Cover Image
close this bookSupport Measures to Promote Rental Housing for Low-Income Groups (HABITAT, 1993, 132 p.)
View the documentA. The nature of owners, tenants and sharers
View the documentB. Preferences for owning, renting or sharing
View the documentC. The choice between different kinds of non-ownership
View the documentD. Who invests in rental housing?

A. The nature of owners, tenants and sharers

Work during the last 10 years in a number of third-world cities has discovered some basic similarities in the kinds of households that own, rent or share accommodation. The descriptions cannot be totally precise, however, because there are important differences between cities. In addition, there is usually considerable overlap between the characteristics of families in different kinds of tenure. As Gilbert (1993: 141) found in a comparative study of tenure in Caracas, Mexico City and Santiago de Chile:

“the basic finding must be that in none of the three cities can owners, tenants and sharers be separated into clear socio-economic groupings. There were too many similarities between owners and tenants, and even between tenants and landlords. Owners, tenants and sharers do not constitute homogenous groups and no single factor seems to determine whether households are tenants, sharers or owners.”

Nevertheless, the following patterns are common in third-world cities.

Tenant and sharer families tend to be younger than those of owners. Certainly, in Santafe BogotCaracas, Guadalajara, La Paz, Mexico City, Puebla and Santiago de Chile, tenants and sharers tend to be drawn from a younger age group than owners (Beijaard, 1992: 71; Gilbert and Ward, 1985; Gilbert and Varley, 1991; Camacho and Ter 1991). In Africa, a similar pattern has been found. In Blantyre and Lilongwe, whereas more than 90 per cent of owner-occupier heads of household are aged 30 or over, nearly 50 per cent of tenants are under 30 (Pennant, 1990: 194). In Kumasi, owners are older than renters and roomers (Tipple and Willis, 1989: 17) and tenants are younger than owners in Tanzania (Peil and Sada, 1984: 296). If the average age of tenants is generally younger than that of owners, it should also be remembered that many older households also live as tenants (Beijaard, 1986, Edwards, 1982; Rivas, 1977).

In Delhi, while tenants are generally younger than owners, less than 1 per cent of tenants are less than 25 years old (Wadhva, 1993: 52). The reason is that young people in India do not set up their own home but share accommodation with parents until they are older. Indeed, sharing is very common in Indian cities. In Delhi, 37 per cent of landlords provide free accommodation for kin (ibid.). In Benin, the age of tenants depends greatly upon the kind of accommodation. In the housing supplied by employers, the main beneficiaries are more senior employees so that virtually all tenants are over 35 years of age (Ozo, 1993: 36). Sharers tend to be the youngest age group with the majority in the 20-40-year age group, they are generally unmarried people or are households at an early stage of their life cycle (ibid., 36).

Tenant families tend to be smaller than owner households. In Santafe BogotCaracas, Guadalajara, Mexico City and Santiago de Chile, owners have larger households than non-owners (Gilbert and Ward, 1985; Gilbert and Varley, 1991; Gilbert, 1993). In Delhi, the average size of tenant households is also much smaller than that of owners: three compared with five (Wadhva, 1992: 98). In Malawi’s two main cities, owner households contain an average of 6.6 people whereas tenant households contain only 3.7 (Pennant, 1990: 194). Owner-occupier households are larger because owners tend to be older and consequently tend to have more children. In addition, in Latin America, Turkey, and in parts of East Africa self-help ownership is particularly attractive to households with large families. In West Africa, owner households are also much larger than those of tenants but for a different reason; the tendency for kin to arrive from the countryside and move in with their home-owning urban relations (Peil and Sada, 1984; Tipple and Willis, 1989).

Tenant families tend to be poorer than owner families. In Indonesia, Hoffman et al. (1989: 2.6) find that “over 75 per cent of renter households have incomes... less than the average household income.” In Santiago de Chile and Mexico City, the incomes of owner households are higher than those of non-owners (Gilbert, 1993) and in Quito “...occupant income increases from rented rooms, to shanties, to houses” (Klak and Hotzclaw, 1993: 270). Similarly, in Ahmedabad, NIUA (1989a: 21) find that

“86.4 per cent of households with incomes about Rs. 2500 had their own houses whereas only 54-56 per cent of households with incomes lower than Rs. 700 owned any house in the city” (NIUA, 1989a: 21).

In the case of Kumasi, Tipple and Willis (1989: 28) report that although some tenants are as affluent as most owners, “roomers - who are just a majority of households include many of the poorest and least advantaged.” In Delhi, Wadhva’s (1992: 98) survey of 200 households found that while 44 per cent of tenants had monthly incomes lower than Rs. 2500, only 7 per cent of landlords had incomes as low as this. In Cairo, sharers are concentrated among the lowest income groups (Serageldin, 1993: 27).

It is important to distinguish, however, between household incomes and per capita incomes. Because tenant families are much smaller, they often have higher per capita incomes. This is certainly the case in Caracas, Mexico City and Santiago de Chile (Gilbert, 1993). Sharers in Mexico City have lower household incomes than other tenure groups but are more affluent than many of the owners in per capita terms. A similar pattern is also true in Kumasi insofar as the extended family system gives owners large households and consequently reduces their per capita incomes (Tipple and Willis, 1989). Linked to this point is the fact that there are substantial groups of very poor families in most third-world cities who move into owner-occupation because they cannot afford to rent. This is particularly likely where the invasion of land is permitted as in Caracas, Mexico City and Jakarta (Hoffman et al. 1989; Coulomb and Shez, 1991; Camacho and Ter 1991). In Caracas and Mexico City, both household and per capita incomes of poor owner families living in invasion settlements are far below those of most tenants (Gilbert, 1993).

There are also some third-world cities where rich households rent in preference to owning. In Cairo, Abt Associates (1982: 125) report that “ownership status, despite being preferred by households, is not positively related to income.” In India and Nigeria, where employers provide housing, the tenants are usually high-income professionals. In addition, certain more affluent Indians rent accommodation, preferring to invest their capital in expanding their business (Wadhva, 1992: 98).

Home owners tend to have lived longer in the city. This finding confirms early thinking on residential behaviour in Latin American cities which emphasised the relationship between tenure and migrant status (Turner, 1967; 1968). The bridgeheader/consolidator model, devised more than two decades ago, still has a certain validity. Migrants still follow the transition from renting to ownership. There is a pattern of movement from more central areas outwards towards the periphery. Poor households are able to build and consolidate homes, at least when they can create an investment surplus. But, now that cityward migration has been underway for so many years, more and more people have been born in the city. As the cities have become larger the housing alternatives become more varied and complex to describe (Brown and Conway, 1980; Gilbert and Varley, 1991; Gilbert and Ward, 1982; van Lindert, 1991).

As a result, interesting variations in this pattern appear. In several Mexican cities, for example, self-help settlements contain a much higher incidence of migrant owner-occupiers (Coulomb and Shez, 1992; Gilbert and Varley, 1991). By contrast, tenants, especially in the central city, are much more likely to be natives. It seems that native Mexicans use their family networks to improve their housing situation. Sharer families, for example, tend to be natives of the city or to have family in the city; non-sharers have no-one with whom they can share. Other natives use their contact systems and sometimes higher incomes to buy better quality plots or homes in the periphery. Given fewer alternatives migrants are often obliged to move into lower-quality owner-occupation.

In West Africa, further variations occur as a result of the tendency for many migrants living in the city to buy property “at home” in the countryside (O’Connor, 1983; Peil and Sada, 1984). In Nigeria, migrants are more likely to rent than urban natives, and partly for this reason, but also because migrants are poorer than natives, they are more represented in the rooming houses (Ozo, 1993).

Female-headed households are more likely to be tenants or sharers than owners. Research in Guadalajara, Mexico City, Querro, and Puebla has discovered that there is a higher incidence of female-headed households among tenants (Chant and Ward, 1987; Gilbert, 1993). A similar finding is true in West Africa (Barnes, 1982; Ozo, 1993; Peil and Sada, 1984).

There are several reasons for this tendency. First, women are often excluded from official housing programmes offering owner-occupation (Klak and Hey, 1992; Moser and Peake, 1987). Secondly, female-headed households tend to be poorer and, since poorer households frequently rent, women tend to be tenants. Thirdly, women tend to lack the kinds of skills required in self-help construction. As Moser (1987: 28) points out: “frequently they lack both skills and time to self-build but are often required to do so in the absence of funds for professional labour”. Unless they have grown-up sons they are less likely to opt for owner-occupation, although of course many wives are deserted by their husbands once the family have moved into a self-help settlement. As Fal Curutchet (1991: 37) describe experience in a Brazilian sites-and-services project: “women-headed households share similar aspirations to other beneficiaries of the programme but, due to their particular situation, they are among the least able to develop their homes.”

B. Preferences for owning, renting or sharing

In the United Kingdom, Saunders (1990: 2 and 311) argues that:

“mass ownership is associated with a strong popular desire to own personal property and is underpinned by deeply cherished and widespread values which emphasize independence, security and the importance of home as a base from which to venture out into the world... “Ownership not only guarantees certain rights which may be denied to tenants, but it also ensures permanency, even across generations. In a world where change is rapid and expectations are forever being turned upside down, the privately owned home seems to represent a secure anchor point where the nerves can be rested and the senses allowed to relax”.

Saunders is certainly correct in underlining people’s attachment to ownership but it should not be thought that this attitude is uninfluenced by any notion of monetary gain. For, over the years, British Governments of every political persuasion have offered people generous financial incentives when purchasing their home. Similar tactics have been employed in Australia where “the owner occupied sector has been heavily protected and taxed at a lower rate than other sectors in the economy” (Flood and Yates, 1989: 208). Fiscal incentives have also been used in the United States where, not surprisingly, a recent survey found that four out of five Americans thought that owning a home was a good investment (Megbolugbe and Linneman, 1993: 660).

Financial rewards have increasingly been offered by governments to formal sector owners in many Latin American countries. In Chile, for example, Governments since 1958 have offered families generous subsidies to buy their own homes (Kusnetzoff, 1990). As IEU (1990) point out “the treatment accorded by the state to owners and tenants is very different and highly regressive with respect to tenants”.

Ownership has also been attractive to low-income households, even if few have ever been offered any kind of financial incentive. Recent surveys conducted in several major Latin American cities, for example, have found a widespread preference for owner-occupation among both owners and tenants. In La Paz, Beijaard (1992: 67) finds ownership to be “by far the most preferred tenure situation for households of all tenure groups.” In Caracas, Mexico City and Santiago de Chile, the vast majority of existing home-owners express clear contentment with their current tenure, practically all of whom had had previous experience as tenants or sharers (Gilbert, 1993). Similarly, most tenants wish to be owners: 86 per cent of tenants interviewed in Santiago de Chile say that they would prefer to be owners and 94 per cent of those in Caracas. Only in Mexico City, is the preference less marked, a mere 58 per cent saying that they would prefer to own. Patterns of residential movement in each city also reflect this general preference for ownership. While most owners had once been tenants or sharers, few families now renting or sharing accommodation had previously been owners. The move into ownership is one-way traffic.

Similar attitudes are found in the three case-study countries. In Delhi, 72 per cent of a sample of 200 tenant respondents prefer owner-occupation, only 4 per cent preferring to rent (Wadhva, 1993: 52). In Egypt, Abt Associates (1982: 124-125) found that

“most households, whether owners or renters, would prefer to own rather than to rent. Among owners in Cairo and Beni Suef, 91 per cent and 100 per cent respectively expressed a preference for owning; among renters, Cairo and Beni Suef proportions who would prefer to own were 75 and 82 per cent respectively.”

In Nigeria, Ozo (1993: 38) reports that “the desire to own one’s house is widespread and cuts across all classes.”

A range of perceived advantages attracts low-income households to ownership. When owners in Latin America are asked about the advantages of owning a home, it is the sense of security and the feeling of independence that comes across (Green, 1988). Having something of one’s own gives an important boost to a family’s self-esteem. No doubt, such self-esteem is enhanced by the constant bombardment of television commercials, aimed at the middle class, about the availability of luxury homes on brand new housing estates. The image conveyed is that all “decent” people own their own home. Ownership in Latin America has become the cultural norm.

In Guayaquil, the most important emotion expressed by tenants about ownership is the need “to feel themselves masters of their own destiny, as symbolized by living in their own houses” (Salmen, 1989: 28). In part this means that they no longer have to share with kin or to put up with interfering landlords or fellow tenants. But, ownership also offers a major financial advantage, freedom from the need to pay rent every month. In Guayaquil,

“the future homeowners looked forward to having their monthly payments be an investment in their own property rather than a contribution to what they regarded as an alien landlord’s ill-deserved affluence” (Salmen, 1989: 28).

Home-ownership also provides something that can be left to the children.

Ownership also promises the chance of capital gain. Houses can be sold when the family move on. Rents are handed over to the landlord but payments for the purchase of land and housing are an investment for the future. In places this is no doubt a real advantage but all too frequently the advantages of ownership in terms of capital gain are vastly exaggerated. Capital accumulation is only possible if it is actually possible to sell. In practice, few owners in third world cities manage to sell their property. As Lloyd (1990: 293) puts it:

“the possibility of purchasing a house from the previous owner... in Africa is very weakly developed. Most of those who have already built a home intend that it shall pass through inheritance to their children - or if they cannot occupy it, to be rented so their children enjoy the income. The inadequacy and complications of legal titles are a further deterrent to transfer.”

Similar problems seem to be faced by owners in the suburbs of Santiago de Chile, to judge from the almost complete residential immobility of the owners in the low-income areas (Gilbert, 1993).

In Egypt, the main advantage of ownership for some appears to be associated with the possibility of living a different style of life. As Wikan (1990: 133) describes it

“the dream of a house of one’s own has been a woman’s dream... Confined to the back streets with the noise, turmoil and neighbourly interference, the desire for peace is one constantly expressed by poor women and is embodied in such dreams.”

Other less qualitatively based assessments in Cairo come up with preferences for “not having to pay rent” and “security” (Abt Associates, 1982). Hardly any prospective owners mention housing as a good “investment”.

Not only are there good reasons for positively opting for owner-occupation but there is also a push factor behind wishing to leave rental accommodation. Reporting on a survey in Oaxaca, in Mexico, Selby et al. (1990: 160) report that “renting is viewed as very unsatisfactory by urban Mexicans... Of all the bills that the poorer people of urban Mexico dislike, it is the most detested.” In the light of other surveys in Mexico this appears to be rather an exaggerated claim, but it is clearly true that in inflationary times ownership frees households from the worry that they will not be able to pay constantly rising rents. However, this is only possible where land is free or very cheap. Where land is expensive, monthly payments on land are a direct substitute for rental payments.

Despite this general preference for ownership, some households remain in rental or shared accommodation even though they have the resources to acquire their own self-help home. Some tenants, despite living in crowded conditions and who could have had much more space as home-owners on the periphery, cling to their rental accommodation. In Delhi, Wadhva (1993: 52) reports that 17 per cent of her tenant sample prefer to stay in the rented house for reasons of location, even though they own houses elsewhere in the city. In Nigerian cities, migrants often rent even though they buy property at home in the countryside (Ozo, 1993: 38). Even those who have lived many years in the city retain a strong allegiance to their rural “home”, the place to which they will retire and where eventually they will be buried. In certain parts of Latin America, substantial numbers of tenants also dissent from the chorus of approval for ownership. In Mexico City, two fifths of central tenants prefer to rent and, in Santiago de Chile, 30 per cent of those living in the conventillos say the same. Four fifths of these central tenants had never looked for their own home (Gilbert, 1993).

There are many reasons why some Latin Americans continue to rent. First, for tenants with secure tenure living in the central areas, a move into ownership is likely to mean a move away from the city centre. If they work in the centre, a move may lead to a journey of several hours per day, especially if they live in a megacity suffering from acute traffic congestion. Poor families in Santiago de Chile who were moved forcibly to the south of the city strongly resent the move. They tell of the many inconveniences caused by the move.

“The central area is a desirable place of residence because of the greater availability of services, its better access to sources of work and the presence of a local authority with a superior capacity to attend to their needs. They have a strong attachment to the centre, they value the contact with their neighbours and the security that the area offers them... In spite of counting with a home of their own, they manifest a clear desire to move; many would prefer to rent in the central area than to own a home in the south of the city” (Rodrez and Icaza, 1993: 5 and 72-73).

Secondly, a move to the self-help periphery may not be what tenants really want. While most families may want their own home, some families are in a particularly weak position to engage in the construction process. This becomes clear when we look at the characteristics of families who do become owner-occupiers. Families capable of undertaking the task of self-help construction are found in much greater numbers among the owners than among the tenants. Thus the proportion of workers in the construction industry is usually higher among the owner households and the number of female-headed households much lower (Gilbert and Varley, 1991). When tenants say that they want a home of their own, they may not mean that they are prepared to build their own house. As Coulomb and Shez (1991) put it: “the great majority aspire to owner-occupation but not under the conditions that are implicit in the ownership and self-help construction of a house in the periphery.” They want to own, but they mean ownership of a custom-built house not living in a shack.

Thirdly, renting may not be as bad as many reports portray. As chapter VI shows, many tenants get on with their landlords and are not constantly being threatened with eviction. For tenants with guaranteed tenure living near the centre of Cairo, Delhi or Mexico City, renting may offer just as much security as ownership. In Delhi and Cairo, tenancy rights can even be passed on to one’s children. Elsewhere, tenure is less secure but in cities like Caracas, Guadalajara and Santiago de Chile, few tenants change home frequently, indeed, in the central areas of Santiago and Mexico City, households stay in the same house for a very long time (Gilbert, 1993). If security of tenure is supposedly one of the main advantages of ownership, many families in Mexico and Santiago seem to achieve this goal perfectly satisfactorily through renting.

Fourthly, while most tenants may aspire to owner-occupation, they may not want that tenure right now. There are several groups for whom renting offers clear benefits. As Sundaram (1990: 126) points out:

“rental housing has certain inherent advantages from the individual’s point of view, such as low initial investment and greater flexibility for future tenure options. This makes it a more preferred alternative for more mobile younger households, the floating population and new migrants.”

In sum, the goal of home-ownership is much cherished. However, the form in which home-ownership becomes available is highly significant. Some families will accept home-ownership even when it means building a home on the unserviced periphery, while for others this is anathema. For some, ownership on the periphery is highly desirable, but it is an unobtainable dream. The result is that we have a highly diverse response to seemingly the same aspiration.

So far, the question of relative costs has not been raised. In some cities, rents are high relative to incomes, elsewhere they are cheap. In some cities, purchasing a plot is expensive, elsewhere land can be invaded or bought cheaply. A critical ingredient in tenure choice is the relative cost of the options. Renting a home in Mexico City is much cheaper than renting a home in Santiago de Chile or Caracas (Gilbert, 1993). Consequently, there are relatively few families in Santiago de Chile who can afford to rent accommodation whereas rent-income ratios in Mexico City are much lower. Under these circumstances fewer families in Mexico will be forced into peripheral ownership by their inability to pay rent. Families in Mexico City face a much more balanced choice between renting and owning than in Caracas or Santiago de Chile. Clearly when rents are very low, families will continue in rental accommodation even when the accommodation is inadequate. Should rents rise they may well reconsider their housing situation, either because they cannot afford the higher rent or because the balance of advantage between ownership and renting has shifted. It is this balance of advantage, not just in costs but in convenience, servicing and location, that seems to be critical in the process of residential choice. This balance, of course, is not determined by individual families but by the political economy of land and housing in the city and country concerned. The basic ingredient in understanding the transition to ownership in third-world cities, therefore, is to consider carefully how readily poor families can obtain land.

C. The choice between different kinds of non-ownership

The choice between owning and renting is not the only option facing households. As table 1 suggests, other kinds of option are available in some cities. In Delhi, Kumasi, Lima, and Santiago de Chile, sharing houses is very common, while in La Paz many households borrow homes from kin, and in Rabat many workers live in accommodation provided by the employer (Beijaard, 1992; Keles and Kano, 1987). Of these various options, sharing is by far the most common. In Delhi, for example, a recent survey by Wadhva (1993: 49) reveals that 37 per cent of households accommodate sharers.

Why do some households choose to rent and others to share? Unfortunately, rather little is known about the motivations of those who share. In some cities, of course, it is clear that many families are forced to share because of a lack of alternatives. In Santiago de Chile, where as many as one fifth of households were sharing in the 1980s, sharing is important because few families can afford to pay the high rents and cheap land is unavailable (Necochea, 1987; Gilbert, 1993). This lack of choice is reflected in the fact that sharers and allegados in Santiago de Chile are typically much poorer than tenants and owners. The allegado population seems to be the group most clearly forced into their current housing circumstances. In Delhi, too, the large numbers of sharers is explained in large part by the lack of alternatives. Households share accommodation because of the high level of rents and difficulties involved in building their own home (Wadhva, 1993: 49).

Yet not too much should be made of the issue of compulsion. For, if some sharers are forced into their present form of tenure, this does not mean that all are unhappy with it. Surveys in Mexico City, for example, suggest that many households who share accommodation are generally content to do so. While sharing is sometimes a tenure of last resort, chosen by those without the resources, 54 per cent of sharers say that there are no disadvantages to this form of tenure (Coulomb and Shez, 1991; Gilbert and Ward, 1985). They do not have to pay rent, they have access to the wide range of consumer durables belonging to their parents, and they have as much or more space than most tenants. Sharing also gives the household a better chance to save money. Their only real complaint is about their lack of independence. Perhaps as a result, sharing is not a short-term tenure, in Mexico City sharers average almost three years in the current home. Similarly, in Santiago de Chile, few who share wish to rent. Sharing is cheap whereas renting is expensive. Among the tenants there are some who would prefer to share were it possible.

If there are so many benefits why do more households not share accommodation? Clearly, in Mexico City some tenants do not wish to do so, 18 per cent saying that they value their independence and a similar proportion stating bluntly that they would not like to share. By far the most commonly cited reason for not sharing, however, is that it is simply not feasible; 28 per cent give this as their reason for not sharing and a further 20 per cent say that they do not have parents in the city (Coulomb and Shez, 1991).

Clearly, kinship ties are crucial in the decision to share, a factor underlined by the finding in Mexico City that every sharer is accommodated by a member of the family. Most sharers are the children (63 per cent), siblings (16 per cent) or parents (8 per cent) of the owners. Mexicans do not share accommodation with friends. The importance of kinship is even more clearly demonstrated in West Africa where sharing seems to have become almost as well-established in the cities as it is in the countryside (Korboe, 1992; Tipple and Willis, 1991; Peil and Sada, 1984). If relations own a home then the rest of family is entitled to stay with them. In West African cities, it seems as if “the very fact of even remote kinship has proved able to invoke free accommodation for the poor” (Korboe, 1992: 1159). In India, too, sharing is confined mainly to extended family groups (Wadhva, 1993). Grown-up children stay with their parents until they are quite old, particularly married sons who will inherit the house after their parents’ death. The next most common kind of sharer is newly arrived kin from outside the city.

Of course, whether all of the accommodating households are happy with sharing is less certain. Some undoubtedly welcome having their children “at home” but others no doubt regret the additional pressure on space. In West Africa, changing attitudes are beginning to increase tensions within extended families. Indeed, in Korboe’s (1992: 1162) survey of sharing families, 45 per cent of owners describe the extended family as parasitic. Much will depend no doubt on the circumstances of the hosts. If they have a large house or plot, they may be pleased to accommodate kin; if they have little room, their view may be more ambivalent. Clearly size of plot has some influence here (Chant and Ward, 1987). In Santafe Bogotwhere plots are relatively small, there certainly seems to be less sharing than in Mexico City where many plots are quite large (Gilbert and Ward, 1985). No doubt, too, the state of the house affects attitudes. In Kumasi, most sharers live in so-called “family homes” which “comprise some of the most neglected housing” (Korboe, 1992: 1168).

What is certain is that sharing can contribute substantially to easing, at least temporarily, a difficult housing situation. As Korboe (1992: 1169) puts it: the housing problem in urban Ghana would be

“decidedly more acute without the contribution which family-housing continues to make. Given the harsh economic climate, this form of housing is too important to be ignored by researchers and policy-makers.”

D. Who invests in rental housing?

The image of the large-scale landlord does not seem appropriate to the majority of landlords operating in most third-world cities. Of course, many cities contain some large landlords but there are remarkably few examples of landlords controlling a high proportion of the rental housing stock. In Africa, Nairobi seems to have many large landlords (Amis, 1984) but even there it is doubtful whether they dominate the market. As Lee-Smith (1990: 182) points out for two low-income areas of the city: “the majority of landlords were not part of the high-income group”. In the slums of Bangkok, Pornchokchai (1992: 152) notes the presence of some “houselords” who operate on a large-scale and in a business-like way. In Nigeria, too, “in city after city there are a few... wealthy businessmen and retired top civil servants, who own dozens of rental units” (Ozo, 1993: 33).

Generally, however, research is indicating that most third-world landlords operate on a small scale. In Latin America, the situation has changed since the nineteenth and early twentieth centuries when there were many large operators; most commercial landlords have long since departed from the scene. In Santafe BogotBucaramanga, Caracas, Guadalajara, Guatemala City, La Paz, Mexico City, Santa Cruz and Santiago de Chile, the typical landlord is now a former self-help builder (Beijaard, 1992; Coulomb, 1985; Edwards, 1982; Gilbert, 1983, 1993; Gilbert and Varley, 1991; Green, 1988; Rodas and Sugranyes, 1988; van Lindert, 1991). Many reside on the same property as the tenants and few have more than a couple of properties. In the consolidated periphery of Santiago de Chile, seven out of 10 landlords rent only to one tenant household, in Mexico City three quarters, and in Caracas two thirds. Even in the central areas of these cities, most landlords operate on a small scale. In Mexico City, subdivision of property through inheritance has gradually reduced the level of property concentration.

A similar pattern of small-scale ownership seems to hold for African cities such as Blantyre, Cairo, Lagos, Lilongwe and Lusaka (Barnes, 1987; Pennant, 1990; Rakodi, 1987; El Kadi, 1988). Indeed, generalizing across Africa, Lloyd (1990: 294) concludes that

“the rental market has been dominated by the small landlord; rarely, except perhaps in North Africa, have private investors built large block of tenements for the relatively poor.”

The same pattern also seems true for much of Asia, certainly in Bangalore the tenant: landlord ratio is only four to one (Malpezzi and Tiwari). In the self-help areas of Pakistan, most landlords operate on a small scale (Wahab, 1984). In Indonesia and Turkey, most landlords are former self-help consolidators (Hoffman et al., 1990; Turan, 1987). In Delhi, the tenant-landlord ratio is 2.5:1 and few landlords are “professional” operators (Wadhva, 1993: 45-6).

In Benin City, Ozo (1990: 264) claims that “the private rented sector is characterised by small-scale ownership.” Just 62.5 per cent of landlords own just a single property, only 12.5 per cent own two properties and a further 5 per cent own more than two buildings. The majority of the landlords/landladies are residents who occupy two or three rooms while letting the extra rooms to tenants. In a more recent survey, 86 per cent of 50 landlords interviewed own only one property, only one having more than three properties (Ozo, 1993a: 41). On average, these landlords provide accommodation for only 3.4 tenants. Few of these landlords appear to be operating in a professional way. Only two are building new rental premises, and only four are putting their rents into another business; the vast majority are using the rents as income on which they can live. Most of those with more than one property are pensioners. The typical landlord is self-employed and built the accommodation originally for his and his family’s own use.

In Cairo, most landlords are operating on a small scale; 91 per cent of landlords in Serageldin’s (1993) survey live in the building and 76 per cent of surveyed landlords have less than 10 tenants. Admittedly, some 9 per cent of landlords have 15 or more tenants but even the few large landlords earn much from their investment because of the rent controls. Certainly, there is little professional interest in rental housing and even less investment in this static business; indeed only 5 per cent of the landlords had begun renting during the last 10 years.

The characteristics of most landlords in third-world cities seem to be generally rather similar to those of other owners and sometimes even to those of their tenants. In Caracas, Mexico City and Santiago de Chile, while they are generally more affluent than other owners they have similar per capita incomes to those of their tenants. Clearly, landlords, owners and tenants in the consolidated settlements are drawn from the same social class. The main thing that tends to distinguish landlords from the rest of the population is their age. Landlords tend to be older than other owners and much older than most tenants. In Cairo, almost half of the landlords interviewed by Serageldin (1993) have been renting for more than thirty years. Because of their age, landlords are much more likely to be retired, live in larger properties than other families and have lived longer in their current home.