“A Patrimony for the Children”: Low-Income Homeownership and Housing (Im)Mobility in Latin American Cities

Abstract

Data are presented from a 2007 restudy of some 300 low-income self-builder owner households across eight settlements in Bogotá and Mexico City originally interviewed in the early and late 1970s, published in the mid-1980s (Gilbert and Ward 1985). Framed within a longitudinal perspective, the article analyzes the level of turnover of household owners living in irregular settlements over a period of thirty years; the current (2007) housing arrangements of households in dwellings and on plots; and the expectancies of ownership and inheritance of (now) adult children and grandchildren. The findings from the resurvey show minimal land-use changes and that more than 80 percent of the original families remain living on the lot. Densities have increased significantly, as has the average number of households sharing the lot. In Mexico City, sharing a lot is almost exclusively done with close kin (adult children), whereas in Bogotá it is both kin as well as renters. Self-estimated property values and tax office assessments show that house values in these consolidated settlements are often so high as to make it very difficult to sell, thereby reducing residential mobility. Also, the use value, and the inheritance expectations for second- and third-generation households living on the lots, gives little incentive (or option) to sell up and exit the settlement. Some of the social, judicial (tenure and inheritance), and housing policy implications and challenges are discussed.

Se presentan los datos de 2007 sobre un nuevo estudio de alrededor de 300 personas de bajos ingresos dueños de viviendas familiares auto-construidas en ocho asentamientos de Bogotá y Ciudad de México, originalmente entrevistados a principios y finales de los años 1970; estos datos se publicaron a mediados de los años 80 (Gilbert y Ward 1985). En el marco de una perspectiva longitudinal, el artículo analiza el nivel de recambio de los propietarios que viven en asentamientos irregulares, a lo largo de un período de treinta años; las características actuales (2007) de los hogares en viviendas y predios; y las expectativas de propiedad y herencia de quienes ahora son hijos adultos y nietos. Los hallazgos del nuevo estudio muestran cambios mínimos del uso del suelo y que más del 80 por ciento de las familias inicialmente entrevistadas siguen viviendo en el predio original. Las densidades se han incrementado significativamente, lo mismo que el número promedio de hogares que comparten un predio. Compartir un predio en Ciudad de México se hace casi exclusivamente con parientes muy cercanos (hijos adultos), en tanto que en Bogotá se comparte tanto con parientes como con arrendatarios. Los estimativos del valor de la propiedad por su propio dueño y los avalúos por la oficina de impuestos indican que los valores de las casas en estos asentamientos consolidados con frecuencia son lo suficientemente altos para dificultar su venta, lo cual reduce la movilidad residencial. También, el valor de uso y las expectativas de herencia de las familias de segunda y tercera generación que viven en los predios genera pocos incentivos (u opciones) para vender y salir del asentamiento. Se discuten algunas de las implicaciones políticas y retos sociales, judiciales (tenencia y herencia) y de vivienda.

Key Words:

• consolidated self-help housing
• innerburbs
• Latin America
• longitudinal studies
• mobility patterns

关键词:

• 综合自建房
• 老郊区
• 拉丁美洲
• 纵向研究
• 流动形态

Palabras clave:

• barrio de auto-construcción consolidado
• barriadas interiores
• América Latina
• estudios longitudinales
• patrones de movilidad

Acknowledgments

I am especially grateful to Dr. Edith Jiménez (project director of the Guadalajara study) and to the graduate students who assisted me in the data collection in Bogotá (Lissette Aliaga and Maria García) and in Mexico City (Erika Grajeda, Alejandra Ramírez Cuesta, and Cristina Saborio). The Latin Housing Network Database (http://www.lahn.utexas.org) and the support of the regional project directors is gratefully acknowledged. The fieldwork was conducted with a 2007–2008 grant from the LBJ School of Public Affairs Policy Research Institute.

Notes

1. Specifically: Argentina, Brazil, Chile, Colombia, the Dominican Republic, Guatemala, Mexico, Peru, and Uruguay.

2. That 1985 text was republished in paperback in 2008 as part of Cambridge University Press's digital publications series.

3. In 1999 Alan Gilbert published a study in which he returned to two of the five settlements in Bogotá that formed part of our PIHLU study in 1979, but he took a fresh random sample of lots rather than following up with the original interviewees, which was the primary aim of the 2007 study reported on here.

4. Katz, Berube, and Lang (2005) have recently begun to analyze the “first suburbs” that formed in the United States between 1950 and 1980. The research findings of the LAHN project described earlier are now beginning to offer parallel insights for Latin America, with the difference that it focuses on the first ring of irregular settlement formation between 1960 and 1980.

5. In the Puebla and Guadalajara study, Varley (1994) reported that around 15 percent of adult children exited the parental home and rented and a further 25 to 35 percent shared (either on the same lot or exiting and sharing elsewhere).

6. One also observes “for sale” signs in consolidated settlements, but this should not be taken to indicate a well-functioning housing market. Our data show that many homes remain on the market for many months or even years before selling (if ever), and others that do sell are invariably sold at bargain basement prices—one third or one half off the apparent market value (information gathered from author's 2009 fieldwork in Monterrey and Guadalajara).

7. A major 1978–1979 database of housing and household characteristics, Public Intervention, Housing and Land Use in Latin American Cities (PIHLU), reported in Gilbert and Ward (1985).

8. In the original study we retained the names and addresses, ensuring confidentiality (rather than anonymity) about our respondents. Today, of course, data records are subject to much greater scrutiny from human subjects’ review boards. If studies like this one are to be replicated in the future, however, then it is important that researchers adopt the confidentiality route rather than that of anonymity (which would require the expunging of all personal identification data). Had we not retained the original addresses, then this study would have been impossible. This underscores the responsibilities of lead researchers to maintain strict confidentiality in constructing their data sets (rather than anonymity) and to ensure that such confidentiality is preserved when offering public access to their data sets (as in the case of the data sets analyzed here; see http://www.lahn.utexas.org).

9. This is why some of the tables analyzed here only relate to data where there was a high level of confidence that the interview was being conducted on exactly the same dwelling site as thirty years earlier.

10. One of the anonymous reviewers expressed surprise that the final survey was so short given the care and time that was expended in tracking down the respondents in the first place. This is a fair criticism, but the primary aim of the study was specifically to measure the degree of turnover of ownership, as well as the nature of household change and structure compared with the earlier baseline data set. We also wanted to get some preliminary insights about property values, lot occupation by second- and third-generation family members, and inheritance patterns and expectations. It also served as a “pilot” survey to the later (2009) and more detailed, broader based surveys of innerburbs settlements in wider range of cities and countries, data that would ultimately allow us to analyze comparatively the current conditions, house structures, mobility patterns, assets, and inheritance, and so on. Furthermore, it was always intended that we would later conduct a small number of detailed qualitative “interesting case studies” with some of the families from the later 2009–2010 surveys. Indeed, in Mexico City the 2007 study led to additional qualitative work with some of the families we interviewed, resulting in a master's thesis by one of the graduate researchers (Grajeda 2008; Ward and Grajeda forthcoming) and an additional round of intensive case study interviews in the summer of 2011.

11. To comply with normal scientific study replication protocols, these data and associated coding guides can be accessed at http://www.lahn.utexas.org by clicking the 2007 Restudy Database link.

12. These results were from 2009 surveys across fifteen settlements (almost 1,200 cases) in Santiago, Chile; Buenos Aires, Argentina; Montevideo, Uruguay; and Guadalajara and Monterrey, Mexico (http://www.lahn.utexas.org). The higher end averages were common in Chile and in Mexico, whereas in Argentina and Uruguay, where settlements are not quite so old, the average number of years in residence on the lot is somewhat less (eighteen and twenty years, respectively).

13. It should be noted that the same degree of lot sharing among independent households is not as high in other cities in which we gathered data in 2009, where the average is commonly 1.4 families per lot (LAHN data). Similarly, the average total number of residents living on these lots is considerably less: an average of five persons in the four other cities not discussed here (Santiago, Montevideo, Guadalajara, and Monterrey). This suggests that the propensity to share on lots relates primarily to the operations of the land market and the low-cost housing opportunities that exist nearby—either for ownership or for rental. Land markets in both Bogotá and Mexico City are highly competitive and there is a scarcity of low-cost housing at affordable prices for new would-be self-helpers (Gilbert and Ward 1985). In Monterrey, Mexico, for example, the state sponsored FOMERREY low-cost land subdivisions generated a significant supply of new low-cost lots that provided a nearby alternative to remaining on the lot with their parents.

14. These “compound” arrangements—akin to those in tropical Africa, hence the term—are made up of several (close) kin-related families living in a single compound or residentially enclosed space (Lomnitz 1976).

15. If there is an inheritance challenge the effect will be to delay or prevent the process of title transfer to the intended beneficiaries, creating a new round of clouded titles to be “regularized” downstream (Grajeda 2008; Ward and Jiménez 2011). This new informality will further inhibit market performance, housing sales, and mobility.

16. In 2009 the following (trimmed) average values were recorded (from much larger samples): Guadalajara ($47,100) and Monterrey ($25,000), Mexico; Guatemala City ($39,936); Montevideo ($12,500), Uruguay; and Santiago ($27,100), Chile. In all cases the formal property (catastral) values, although somewhat lower, confirm that these estimated property values are realistic locally.

17. For further details on the methods see http://www.lahn.utexas.org (and click Methodology and Intensive Case Studies).