material standards. 4/ 1/ Imposed strategies might be a better term to convey the reasonable assump- tion that neither of these two families who were relocated would be spending 55 and 35 percent of their very low income in high standard dwellings if they had been left to make their own decisions. 2/ 70 percent of the cases that rank general security very high also give top priority to secure tenure of their dwellings. 3/ All eight households in which opportunity-seeking is a high priority also have a high or very high priority for physical proximity between residence and work place, or for housing location in certain areas of the city, or for location near peer and kindhip groups. The other seven cases with high priority for location appear to be motivated by other reasons, not so obviously related to their general priorities. 4/ The eight households with primary or secondary general priority for non- economic needs expressed either top or secondary priority for housing standards. The corollary is not so strong as in the previous proposition, i.e. there are 15 cases of households with significant priority for material standards and comfort which cannot be clearly explained in terms of their general priority needs. -28- Household budget strategies are even more closely related to specific housing priorities and expenditure patterns than are basic reeds. Again, significant discrepancies can be found but the analysis of these would require a larger sample of households than that available in the present study. In general, the observed association between housing priorities and budget strategies are: (a) households that minimize costs are also those that minimize housing expenditures; 1/ (b) households that optimize their general budget also tend to optimize their expenditures on housing; 2/ (c) households that maximize their housing expenditures are either those oriented by exogenously set priorities or those going through an intense investment period which deviates from their normal budget strategies. 3/ It is necessary to know the major resources commonly available for housing purposes, and the factors influencing the use of resources. Monetary savings and allocation of current income do not represent all available resources. The question of how money is used depends on the availability and use of complementary resources. Likewise, the social and material values derived from similar sums of money depends upon the combi- nation of monetary and non-monetary resources. The many rescurces commonly used by households in the sectors investigated can be grouped as follows: (a) financial resources - accumulated household savings - percentage of current income from regular jobs and earnings from outside activities - gifts and loans from relatives - wage advances and loans from employers - commercial and public sector loans, including rent- purchase of building plots 1/ Hackenberg associates "income conservation strategy" with low class families that eliminate housing and utilities' expenditures from their budgets. See R.A. Hackenberg, op. cit., p. 39. 2/ Among the 20 households classified as "budget optimizers" 18 are also optimizing their housing expenditures. 3/ There are eight cases of variations between budget strategies and housing expenditure strategies, five of which are households that minimize their housing expenditures as part of their optimization strategy. Households maximizing housing expenditures are typically those who have built their own homes. -29 - commercial, bank or public mortgage credit for complete dwellings (b) other material resources - free land (through inheritance, membership in an ejido cmmunity, or squatting) - useable materials on site - waste or discarded materials obtained from employers or elsewhere - free utility services (e.g. public water taps and pirated electricity). - garden produce and livestock generating income (c) local hman resources - spare-time for labour and management - construction skills - management skills - cooperative organization (between extended family networks or neighbors) (d) supra-local societal and institutional resonrces - national or regional political organizations and representatives - charities and other private institutions or organi- zations - public housing agencies (providing programmed services and projects) - other specialized agencies (providing specific services, such as land transfers and entitlement). Another category of factors, which may be regarded as resources, are facilitators of resource utilization including: (a) 'information and the networks through which information is obtained (b) imagination and the capacity to anticipate change -30- (c) initiative or enterprise (d) determination and sustained effort (e) tolerance for sacrifices demanded by courses chosen (f) incremental or progressive development and change, allowing for the maximum use of resources through the flexibility and adaptability of the process. The variety of resources available to households, together with the variety of motives determining their use, illustrate and to some extent explain the immense variety of housing demands observed among lower-income sectors that have access to this range of resources and at least some freedom to utilize them. The case studies support earlier arguments that managerial skills are as important, or even more important than manual skills and that it is misleading to suppose that self-help potential lies in the latter alone. It is difficult to quantify the gains made from specific operations that have already been carried out, and the study could not include a sample for the monitoring of building procedures. However, it seems clear that equally considerable sums are saved by low-income Mexicans as by relatively very high income US owner-builders. I/ As the use of these and other monetary and non-monetary resources demand considerable time and effort on the part of the owner-builders, motivation is obviously vital. The relatively low priorities for home ownership among the households with substantially higher incomes (as for example Cases 1, 2, h, and 6 who are vecindad-renters) have evidently been insufficient to overcome the inhibition of their capacity to invest derived from their access to excessively low rents. It is also important to note that informal financial resources are commonly used. Several households reported loans from employers, some of them fairly substantial and all probably essential to the realization of their housing achievements. Not infrequently, these credits are interest free and, in all cases, they involve direct, simple and therefore, cheap arrangements between borrower and lender. Recent developments creating a centralized public housing mortgage system (INFONAVIT) might affect these informal credit lines. 1/ See John F.C. Turner and Robert Fichter, o2. cit., for evidence on US owner-builders. -31- IV. PATTERNS OF HOUSING PREFERENCES The time periods covered by the case studies do not coincide because of the different arrival dates of each household and of various household members, thereby complicating the identification of precise pre- ference patterns. Continuing research and the accumulation of more data should simplify the analytical process somewhat. Until such time, the propositions in this chapter provide a working hypothesis. Although recurring patterns do emerge, it is only in the context of in-depth studies of household decision-making processes that these patterns may be observed in order to determine household preferences. The 25 Mexican households show three basic types of budgeting: (1) the poorest households (Cases 5 and 13) minimize all expenditures and spend only on essentials; (2) the majority of households optimize their expenditures to varying degrees, including those with very low incomes, 1/ and (3) a few households budget on the basis of priorities for non-essentials, such as the desire to enhance social status (Case 2) or to enjoy environmental amenities (Case 1). Two of the three exceptions to this are the very low income families whose budgets have been grossly distorted by external inter- ventions in their housing situations (Cases 24 and 25). The third exception (Case 8) is similar, in that the family is threatened by eviction from its shanty town dwelling, 2/ and the fear has exacerbated the alcoholism of the elder male, thus making the purchase of liquor a dominant item in the household budget. In the majority of cases housing expenditures are closely asso- ciated with household budget strategies, reflecting the same general break- down of spending. Three general categories are: (1) cost minimizers or households that minimize housing expenditures; (2) budget optimizers for whom housing services have a low priority and (3) housing expenditure maxi- mizers or households investing all they can in housing (typically owner- builders who use housing to secure their positions or to gain access to higher socio-economic levels). In the 1968 sample, housing expenditure maximizers clustered below or at the median household income level of the low income group and on the side of positive expectations for the future. 1/ Although low-income families have very small savings margins, and those with very low incomes have none, still all but the most desperate or depressed make deliberate trade-offs beiween expenditures on essential items. Cases 12 and 22--the fruit-seller and the car-painter families-- are typical examples of budget optimization despite their very low incomes. The first selected to invest in housing at the expense of other expenditures, while the latter minimized housing expenditures in order to be in the position to take full advantage of opportunities demanding a move at short notice. The contrast illustrates the important differences between general budget strategies and housing expenditure strategies. 2/ Ciudad perdida La Marranera. - 32 - The housing expenditures and priorities of households are influenced by known and expected housing services and by the services that have been made use of by the family. The case studies suggest that government interventions in the housing market are especially powerful determinants of housing expenditures and priorities. The original Turner hypothesis for the distribution of housing priorities is confirmed by this study, although there are some modifications and refinements suggested by the data. All levels or kinds of preferences are the products of priorities between the needs for residential location, shelter, and tenure, and between the essential means for satisfying these needs, that is, the knowledge of what is available, the capacity to use that knowledge, and the means to implement or pay for the chosen housing alternative and to maintain the services associated with it. Among the suggested modifications of the Turner thesis is a redefinition of the locational factor and a division of this factor into two sub-factors. The case studies show that there are two principal criteria determining household decisions with regard to location: 1/ (1) socio-economic location and (2) geographic location. The observed data show that the latter is less important than originally supposed by Turner. Only three out of 25 households give top priority to "geographic" location. The fact that the three households that gave top priority to inner-city location also have occupations in that area, further reinforces the need to re-define the concept in socio-economic terms or to sub-divide location into two component factors. Households expressing high priority for inner-city location for other than socio-economic reasons (six cases including the three mentioned above), are all inner-city 2/ or inner-ring 3/ residents, suggesting that they are strongly influenced by their present situations. There are only three cases in which locational priorities for inner-city residence are expressed independet of inner-city job locations. 4/ Housing location priorities point to the need for examining the location of employment opportunities. In Mexico City employment opportunities are widely scattered. In earlier stages of metropolitan growth, it may well have been true that most employment opportunities for the least skilled-- and especially for recent migrants--were concentrated in the densest areas of the city, but this is no longer the case. A look at the employment and residential itineraries of the households indicates that these itineraries are highly coincidental and that they are taking place mainly outside the central areas. This explains why the great majority of the very low income households (nine out of 10 cases) have top priority for housing location 1/ 'Or their level of satisfaction with a given location. 2/ Vecindad residents. 3/ Ciudad perdida residents. 4/ Cases 1, 4 and 6 which, incidently, have incomes above the sample mean. -33- close to their work or peers, or both, but not necessarily to inner- city location. 1/ Security of tenure, or the assurance of continued residence in a given dwelling, is also a high priority for most of the very low income households. This finding also suggests a modification of the original hypothesis in its proposition that there is a need for residential mobility in order to maximize earning opportunities among very low incone groups. Only two families (Cases 14 and 22), both with high expectations for the future, seem to attach considerable importance to residential mobility. All other very low income households search for secure tenure, this priority coincides with their expectations that secure housing is needed to anchor the 'family as much as possible. While priority for secure.tenure predominates at the intermediate income level, the most frequently observed second priority among this group is for improved housing standards. 2/ Locational factors, as a second priority, appear only in five of the 13 intermediate income households. 3/ The predominant concern for security of permanent residence at the inter- mediate income level must be considered in light of the past income experience of the families, all of which had had major income gains during the working lives of their members and presently seem to have reached their upper mobility ceilings and the desire to secure what has already been achieved. Under present circumstances, the only really secure housing tenure in Mexico City is pbssession, de facto or de jure. Among the households studied, those that have secure tenure are preponderantly owner-builders, 4/ beneficiaries of very low frozen rents in centrally located vecindades, 6/ and a case benefiting from a subsidized government housing project. In all probability the families which have acquired secure tenure through rental or other types of subsidies, would have built their own homes if not for the exceptional circumstances introduced by government action. Thus, in general 1/ The exception which proves the rule is the factory watchman's family (Case 17) in which the family lives at a distance from the principal earner's work place limiting lodging time spent with his family. 2/ 12 of the 13 cases with incomes between four and seven times subsistence. 3/ These households have incomes of 6S to 73. They form a cluster of high locational priorities which seems different in kind from the cluster made up of very low income households. However, the number of cases is not sufficient to prove that this is more than a coincidence. 4/ Ten out of 15 cases at the 3S to 7S income group. 5/ Cases 2, 4, and 6. -33- close to their work or peers, or both, but not necessarily to inner- city location. 1/ Security of tenure, or the assurance of continued residence in a given dwelling, is also a high priority for most of the very low income households. This finding also suggests a modification of the original hypothesis in its proposition that there is a need for residential mobility in order to maximize earning opportunities among very low income groups. Only two families (Cases 14 and 22), both with high expectations for the futule, seem to attach considerable importance to residential mobility. All other very low income households search for secure tenure, this priority coincides with their expectations that secure housing is needed to anchor the Tamily as much as possible. While priority for secure tenure predominates at the intermediate income level, the most frequently observed second priority among this group is for improved housing standards. 2/ Locational factors, as a second priority, appear only in five of the 13 intermediate income households. 3/ The predominant concern for security of permanent residence at the inter- mediate income level must be considered in light of the past income experience of the families, all of which had had major income gains during the working lives of their members and presently seem to have reached their upper mobility ceilings and the desire to secure what has already been achieved. Under present circumstances, the only really secure housing tenure in Mexico City is possession, de facto or de jure. Among the households studied, those that have secure tenure are preponderantly owner-builders, h/ beneficiaries of very low frozen rents in centrally located vecindades, / and a case benefiting from a subsidized government housing project. In all probability the families which have acquired secure tenure through rental or other types of subsidies, would have built their own homes if not for the exceptional circumstances introduced by government action. Thus, in general 1/ The exception which proves the rule is the factory watchents family (Case 17) in which the family lives at a distance from the principal earnerts work place limiting lodging time spent with his fam:nily. 2/ 12 of the 13 cases with incomes between four and seven tines subsistence. 3/ These households have incomes of 6S to 7S. They form a cluis'er of high locational priorities which seems different in kind from the cluster made up of very low income households. However, the nunber of. cases is not sufficient to prove that this is more than a coincidence. 4/ Ten out of 15 cases at the 3S to 7S income group. 5/ Cases 2, 4, and 6. -341 - households with low, but not very low incomes have very high preferences for owner-occupancy, even when this .means building their own houses over extended periods of time. First, it is important to distinguish between what households could and would invest in their housing if the resources and services were available to them. Second, there are differences between what households can do with the resources and services actually available to them and what they do in practice. These distinctions bear considerable relevance for policy formulation and implementation in the long and short run. In the case of Mexico, the gap between existing and available key resources, such as building land, is very wide for a majority of poten- tial investors in low income housing. The same is true of financing, building tools, materials and skills. The field observations show that structural changes in the market for land and credit are essential for the closing of this growing gap and, therefore, for any major improvempnts or even the stabilization of the housing situation. In terms of direct housing expenditures--in rent or amortization and rates and utilities--household spending patterns are distributed as follows: (a) those spending substantially less than they can afford (16 cases); (b) those spending much more than they can afford (7 cases); (c) those having a level of housing expenditures proportional to their incomes (2 cases). This distribution 1/ and corresponding investment estimates substantiate the proposition that there is a gap between what people can invest in housing and the available housing. 2/ For the 16 households in situation (a), the estimated total underinvestment is of the order of 65,000 pesos per annum, while for the 7 overspenders it is probably about 20,000 pesos per annum in excess of their combined capacity. Interestingly, over half of the total estimated underinvestment (33,400 pesos) is attributable to four upper-lower income households living in rent-controlled vecindades, and nearly half of the total over-investment (9,500 pesos) is associated with two very low income families occupying project dwellings. This suggests that a possible major constraint to the release of pent-up needs is introduced by legis- lation and public administration. When indirect effects of government 1/ If the conventional norm were used, i.e. that households can spend 25 percent of their incomes in housing, irrespective of their income levels, then 18 families would be underspending, h would be within 5 percent of the norm, and 3 would be overspending. 2/ A similar pattern has been observed in Tanzania. See John F.C. Turner, "Consultant's Report to the World Bank Appraisal Mission for the Sites and Services Project of the Government of Tanzania," IBRD, August 1973, pp. 6-10. action in housing are taken into account, such as legislation inhibiting the construction of low-rental accommodations, the case for increasing present forms of government housing expenditures (or imposing further price controls or raising standards) becomes untenable. In the present Mexico City context, low-income groups have basically three alternatives for housing: (1) low-rental tenements, (2) urban farmsteads, and (3) undeveloped plots, allof which already exist,or could be generated on an adequate basis through politically feasible policy changes. Sites and demands for low rental tenements can be found in all sections of the metropolitan area, including the central or inner ring zone. All the households living in ciudades perdidas, in condemned tene- ments or in extremely rough, provisional shelter 1/ could afford the rent for simply built, short-life courtyard tenements. If appropriately located, these households would probably opt for this alternative. Urban farmsteads are conceived to accommodate households with virtually no income. The option involves a plot of land sufficient for sub- sistence, cultivation and raising livestock. The plot could be used to feed the households and possibly provide some income. Tracts of land held in reserve for future urban development could be rented by the government or private sector thereby serving the housing interests of the poorest of which the sunflower-seed vendor's family is a case in point. Given the high priority that this group attaches to location, tenure, or both, farmsteads would best meet the demand if they were well scattered between and beyond suburban developments. 2/ Minimally serviced plots make up the third option for the low income households. These plots are conceived as "progressive development" sites equipped with a few basic services which are upgraded according to the needs of the residents. This option is intended to satisfy households with high priorities for secure tenure and sufficient margins for investments in improvements. Ten of the 25 households would be best served by these owner- occupied, unmortgaged plots for the staged construction of their own dwellings and for stages infrastructure installation. The argument for this option is that, at the lower end of this income group, priorities are overwhelmingly centered on security of tenure. When incomes and expectations rise, priorities for higher physical stmnd-rds also rise, and these are generally sqtisfied in the form of improved !wnl ing and locality. This match is mainly dictated by the nature of the household economy which may tie them to a particular locatiLon where other altern,tivos are not available at prices they can afford. 1/ This group comprises Cases 7, 8, 0 and 22. 2/ Clearly, however, those attributing critical importance to location are more likely to find the best match with their priorities in a tenement, as these can be provided in almost all sectors of the city. -36- At higher income levels, particularly when households have reached six or seven times the subsistence level, the option of fully developed plots can be considered. This option seems particularly attractive for households holding high priorities for both ownership and improved housing standards, such as Cases 15, 16 and 23. All these have incomes high enough to pay for serviced plots and still build their own dwellings of moderately high standards over relatively short periods of time. 1/ At even higher levels of income, households can be adequately served by rented or condominium apartments in more central locations. 1/ Two of these households, Cases 1 and 16, have actually done just that, and the third, Caso 23, has bought a plot but was diverted from building by the offer of a subsidized project dwelling. -37- V, SUMMARY In sumation, the following matches income levels with housing options taking preference patterns into consideration. Income Levels Housing Options Subsistence 1 or below Semi-subsistence urban farmstead Subsistence 1 to S2 Lowest price tenements and farmsteads Subsistence 2 to S3 Undeveloped plots, when household also have high expectations, or tenements, or farmsteads Subsistence 3 to S5 Undeveloped plots and tenements Subsistence 5 to S7 Fully serviced plots, rented or condo- minium apartments, for those with high expectations; undeveloped plots and tenements for the others Subsistence 7 to S10 There will be a few stable or upwardly mobile households that will opt for tene- ments or for minimally serviced plots; in addition to the apartment and serviced plot options, however, those above the 7 level can afford the rent-purchase of modern standard single-family dwellings in contract-built developments Above subsistence 10 Those households with incomes 10 times the minimum subsistence level or above are beyond the terms of reference of this study and are likely to choose between the rented or owner-occupier apartment or single-family dwelling as supplied by commercial developers. Only those with very high incomes indeed a likely to build privately designed and built homes, which, of course, embody all the functional advantages enjoyed by the lower- income owner-builders. ANNEX Page-T -o 4 pages METHODOLOGICAL NOTE ON THE SELECTION AND 1EXECUTION OF THE CASE STUDIES I. SELECTION The 25 case studies in this report have been selected from a sample of approximately 500 households which were interviewed in the course of three research projects directed by the investigators. These projects are: (1) "Low Income Housing Study.," Master Plan Office, Federal District Government, 1972-1973; (2) "Low Income Settlement System Study in Metropolitan Mexico City," UNAM-National University, School of Architecture, Spring 1973; and (3) "Low Income Housing Study," UIA-Iberoamerican University, Spring, Summer and Fall of 1973. No precise level of confidence has been estimated for the sample size, but on the basis of extensive field experience it can be taken to represent the main types of low-income localities in the city. The selection of localities was based upon the analysis of: (1) Census employment and income data; (2) housing quality data from the Census and other sources; (3) data on the quality of services and utilities from the Census and from branches of local and city governments; (4) environmental quality data collected by the investigators; (5) specific materials and studies on relevant aspects of the system such as low quality housing, illegal sub- divisions, and areas under rent control. Within each selected locality, families have been randomly chosen to be interviewed. Mexican university students and social workers interviewed 500 households using a uniform questionnaire. The average interview time was between 60 and 90 minutes per family. From these 500 cases, 25 families were interviewed in greater depth. Each case study consists of: (a) A narrative report covering basic household/housing characteristics, background history of the household, socio-economic and housing character- istics over time and priorities over time. (b) Tables and charts showing household composition by age, sex, pera- nence of residence, education and occupation; monthly income, expenditures, location of residence, residential itinerary, tenare situation, history of household's relationship to housing market, shelter quality and housing priorities. (c) Maps indicating the geographic location of the residence in Me1Jico City and the proximity of the household to educational, health, recreational, shopping and transportation facilities. (d) Floor plans of the dwellings. ANNEX Page 2 of 4 pages The case studies provide examples of households occupying low cost tenements (6 cases), shanty towns (2 cases), low cost subdi-visions (11 cases), squatter settlements (2 cases) and subsidized housing projects (4 cases). The attached schematic map of the metropolitan area indicates the geographic distribution of the localities covered by the case studies. CASE STUDIES BY SUB-SYSTE4S AND LOCALITY Sub-System Locality Case No. Vecindades (low cost tenement) A. Tepito 1 to 4 Vecindades (low cost tenement) B. Centro, 1st block 5 and 6 Ciudad perdida (shanty town) C. Magdalena Mixhuca 7 and 8 Fraccionamiento popular (low D. Magdalena Atlazalpan 9 and 10 cost subdivision) Colonia paracaidista (squatter E. Sto. Domingo de los 11 and 12 settlement) Reyes Colonia paracaidista (squatter F. San Nicolas de 13 and 14 settlement) Padierna Fraccionamiento popular (low G. Tlacoligia 15 and 16 cost subdivision) Fraccionaniento popular (low H. Chamapa 17 to 21 cost subdivision) Conjunto subsidiado (subsidized I. Sta. Cruz Meyehualco 22 housing project) Conjunto subsidiado (subsidized J. Vicente Guerrero 23 to 2 housing project) II. CASE STUDY PROCEEDINGS Interviews were conducted from July to Septinber 1973 by Mexican university students and by one of the investigators. Each case study included a nunber of structured standardized interviews and unstructured informal con- versations with the members of the fanily. The case studies are divided into four sections. The first section covers basic data on the family history, with emphasis on the economic and ANNEX Page 3 of 4 pages demographic characteristics. The second section includes information on place of origin of the household head and his wife and about their life at the place of origin. In addition, the second section examines each place of residence before the move to Mexico City and records visits to Mexico City before installation in the city. The third section covers the family history as concerns residential moves in the city. The following aspects of each placo of residence were recorded: changes in family cycle, changes in employment and income situation, other important changes in the house- hold situation (such as rural family assets) and, important contextual changes either in the locality or in the city that affect the family. The fourth section focuses on family priorities and their changes over time. It includes a discussion of housing preferences in relation to priorities for locality and other expenditures. The priorities were discussed within the boundaries of the economic possibilities of the families, assuming no housing supply constraints. The last questions on priorities were designed to facilitate the ranking of priorities and trade-offs between different categories and within them. Transcripts of the case studies are available at the Urban and Regional Economics Division, Development Economics Department, IBRD. 叫