Urbanisation, Migration and Inequality in the 1960s and 1970s

In the mid-twentieth century, Guatemala had low levels of urbanisation compared with Argentina, one of the most urban countries of the time in Latin America, and Mexico, which had intermediate levels of urbanisation (see Table 1). Manufacturing industry growth was weaker in Guatemala than in many other countries during the ISI period in Latin America from the 1950s to the 1980s, based on tariff protection for domestic industry, but it shared with those countries other characteristic results of such policies, particularly economic concentration in the largest city.Footnote ⁶ The 1950s and 1960s saw high rates of urbanisation in Guatemala characterised by the increasing concentration of the urban population in Guatemala City. In 1950, Guatemala City was over four times larger than the combined populations of the next four cities; by 1981 it was over seven times larger.Footnote ⁷ Velásquez provides a detailed analysis of economic trends in these years, showing that economic and employment growth in Guatemala was primarily located in the city.Footnote ⁸ Primary sector exports prospered but the benefits from these were concentrated in large, often foreign-owned, agricultural enterprises and in the city-based financial and marketing firms that serviced them, and were diminished internally by unfavourable terms of trade. The share of GNP provided by small-scale subsistence-oriented farming declined sharply up to 1980, whereas the number of subsistence farmers increased as the rural population grew apace. Industrial growth, stimulated by the General Treaty for Central American Integration of 1960, concentrated in Guatemala City, as did investments in construction.

Table 1. Urban and Economic Indicators: Selected Latin American Countries

Source: United Nations Population Division, World Urbanization Prospects: The 2001 Revision (United Nations Common Database, 2002).

Economic concentration brought large-scale migration from rural areas and small towns to Guatemala City, as it did to the major cities of other Latin American countries. The implications of this migration and of ‘explosive’ urban growth became a major research focus for social scientists interested in development issues.Footnote ⁹ The question that preoccupied many researchers of the time, including myself, was whether rural people with low or no education, accustomed to the ways of life in very small places, could get by in large, heterogeneous cities. These issues of ‘peasants in cities’ were approached from diverse theoretical perspectives.Footnote ¹⁰ For some theorists, rural–urban migration was viewed as potentially positive when accompanied by urban planning for low-cost housing, not only in terms of providing the labour needed for industrialisation, but also in terms of instilling the values and skills required for economic development through education and the experience of ‘modern’ work.Footnote ¹¹ Other theorists pointed to the anomie that could result from migrants' experience of the impersonality of the city, and its possible negative political implications. This was a preoccupation even in the Argentina of the 1950s, where urbanisation was well advanced, as in Gino Germani's comments on the role of internal migrations in the creation of a society of masses in Buenos Aires.Footnote ¹² Whatever the perspective, rural–urban migration was seen as essentially a centripetal process in which over time rural and small-town inhabitants made permanent homes in the city.

My research in Guatemala City concentrated on two low-income neighbourhoods: a squatter settlement, San Lorenzo, and a low-income legally established neighbourhood, Planificada.Footnote ¹³ In both neighbourhoods, the majority of heads of family were migrants, but some 27 per cent in each neighbourhood were city-born; education and occupations among the latter were of a somewhat higher level than those of the migrants.Footnote ¹⁴ The focus of the research was, however, less on migrants than on showing how the poor, whether migrants or natives, were active agents in managing the challenges of living in Guatemala City. It was influenced by earlier anthropological research carried out in Central African towns, which showed how rural- or small-town-origin migrants could adapt to large towns and cities through social networks and urban identities based on the cultures of places of origin, kinship and religion, as well as urban work and residence.Footnote ¹⁵ The organising capacity of migrants in the cities and their capacity to adapt economically to urban life had also been shown in the Latin American cities of the 1960s, in such case studies as Mangin in Lima, Leeds in Rio de Janeiro, and Balán, Browning and Jelin in Monterrey, Mexico.Footnote ¹⁶

In the 1960s, there was relatively little migration to Guatemala City from the indigenous rural areas to the west. Indigenous identity was therefore not a significant organising factor in urban life, unlike in the case of La Paz in the same time period and, to a lesser extent, Mexico City.Footnote ¹⁷ In the 1980s, Bastos and Camus found a more substantial indigenous migration to Guatemala City, partly as a result of the violence against indigenous areas during the military regimes of the 1970s and 1980s.Footnote ¹⁸ Except in one of their neighbourhood cases, however, they report little evidence of indigenous families from the same place of origin settling in the same urban neighbourhoods and find relatively low levels of ethnic segregation.

In Guatemala, the relative weakness of rural–urban property and commercial ties meant that place of origin did not determine urban residence and relationships as much in Guatemala City as it did in Lima and Huancayo in Peru.Footnote ¹⁹ Kinship, however, was an important part of coping with urban life in Guatemala City, as elsewhere in Latin America.Footnote ²⁰ Also, my research and that of Lauren Bossen has shown that religion was an important source of material as well as emotional support for women in San Lorenzo, who were often the main breadwinners of the family due to the desertion or lack of contribution of the male head.Footnote ²¹

The Making of the City

The focus on adapting to urban life highlighted two key elements of the nature of poverty and inequality in the ISI period. Poverty was a general condition that in terms of low income and poor facilities constrained the lives of most of the urban population, but it did not prevent them actively seeking to improve their situation. Despite Oscar Lewis' culture of poverty thesis, I and others found no evidence that the urban poor were mired in fatalism.Footnote ²² Likewise, inequality, while clearly apparent in housing and lifestyles, was not perceived by my poor informants as a structure excluding them or their children from the possibility of obtaining education, employment and shelter. For rural migrants, the city offered opportunities for work, improved housing and education for their children. Among the first generation of migrants, poverty did not provoke violence or radical politics.Footnote ²³

A key to understanding why even the poor saw opportunities in their environment is the informality of many of the rapidly growing cities of Latin America. Poor migrants and city-born people could see themselves as ‘making’ the city because that is what they did, in terms of both work and housing. An important part of adapting to the city in this early period was the informal pattern of settlement. The Latin American city in the mid-twentieth century showed a chaotic pattern of urban development. Neither markets nor government planning ordered the city's ecology into clearly defined zones of commerce and industry, or of working-, middle- and upper-class residential areas. Land invasions, ambiguous or contested land titles, and the ease of subdividing land into building plots without providing infrastructure generated a heterogeneous pattern of residential and business settlement. Lúcio Kowarick labelled the process ‘the logic of disorder’, referring to the speculative, ad-hoc ways in which areas of the city were urbanised through the combined efforts of politicians, developers and people seeking cheap accommodation.Footnote ²⁴ The low incomes of the majority of the Latin American urban populations meant that there was little incentive for the private sector to build extensive middle-class suburbs on the US model, nor to provide housing for the working class.

Government-constructed housing for the working class was relatively uncommon until recently; the housing that was built tended to go to the middle class. Consequently, self-construction was the rule rather than the exception. Squatter settlements and other types of informal habitation occupied both peripheral and central locations. Low-income housing appeared even in zones mainly occupied by the wealthy. At times, this was facilitated by topography, as in the hillsides of Rio de Janeiro and Mexico City or the ravines of Guatemala City, but even in relatively flat cities, squatter settlements or semi-legal low-income housing found their way into the interstices of high-income neighbourhoods, such as the squatter settlement Villa La Cava, in the middle of the wealthy neighbourhood of San Isidro in Buenos Aires.

There was no one path of settlement in the city. The influential argument of John Turner posited a model of first seeking cheap rental accommodation in the central city, which made it easier to seek work, and then eventually moving to an irregular settlement on the urban periphery in order to self-construct housing to meet the needs of a growing family.Footnote ²⁵ Migrants also settled directly in more peripheral locations, however, as argued by Perlman for Rio de Janeiro and in Ward's analysis of patterns in Mexico City in the 1970s.Footnote ²⁶ Renting remained an important means of access to the city and to locating close to jobs.Footnote ²⁷ Most studies reported, though, that after some initial residential mobility, migrants became long-term residents of neighbourhoods, whether in legal or illegal settlements. Thus, even in neighbourhoods where tenure was precarious, most residents had lived there over the long term. In a 1993 survey of 61 precarious settlements of Guatemala City, Morán found that most households had lived in their settlement for ten years or more, and with longer residence, their housing was more likely to be built with solid materials.Footnote ²⁸ Obtaining housing, upgrading it with time and working with neighbours to improve neighbourhood amenities became in the 1970s and 1980s a major way in which even the poor could experience some improvement in their situation.

There was a clear synergy between informal settlement and finding jobs in the informal or non-state-regulated economy of self-employment and small businesses.Footnote ²⁹ In many cities, informal settlements were close to city centres or to the residences of the wealthy, facilitating service employment and street vending. Self-construction was itself a source of informal employment, as owners hired labourers and craftsmen to build their houses. Also, the flexibility of construction in informal settlements meant that housing could accommodate workshops, shops and other domestic industries.Footnote ³⁰ Even in formally settled neighbourhoods, zoning regulations restricting industry from locating inside residential locations were rarely enforced even when they existed. Lax government regulation stimulated many types of informal activity. The most successful entrepreneur in San Lorenzo in 1968 made his money by smuggling goods from Mexico to sell in the city, taking advantage of the tariffs that put a high price on legally imported goods.

In the Latin American ISI city, poverty and inequality were facts of life but not inflexible barriers to social mobility. Most researchers of the period did not view poverty as a social condition that required a government welfare policy. From the marginality perspective, the problem of the poor was their lack of integration into society and was to be remedied by education and community action aimed at promoting local organisation.Footnote ³¹ In cities such as Buenos Aires, where industrialisation had proceeded apace and formal work dominated, it was labour conditions rather than poverty that dominated the agenda. Gino Germani's classic Estructura social de la Argentina reviewed a host of social and political issues, such as social mobility, but had no section on poverty.Footnote ³²

Levels of occupational mobility were high in the cities of Latin America in the 1970s as educational levels increased and economies moved structurally away from unskilled manual work and toward skilled and white-collar employment.Footnote ³³ Even in Guatemala City there were high levels of structural mobility as white-collar service employment grew. Part of that growth was provided by state employment. Even in weakly developed states such as Guatemala, public employment was estimated in 1989 as comprising almost 17 per cent of the city's labour force and 27.5 per cent of the formal sector labour force.Footnote ³⁴ These figures are comparable to countries that had extensive bureaucracies, such as Argentina, where the public sector accounted for 52 per cent of non-manual workers, or Mexico, where the public sector comprised 17 per cent of total employment in 1980.Footnote ³⁵ In the 1970s the Guatemalan state provided some 4,000 houses mainly for its mid-level employees, offering them finished homes with utilities and neighbourhood parks and facilities.Footnote ³⁶

Change and Guatemala City 40 Years On

At first sight, the ending of ISI and the increasing integration of Latin America into the global economy mark a radical break for the cities of the region. Neoliberal economic policies opened markets (including land markets), privatised state-run enterprises, de-regulated labour markets and led to increased foreign direct investment (FDI). In the cities, the private sector had a greater freedom than in the past to provide housing, transport and communications infrastructure, and to develop large-scale commercial enterprises such as megastores and shopping malls. Alejandro Portes and I characterised the new urban environment in Latin America as ‘The Free Market City’, basing our interpretation on studies by colleagues in Buenos Aires, Lima, Mexico City, Montevideo, Rio de Janeiro and Santiago.Footnote ³⁷ We identified a set of social problems that accompanied the economic changes in the six cities: high levels of income inequality; job instability and informal employment, particularly marked in cities such as Buenos Aires and Montevideo, which previously had low levels of informality; a residential segregation marked by an increase in small-scale segregation through the proliferation of gated communities throughout the city; and increasing rates of crime and violence.Footnote ³⁸

The case of Guatemala City fits this profile, but in an exaggerated form as a result of the slow and unbalanced development of its economy and its tenuous insertion into the global economy. Up until 2000, the value added by agriculture in Guatemala to the country's GDP was greater than that added by industry. This is in sharp contrast to Argentina and Mexico, whose agricultural sectors contributed relatively small amounts to its GDP (see Table 2). Table 2 shows that in 2000, Guatemala was still primarily a rural country and had a rate of population increase that was higher than that of either Argentina or Mexico; it also shows that the GDP per capita was substantially lower than that of Argentina or Mexico.

Table 2. Selected Indicators of Development in Guatemala, Argentina and Mexico

Source: World Bank, World Development Indicators (WDI) database, http://data.worldbank.org/data-catalog/world-development-indicators, accessed 8 Sep. 2009.

Guatemala is less tied to global investments than most other Latin American countries, receiving between 40 and 80 times less FDI than either of the other two countries (see Table 2). FDI has grown rapidly in recent years in Latin America, increasingly concentrating in ‘urban’ investments such as communications, financial services (including real estate) and commerce. Only a minority of FDI is now placed in the traditional areas of agriculture and the manufacturing industry. In the first years of the new millennium, Guatemala has succeeded in increasing its value added through industry and also its FDI flow, which totalled US$ 658 million in 2007. Much of this has been due to investment in textile plants exporting to the US market; largely Korean-owned, these plants are located mainly to the south of Guatemala City and employ some 113,000 workers.Footnote ³⁹

Guatemala has high income inequality, with the urban Gini coefficient barely changing between 1989 (.558) and 2002 (.524). The percentage of the city's labour force working in the informal sector was approximately 57 per cent in 1964 and 49 per cent in 2004.Footnote ⁴⁰ Other estimations put the contemporary percentage of informal sector workers as much higher.Footnote ⁴¹ Neoliberal reforms in Guatemala may actually have energised the heterogeneous collection of activities that comprise the informal sector through the import of cheap goods and reduced government control of markets.

People that we interviewed in both neighbourhoods were positive about their economic situation, seeing crime rather than economic opportunities as limiting their possibilities. Self-employed workers abound in San Lorenzo today just as they did in 1968, with occupations such as tailoring, street sales, car repair and painting, and television repair work. The current most successful entrepreneur in San Lorenzo makes his money through distributing contraband goods as did his predecessor in 1968, through a chain of street sellers, renting space in his house to some of these. The contemporary contraband business is no longer based on avoiding tariffs. It is a product of neoliberal reforms that create incentives for the political and military elites to siphon off a proportion of private sector imports to be sold without sales taxes through intermediaries in the numerous markets that crowd the streets at the centre of Guatemala City. Apart from the vibrancy of the informal economy, an important factor that makes people in both neighbourhoods relatively positive about their economic situation is the fact that they have no golden age of the past to regret; this is in sharp contrast with the worsening situation of Buenos Aires, where the percentage of informal sector workers went from 12.9 per cent to 44.1 per cent and the Gini index went from .411 to .540 between 1980 and 2002.Footnote ⁴²

Rethinking Migration

Migration patterns in Guatemala show a greater break with the predominantly rural–urban migrations of the past and more similarities with other Latin American countries than might be expected from Guatemala's low level of urbanisation. Important explanatory factors are the ethnic divide in Guatemala and its troubled recent history. The earthquake of 1976 had severe negative impacts on the economies of the rural poor, as did the brutal civil wars of the 1970s and 1980s that resulted in the deaths of hundreds of thousands of Guatemala's population, mainly in the indigenous rural areas to the west of the country.

The resultant instability led to a rise in migration from the affected areas; the metropolitan area of Guatemala did not show any unusual rise in population during these years, however, with other, less urban regions showing more rapid growth in each of the intercensal years of 1964–73, 1973–81, 1981–94 and 1994–2002.Footnote ⁴³ The dense indigenous populations to the west of the country had not migrated substantially to the city in the 1960s, and despite an increase in the 1980s, their share of the city's population remains very low. They, like many other Guatemalans, have sought refuge elsewhere, mainly in the United States, but also in Mexico.

As was the case in the 1960s, Guatemala City is predominantly made up of owner-occupiers, with some 70 per cent of households owning the places where they live; these buildings range from the modest concrete-block houses of the poor to the mansions of the rich. The major zones of renting are Planificada, at the north-western edge of the city, where renters are just over half the population, and the more centrally located Zone 8, where 47 per cent of households are renting. In Zone 5, where San Lorenzo is located, 30 per cent of households are renters.Footnote ⁴⁴ In the heavily populated metropolitan municipalities, owner-occupiers are also the majority, as a consequence of the waves of squatter settlement and irregular subdivisions offering cheap housing to those living in overcrowded conditions in the central city. In 1964, some 61 per cent of the population of the municipality of Guatemala was born in the city, compared to 77 per cent in 2002.Footnote ⁴⁵ This latter proportion is similar to the 70.4 per cent reported for the central city of Buenos Aires in 1999, and the 77 per cent reported for the Federal District of Mexico in 1999.Footnote ⁴⁶ In contrast, the outlying metropolitan municipalities of Guatemala City show 52.6 per cent of their population born in those municipalities in 2002, but an additional 24.7 per cent of their population born in the central city. The central city is now predominantly a city of people who were born there. Of those that leave, over 80 per cent now live in the outlying metropolitan municipalities.

This pattern occurs elsewhere in Latin America, where many of the cores of the large cities began to report net losses of their populations in the early years of the twenty-first century, as in the case of Mexico City in 2000. As the metropolitan areas of cities grow, a considerable part of total residential moves are within metropolitan areas, particularly moves from the old cores of the cities, such as the Federal District of Mexico and the City of Buenos Aires, to their outer metropolitan rings. Emilio Duhau has analysed the population exchange between core and periphery in Mexico City, in which the poor are pushed out of the core through redevelopment and lack of space and the wealthy return to the core to occupy new condominium developments or gentrified central areas.Footnote ⁴⁷

Demographic changes contribute to this population dynamic. Fertility has declined everywhere in Latin America, so that rates of population growth are converging below 2 per cent. Average family sizes are now smaller than they were in the mid-twentieth century. Family structure is also changing, with extended families comprising a smaller percentage of households than in the past and with higher percentages of households made up by a single person, a single parent or couples without children. These changes in structure increase the number of households as couples break up and because the decline in extended family households means that children leave their parental household upon marriage or to live alone. These trends combine with previous high levels of fertility to drive intra-urban mobility.

The restudy of the two Guatemalan settlements first studied in 1968 illustrates these migration and demographic dynamics. There is considerable continuity in residence in both settlements, with the surveys showing that the average length of residence in San Lorenzo is 37 years; Planificada owners likewise have an average length of residence of 37 years. Renters in Planificada have been living for an average of 18 years in the neighbourhood and an average of eight years in the same rental. Although Planificada's large rental market could potentially attract recent migrants from outside the city, 60 per cent of renter household heads in Planificada have either lived all their lives in Planificada or come from another zone of the city. A further 31 per cent come from the outlying municipalities of Guatemala City's metropolitan area. The relative absence of rural migrants is highlighted by the fact that just 9 percent of renter heads of household came to Planificada from outside the metropolitan region of Guatemala City.

Despite the continuity in family residence revealed by the restudy, there has been considerable residential mobility among the families of the original informants. When the study was done 40 years previously, there was an average of three children per household in San Lorenzo. In the succeeding years, these children grew up, married and had children themselves. Space is at a premium in San Lorenzo. Although two-thirds of current respondents have added a second floor to their original house, there is not enough space to accommodate the increase in population. In Planificada, the lots were originally much larger and could have lent themselves to the type of remodelling to accommodate extended families that has been noted elsewhere in Latin America. However, renting space has proved a more attractive alternative to most lot owners, who stress in interviews how renting has become one of the most reliable sources of income in the city.

In both settlements, the families of the original residents have dispersed. In San Lorenzo, 19 per cent of the children or grandchildren of the original residents are still living in the same house, but 30 per cent have moved elsewhere in the same neighbourhood, 36 per cent have moved elsewhere in the city, 4 per cent have moved elsewhere in Guatemala and 11 per cent have moved to the United States. In Planificada, 22 per cent of the children or grandchildren of the original residents are living in the same house, 31 per cent are living in the same neighbourhood, 35 per cent are living elsewhere in the city, 1 per cent are living elsewhere in Guatemala and 12 per cent are living in the United States. Compared to 1968, international migration has become a much more evident feature of both neighbourhoods. Indeed, 32 per cent of household heads in Planificada and 59 per cent of household heads in San Lorenzo reported having siblings who are now living in the United States.

The family histories that we have collected suggest that these patterns result from the outcome of calculations that balance the need for space, the value of established community relations and the desire for new economic opportunities. Demographic pressures mean that a high proportion of people have to move out of overcrowded San Lorenzo, but they have the option of buying houses left vacant by those who have left the neighbourhood. One of the original informants that we re-interviewed was Lupe, the widow of Felix, a community leader and construction worker of indigenous background.Footnote ⁴⁸ She identified nearly 60 descendants, children, grandchildren and great-grandchildren. This family has remained tightly ensconced in San Lorenzo; its members help each other and see the neighbourhood as a supportive community. The children and grandchildren buy housing when it becomes vacant in San Lorenzo, so that they now own eight houses in the neighbourhood. Only one of the children has moved to a peripheral municipality. He is the only one of the children with a middle-class job, and his sons were the first in the family to complete college. He returns to visit every week despite now living far to the south in the city.

In two other cases, there has been more dispersion. The wife of one of my original informants, Don Luis, left him and moved to California, where two of his daughters also live.Footnote ⁴⁹ He has two other children living in the south of the city, where they moved because of the job opportunities created by the textile maquiladoras and the availability of cheap land for building. One of these two is a return migrant from the United States. A third son lives in the north of the city, and another, Luisito, has migrated twice to the United States but has returned each time to San Lorenzo because, as he puts it, he feels more comfortable there than anywhere else. Another case is Beto, who was born in San Lorenzo and was a child when I was there in 1968. Only one of his siblings remains in Guatemala, living in the north of the city. Beto's two adult sons have stayed in Guatemala, one living with him and the other in a neighbourhood in the same zone as San Lorenzo. His mother left for the United States in the 1980s with another of Beto's brothers to join his two eldest sisters in Los Angeles. His eldest sister was the pioneer migrant of the family in the early 1970s; her move to the United States was, according to Beto, probably occasioned by her conversion to evangelicalism and her church contacts there.

Leaving the city and a poor neighbourhood is viewed as both a survival and a social mobility strategy. Even Luisito admits in an interview that he would go again to the United States despite his liking for San Lorenzo and Guatemala. Remittances from the US are used to improve or buy housing. Two new formally constructed houses in San Lorenzo, one of three floors and the other of four, have been built with remittances. The three-storey house belongs to a return migrant who proudly told me that he had built it with 11 years' work in Los Angeles. The bottom floor is used as a garage where he rents spaces for neighbours to keep their cars; this is a business that he established speculatively on his return even though at that time there were no roads into San Lorenzo and hardly any cars. Those of our informants in Planificada who had lived in the neighbourhood from the beginning commented extensively on how the neighbourhood had become mainly a commercial and rental one. They cited examples of their neighbours who had gone to the United States and left their houses in charge of a relative, who had carried on a business there and rented out accommodation.

Inequality and the Consolidation of the City

In the contemporary period, marginality inadequately characterises the situation of the urban poor, who are now integrated into the city in terms of the provision of basic services, infrastructure and democracy, although their rising levels of education have not improved their job opportunities significantly. The term ‘exclusion’ better characterises contemporary inequality in the Latin American city, because it draws attention to the underlying processes that perpetuate disadvantage.Footnote ⁵⁰ Exclusion also draws attention to actions that keep the poor in poverty, particularly those of the wealthy and the markets. In this section I use the data from Guatemala to look at exclusion as a phenomenon of urban spatial organisation, including the physical barriers that contribute to a growing gap between the quality of public and private provision of education and other services, that stigmatise the poor and that concentrate crime and violence in low-income neighbourhoods.

Research in Latin America, as elsewhere, shows that the characteristics of neighbourhoods have an effect on the employment and educational opportunities of residents over and above the effects of the residents' individual and family characteristics. In homogeneously poor neighbourhoods, the poor are more likely to be unemployed or to fail in school than are the poor in more socially heterogeneous neighbourhoods.Footnote ⁵¹ Despite the concentration of poverty on the periphery, Latin American cities remain relatively less residentially segregated than their US counterparts.Footnote ⁵² In the contemporary city, however, residential segregation is acquiring a new significance in contributing to inequality.

The contemporary patterns of segregation are more malignant and exclusive than in the past because of three cumulative changes. Firstly, there are the consequences of the physical and population expansion of cities. This expansion results in larger areas of homogeneously poor neighbourhoods, and in substantially increased distances between peripheries and centres, which Iracheta and Smolka see as exceeding an economically tolerable level in terms of travel costs.Footnote ⁵³ Also, there is likely to be a worsening spatial mismatch between jobs and residence, as in the case of Mexico City, where the poor populations live at some distance from the major sources of employment as measured by the economic censuses.Footnote ⁵⁴ Secondly, the middle classes and elites are able to escape the poor quality of public facilities through private education, private healthcare and gated neighbourhoods. Thirdly, whereas in the ISI period even becoming literate brought rewards in an expanding labour market where non-manual jobs were an ever-increasing proportion of the total, it is not literacy that makes the difference in the contemporary labour market but acquiring professional and high-level technical skills. Studies in Monterrey, Mexico, show that compared to the 1960s, education makes relatively little difference to income except at the very top of the occupational structure.Footnote ⁵⁵

These changes are evident in Guatemala City, but in a more exaggerated form as a result of a highly unbalanced pattern of economic growth and low levels of infrastructure development. In some major cities, such as Santiago de Chile and Buenos Aires, the pattern of residential segregation has been changed by the provision of superhighways that circle the city and give rapid access to the peripheries. Their impact on segregation is to increase small-scale segregation by breaking up the large-scale segregation associated with the homogeneously poor peripheral areas of the cities through the creation of gated communities in peripheral areas.Footnote ⁵⁶ Guatemala City also now possesses a superhighway that circles the city and around which gated communities cluster. It is not complete, however, and access to the densely populated municipalities of the west and south of the city is limited by poor road links. The transport system in Guatemala has long been plagued with problems of old, overcrowded, unreliable buses. In 2007 a Transmetro system similar to that in Bogotá was introduced, but with only limited initial coverage.

Guatemala City's expansion has been a horizontal one, mainly extending some 20 miles towards the south and to the west, with little high-rise housing for low-, middle- or even high-income families in either the centre or the periphery. The city and its surrounding municipalities are more socially heterogeneous in terms of housing types and population than are most Latin American cities. Guatemala lacks the large-scale government housing projects and private investments that have built large homogenous tracts of middle- or low-income housing in cities such as Santiago.Footnote ⁵⁷ Villa Nueva and Pinula, the southern municipalities that have been popular destinations for people leaving San Lorenzo because of the offer of large, cheap lots, contain large areas of older low- and middle-income housing, the largest squatter settlement in Guatemala (with a population of around 55,000), and several gated communities. One of the latter that we visited is adjacent to the squatter settlement and has two sets of walls and entrances to protect its residents.

The heterogeneity of most municipalities in the metropolitan area and of the city's 22 zones means a proliferation of relatively small gated communities throughout the metropolitan area, some purpose-built and some where walls and gates have been built around existing neighbourhoods at the residents' initiative.Footnote ⁵⁸ Wherever you go in Guatemala City, there is physical evidence of a socially divided environment. One striking example is the recent enclosure by fences and gates of older low- to middle-income neighbourhoods across a major highway from Planificada. Planificada's boundary streets have also been blocked at some access points.

In various parts of the city, walls, guard posts and entrance roads have been built to monitor what are perceived as crime-infested neighbourhoods. Neighbours in El Gallito, a once socially heterogeneous neighbourhood in Zone 3 of the city, complained on the municipality website's ‘Barrio Querido’ section of the police closing 12 streets and leaving only three entrances and exits. The neighbours posted commentaries saying that this made life more difficult for residents and reinforced in the eyes of outsiders the neighbourhood's reputation as the home of criminals and drug traffickers.Footnote ⁵⁹ In a similar vein, the new through road in San Lorenzo has created controversy within the neighbourhood between those seeing it as a police attempt to control illegal activities within the neighbourhood and those favouring it for that reason. In 1968, a through road would, I suspect, have been seen by almost all neighbours as a major sign of progress. Gating in Guatemala, then, serves the dual purpose of protecting the inside from the outside and the outside from the inside.

A city of walls is also likely to intensify educational segregation in the city.Footnote ⁶⁰ At the primary level, public schools enrol some 54 per cent of Guatemala City's children, with the remainder attending private schools. The public schools serve small, relatively homogeneous areas such as San Lorenzo, whose primary school draws on the neighbourhood but loses children to what neighbours with more resources see as better private schools in the neighbourhoods above the ravine. At the secondary school level, segregation occurs mainly through families' varying ability to pay private school fees of at least US$ 40 a month. In Guatemala City, of the 40 per cent of children who attend secondary school, public schools educate 22 per cent and private schools educate 78 per cent.Footnote ⁶¹ Private secondary schools are concentrated in the centre of the city or close to the more middle-class zones of the city, entailing transport costs for those living in peripheral zones. One of our middle-class informants from Planificada arranges for her children to go to a city-centre private school, a journey that could take up to an hour.

These patterns of segregation are associated with high levels of crime and violence that make Guatemala City one of the most dangerous cities in Latin America. Insecurity is a general feature of urban life in Guatemala; it is a constant theme in all our current interviews in San Lorenzo and Planificada, and one that was almost absent in the 1960s. Various factors explain the high levels of crime and violence present in the city. Persistent poverty, unemployment and lack of good job opportunities, particularly for the young, are important explanations. There are at least 14,000 gang members in the city, predominantly aged between eight and 24.Footnote ⁶² They are concentrated in certain peripheral zones of the city and in outlying municipalities. These are a ‘second generation’ that experience the city as a place that frustrates rather than helps them realise their aspirations, unlike their migrant parents or grandparents, who found jobs and acquired housing and basic amenities.Footnote ⁶³ The insecurity and availability of firearms associated with Guatemala's long civil war and the expulsion of Guatemalan gang members from the United States in the 1990s are other factors in the formation of armed gangs (maras) throughout the city and metropolitan area. These gangs survive by robbery (including bus robberies), kidnapping and drug dealing. The victims are predominantly low-income populations who cannot afford the private security that protects middle- and high-income areas and businesses. Even gated communities cannot escape insecurity, however. The leader of the gang accused of the murder of the Guatemalan attorney Rodrigo Rosenberg lives in a gated community in the south of the city, with other gang members living close by.Footnote ⁶⁴ He is an ex-policeman, as is one of his associates; two members of the gang are current members of the police force, and another is an ex-soldier.

As has been reported for Rio de Janeiro, communities like San Lorenzo learn to live with drug dealing.Footnote ⁶⁵ In interviews, San Lorenzo residents recognised the presence of drug dealing and the violence occasionally associated with it but stressed that they still felt safer in the neighbourhood, where neighbours know each other, than on the outside, where they were prey to random crime and violence. Luisito, despite being robbed at gunpoint on a city bus shortly before one of our meetings, insisted that he felt safer in San Lorenzo than anywhere else because he was familiar with his neighbours and they with him.

Planificada informants worry about insecurity in the neighbourhood, which they see as having worsened over the years and whose visual sign is the prevalence of iron gratings covering shop fronts in most street businesses. Neighbourhood businesses pay protection money to gangs on a regular basis. The informants comment, however, that insecurity is bad throughout the city and that little is to be gained by leaving. This kind of calculation is also made by renters. One informant renting in Planificada has two daughters and one son, and the family is middle-class. The parents have been looking for housing with one of the daughters, who is engaged to a government employee. They want to buy a house, but say that the neighbourhoods where prices are affordable are too insecure and remote (‘bien adentro’); consequently, they have chosen to continue renting in Planificada until a suitable low-cost property comes on the market in the neighbourhood.

Changes in Civil Society

In our case studies, we have shown that the organisation and goals of urban popular movements vary dramatically between the six cities.Footnote ⁶⁶ There is, however, a common underlying matrix that is the changing class and spatial structure of Latin American societies during the neoliberal period; to explain the variation, we emphasise prior experiences of mobilisation and, importantly, differences between cities in the kinds of relationships that have developed between states, NGOs and citizens as a result of administrative and social policy reform. These new social policies decentralise administration, target low-income populations and directly and indirectly promote community and NGO participation. We argue that, depending on the capacity of the state, the reforms result in more pervasive relationships between governments and citizens, which encourage but also fragment grassroots responses to the challenges of the Free Market City.

The experiences of the two neighbourhoods (and particularly San Lorenzo) over 40 years provide an opportunity to explore these hypotheses with longitudinal data. The caveat here is even stronger than in the case of rethinking migration and inequality. Guatemala is, in many respects, a worst-case scenario of a failed state plagued by violence whose citizens have increasingly chosen the exit option in Hirschman's trilogy of exit, voice and loyalty.Footnote ⁶⁷

The consolidation of democracy in the face of military rule has been fragile. The Christian Democrat party and its affiliated movements had begun to establish a strong grassroots presence in the city when I was there in the late 1960s.Footnote ⁶⁸ The party failed to establish a strong political base that made it independent of the military and economic elites, however, and it disintegrated at the beginning of the new millennium. Urban and national politics are today dominated by a number of parties of recent formation that are closely identified with particular leaders. None of them has a strong or consistent electoral base. A telling indicator of the continuing weakness of governing institutions is the prevalence of private security forces guarding gated communities and neighbourhood businesses. These outnumber the national police force, with approximately 80,000 private security guards compared to 18,500 police officers.Footnote ⁶⁹

In other respects, Guatemala City shows the trends in state–civil society relations that are occurring elsewhere in Latin America. There are some 300 NGOs located in Guatemala City, both national and international, many of which provide social and economic services to the urban low-income population. In 1968 the NGO presence in the city was very limited, apart from religious organisations, both Protestant and Catholic. These latter organisations have also grown in number, and several of them have missions in San Lorenzo, where students from North America are sent to distribute food and provide literacy or other training programs for neighbourhood inhabitants. San Lorenzo has a kindergarten and primary school set up by an Indian-origin educational movement for peace and harmony. Recently a computer training facility has been established with NGO funds in the communal centre of the neighbourhood, with over 20 computers. In general, projects to improve irregular settlements in the city are mainly financed with help from NGOs, international agencies or foreign governments, with Guatemalan government or municipal agencies providing technical assistance.

Guatemalan government has modernised considerably in recent years, providing internet access to its ministries and following international standards in applying surveys and censuses. It has also collaborated with the World Bank and a number of other international organisations (the Inter-American Development Bank, the United Nations Development Programme, the United States Agency for International Development, the United Nations Children's Fund, the Soros Foundation and the International Labour Organization) to develop a state-of-the-art national poverty plan, the Guatemala Poverty Assessment Program (GUAPA).Footnote ⁷⁰ The Secretaría de Planificación y Programación (Ministry of Planning, SEGEPLAN) promotes the new social policies of targeting, decentralisation, multiculturalism and participation that have become standard in Latin America.Footnote ⁷¹

Although Guatemala City lacks an effective metropolitan authority, the municipality has an active policy of community development and outreach. Its website promotes local community identity through the ‘Barrio Querido’ section, which features the histories of low-income urban neighbourhoods including Planificada and a neighbourhood adjoining San Lorenzo, records the memories and opinions of residents, and reports on current activities and organisation. The municipal government has organised the municipality into 14 districts, 54 delegations and 700 neighbourhood committees, and these have generated a further 300 security committees responsible for patrolling the streets. Whereas in 1968 there was little formal municipal presence in San Lorenzo, there is now an official neighbourhood committee of nine members and a municipal representative, all of whom are active in promoting the neighbourhood's welfare.

Our interviews with neighbours and leaders in San Lorenzo are still under way, and provide only preliminary insights into the changes in community organisation. Projects of neighbourhood improvement continued after those described in 1968, and San Lorenzo has had a relatively continuous history of neighbourhood committees and participation in neighbourhood projects, of which the most important are a neighbourhood primary school, a municipal clinic, provision of piped water to houses, providing concrete surfaces for the roads and alleys of the neighbourhood, and building a through road that links San Lorenzo to the centre of the city. Women have been as active as men in leadership positions, as Bossen makes clear.Footnote ⁷² One example is María, a long-term resident of San Lorenzo; a somewhat volatile member of community organisations in 1968, she later became president of the Betterment Committee and an effective community leader who helped obtain the school.Footnote ⁷³ She, like other former leaders, retains clear memories in our current interviews of the value of collective organisation in the neighbourhood, detailing the various projects and giving dates for the arrival of water, electricity, drainage, schools and clinic.

These leaders are also aware of the growing formality of a neighbourhood administration that began with ad-hoc committees to promote projects; these became officially recognised Betterment Committees, which then turned into the current Comités Unicos de Barrio (Neighbourhood Committees), with municipal officials organising the elections and distributing leaflets calling for meetings. Residents finally received official titles for their houses in 1998, and are an integral part of the city's administration and utility services; water and electricity meter readers now walk the alleys, a possibility that would have been unthinkable in 1968. Our neighbourhood interviewees calculate that they now spend some US$ 35 a month on utility payments, costs that were not present in 1968. They talk about the growth in drug-related violence, but emphasise the considerable improvements that have been made in the neighbourhood.

Contemporary leaders stress the value of collective action as much as they or their counterparts did in 1968, though they note that neighbours rarely join together in projects and the relationship with the municipality or NGOs is an individual one. Also, they claim that participation is not as extensive as it has been in the past – the mandated yearly meeting of the Asamblea de Barrio (Neighbourhood Assembly) rarely attracts more than 50 attendees. Collective action today, as in 1968, is not dominated by political parties. Leaders made clear in our interviews that the neighbourhood should seek to obtain what it needs by approaching every political party with candidates in a municipal or national election and seeing what they are prepared to offer. Luis had been active in neighbourhood committees in 1968, and though he worked for various political parties, he had been mainly sympathetic to those on the Left.Footnote ⁷⁴ In 2009 he had retained his political sympathies, but was as clear as he was in 1968 that he is primarily a neighbourhood activist, ready to approach any political party for help.

Don Beto holds a municipal position that gives him responsibility for advancing infrastructure and social development in San Lorenzo. He has been active in promoting various projects, such as the through road and the computer centre. He even created a sophisticated PowerPoint presentation for meetings with potential donors in which he juxtaposed images of San Lorenzo from 1976 with pictures taken from the same position in 2008 to show the improvements in the neighbourhood. Beto's 28-year-old son, who lives with Beto, helped put together the presentation. He has finished secondary school and works as a gas meter reader. From the time that he was in secondary school, he has been active in city-wide projects sponsored by the Kansas-based organisation Children International. He likes San Lorenzo and sees it as having community spirit, dismissing its insecurity – ‘hay mas peligro afuera que adentro’. He is not interested in migrating to the US, where other members of his family live, and sees a future for himself in community organising.

Levels of awareness of political and international issues are much greater now than in the 1960s. Our informants talk knowledgeably about city, national and international events. Mobile phone and internet use is common in both San Lorenzo and Planificada, as is access to television. In other respects, however, the contrast with 1968 is far less sharp. Beto is a better informed, more self-assured and organised leader than his counterparts in 1968, but sees himself as working very much in their tradition, constantly invoking the achievements of the previous Betterment Committees. Current local leaders in San Lorenzo need to be adept at working with the city administration and national and international NGOs, but that was also the case in 1968 when municipal de-centralisation had begun. What is less evident now is involvement in city-wide coalitions to improve conditions in poor neighbourhoods. These had been present in the late 1960s and 1970s, in the form of city-wide organisations such as the Movimiento Nacional de Pobladores (National Movement of Settlers, MONAP) and the Instituto para el Desarrollo Económico y Social de América Central (Institute for Socio-economic Development in Central America, IDESAC), and the network of Women's Clubs.Footnote ⁷⁵ These and other organisations that promoted city-wide movements of community development began to withdraw from San Lorenzo in the 1980s as political tensions and government repression grew. The last of those that I had known in 1968, IDESAC, withdrew in the late 1990s to concentrate on rural issues.

Conclusion

Re-studying migration and inequality through the lens of Guatemala reminds us that change is often more incremental than we might assume, and that making sharp contrasts between economic periods can be misleading. The way in which poor people adapt to urban conditions in 2009 is basically similar to the way in which they adapted in 1966 and 1968. Their reliance on mutual aid among family and neighbours is similar, as is the way in which they actively and eclectically seek the help of outside government agencies or national and international NGOs. Nowadays the poor in Guatemala are faced with a larger and more complex city, more insecurity and more difficult housing choices than they did in the 1960s. Unlike other cities in the region, however, Guatemala City has never been an industrial city with high levels of labour organisation and social protection for the working population. Our interviewees regret the passing of neighbourhoods secure from crime and violence, but they also value local community organisation, the gradual improvements in housing and the new economic opportunities made possible by international migration and a flourishing neoliberal informal economy. They have been badly let down and exploited by political and economic elites, but they never expected much of them in any event, other than individual favours and patronage. They continue to exemplify the concept of popular rational adaptation in which the poor seek means to cope instrumentally with the specific social and political context of their city in order to ensure survival, physical security and neighbourhood improvements.Footnote ⁷⁶

Despite the particularity of Guatemala City, the research issues seem clear. Migration is no longer the end point of the rural–urban transformation that results in finding, after a few trials, a permanent urban location. Instead, it has become part of the constant re-ordering of settlement in the large cities, particularly the metropolitan areas, as people, individually rather than collectively, seek accommodation that best meets their budgets and family stage and structure. It is also a starting point for city dwellers seeking to capture resources elsewhere in their country or abroad. The extent and consequences of these movements, including return to place of origin and the ties that are created, require more research than they have yet received.

If we link urban inequality to social exclusion, we need to examine empirically the mechanisms by which this occurs. To what extent are there diminishing opportunities for social mobility, and how central have high levels of education become for obtaining the relatively few good jobs available in a city? To what extent are the poor excluded because they are locked into neighbourhoods where there are few local job opportunities and where schools are of low quality? Where once welfare was closely tied to work and the formal economy, it is now increasingly based on an individual household relation to the market and to the state. In cities like Guatemala, informality may spread wealth to the population through illegal as well as legal means more effectively than does the formal sector. Corporate, employer-based provision of welfare, negotiated by trade unions, slowly disappears. Inequality in this situation does not mean lack of opportunities for the poor, but lack of the means to assure a sustained improvement in their and their children's social and economic situation; neither education nor the labour markets provide that assurance. Only the highest levels of education guarantee a stable, well-paid job. The labour market does not easily lend itself to building a career in one firm or field of expertise because of the volatility of an increasingly globalised economy in which foreign investment comes and goes with changes in international trading agreements, as has happened in the Guatemalan textile industry. Spatial mobility becomes a means of securing opportunities, but equally, as we have seen, immobility is seen as a less risky strategy by many.

Rethinking inequality in this context entails separating the formal analysis of inequality from its substantive meaning. Formally, the urban social structure has become increasingly rigid as social mobility is reduced by an increasingly physical pattern of spatial segregation, deep divides in the quality of private and public provision in health and education, and the scarcity of managerial and professional positions. Substantively, inequality is obscured in Guatemala as in other Latin American cities by an improvement in living standards and the range of consumer goods accessible to the poor, by an insecurity that is shared by rich and poor alike, and by a likely decline in encounters between poor and rich, whether in the workplace, in public spaces or in markets or malls. The final aspects of contemporary inequality that need research are the ways in which inequality and poverty may be acquiring both new form and new meaning as they become the objects of social welfare policies that create new direct relationships between governments, NGOs and their client populations. As Simmel argued, when poverty is a relationship between a state and part of its population, then it becomes a relationship of second-class citizenship.Footnote ⁷⁷ The poor are targets of state policy, and their receipt of aid comes in return for conforming to certain patterns of behaviour.