Latin American Research Network

Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

June 2007

I.

Justification and Background

Quality of life (QoL) is associated with the concept of social wellbeing and in mainstream economics has been mainly linked to consumption possibilities as measured by, for example, real GDP per capita. However, as surveyed in Biagi et al (2006), economic thinking in the area of development and welfare economics has pushed for a less simplistic vision of QoL. In this respect Sen (1987, 1993, 1997) recognizes income and consumption as components on QoL, but also emphasizes the “capability” and “opportunities” of individuals to achieve broad objectives which go beyond economic aims including political participation and social interaction.1 Based on these ideas the UN has proposed as an alternative indicator of human wellbeing the well known Human Development Index (IDH) (United Nations, 1990), which combines GDP per capita with an indicator of life expectancy at birth and educational attainment (illiteracy rates and gross enrollment rates). Inspired also by Sen’s ideas, Slottje (1991) has developed a multidimensional approach to the concept of QoL providing a methodology for measuring it and making cross-country comparisons. He proposes using 20 indicators that in addition to income per capita, life expectancy and educational attainment, includes political rights, freedom of the press and national infrastructure (roads). Quality of life indicators also have a long tradition in the field of Urban Economics as described in an early paper by Wingo (1973). Gillingham and Reece (1979) note that QoL is the result of the satisfaction the individual achieves as a consequence of the consumption of market goods, leisure, public goods and other characteristics (physical and social) of the environment in which the individual is located. Thus, one reason to focus the analysis of QoL at the urban level is that many of its key attributes, like crime and the environment, vary widely between cities and even across neighborhoods within cities. For example, in Latin America, some basic public services, determining key aspects of the QoL of households such as running water and sewage are generally supplied in the core central districts leaving the peripheral areas without coverage. QoL considerations enter into the decision-making process of households when they choose where to live and where to buy or rent a house. Thus, in the urban economic literature we find a strand of research that studies the role of QoL as an input to this decision-making process. Decisions about where to locate implicitly reflect preferences about a particular

1

See also Basu and Lopez Calva (2002).

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

location that offers a given level of QoL. These preferences will end up affecting the prices of land and house rentals. In other words, the value of QoL is capitalized in locally traded goods such as land and housing facilities. Thus, as we will describe below, a number of authors (see, for example, Roback (1982) and Blomquist et al (1988)) have provided hedonic price methods to estimate implicit prices of local QoL attributes, which then can be used to construct price-weighted QoL indexes that allow for across or within city comparisons. Thus, another reason to focus on the urban level is that local land and housing markets provide a method of assigning implicit prices to

many nonmarket attributes of

QoL.

II.

Objectives of the Study

The general purpose of this study is to provide updated estimates of QoL indicators for urban neighborhoods in a sample of Latin American cities. More specifically, there are two objectives. First, to provide a diagnosis regarding the behavior of these QoL indicators in order to address questions such as the following: Do we find important disparities in the QoL of urban neighborhoods both across and within cities? Are within-city disparities associated with a strong spatial pattern (low and high QoL neighborhoods are mainly concentrated in certain areas: central districts vs periphery?) What are the main driving forces pushing down QoL in different neighborhoods? Are the differences in QoL closely correlated with differences in income? The second objective of the study is to provide inputs for policy action. In this respect, at the very least, the information presented in the various studies should influence the public debate and place on the policy agenda issues associated with QoL in city neighborhoods. A more ambitious aim is for the analysis to help determine the priorities and actions of urban infrastructure and other intervention policies. To facilitate both objectives the studies will provide a methodology/model to perform a survey/monitoring system on QoL indicators to be implemented on a regular basis by municipalities.

III.

Conceptual Framework: Measuring QoL. 3

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

Despite the implicit consensus on the multidimensional nature of QoL, there is no clear definition of the term, and various indicators have been used in applied work to measure different aspects of QoL. What is implied by QoL varies according to data availability, the aims of each study, the methodology used and the spatial disaggregation level examined. Most studies include indicators describing weather, crime, pollution/environment, health, education and housing. There are some studies that provide a more detailed analysis of city amenities. For example, Blomquist et al (1988) in their study of 253 urban counties in the United States include distance of the city from coast/lake; Giannias (1998) in his study of 13 Canadian cities includes the number of professional sports teams in the city; Glaeser et al (2001) for 19 major Urban Agglomerations in the U.S. (plus London and Paris) includes restaurants, art museums and theaters per capita. Restaurants per capita are also used by Shapiro (2006) in his analysis of 222 U.S. metropolitan areas. Beyond the selection of QoL indicators, the difficult question is whether to combine them in a summary QoL index. A large branch of the literature simply combines in an atheoretic manner the different variables into single indices and then uses these indices to rank cities. There are several problems related with this approach. The main one is the inherent subjectivity in the choice, firstly of the indicators, and secondly of the final weights of every indicator used for the construction of the final composite QoL index. A different approach that has a strong theoretical background is to estimate weights based on preferences revealed in market decisions. Many QoL aspects do not have a market price (i.e. climate) and so a value needs to be assigned to them. For this strand of literature, this value is embodied in house prices and wages. In other words, the value of QoL is “capitalized” in the price of locally traded goods like land, housing and labor. Through what is called a hedonic price method, the implicit prices of local attributes can be calculated and then used either separately or as part of a QoL index, in order to rank cities. The tradition in the QoL literature using this approach began with the seminal works of Rosen (1979). His particular insight is that wages and housing cost should compensate for differences in other amenities among jurisdictions. Extending Rosen’s work, Roback (1982) constructed a model to show how interregional amenity differences are bid into interregional differences in both wages and land rents. These differences yield implicit amenity prices that

4

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

are appropriate weights for a QoL index. Following Rosen and Roback, many economists have used this methodology to rank cities, with variations in the functional form of the models they used and the indicators they included in measuring QoL. (Blomquist et al, 1988, Gyourko and Tracy, 1991, Stover and Leven, 1992, Giannias, 1998).

IV. Proposed Research Strategy and Methodology The proposed methodology for measuring QoL in urban neighborhoods in Latin American cities draws in part from the above urban economics tradition. A key aspect for determining the research strategy is that, in comparison with available academic and policy practice work, the aim is not only to obtain city level indicators but also to perform a more disaggregate, within-urban agglomeration analysis. The presentation of the methodology will be divided into three parts. First, we will define the proposed QoL indicators and discuss issues regarding the spatial disaggregation and statistical methods to interpret the data. Second, we will present alternative methodologies that can potentially be used to construct a QoL indexes. Finally, we propose to complement the quantitative, secondary-source construction of QoL indexes with the realization of direct surveys to obtain additional information on QoL variables and to gather more subjective opinions on QoL perceptions and satisfaction levels. 4.1 City Limits, Sub-City Disaggregation and QoL Indicators. Two important methodological decisions that have to be made are the boundaries of the urban agglomeration to be analyzed and the level of within-city disaggregation. Regarding the first issue, in some cities, the boundary of the city authority does not reflect the boundary of the whole urban area. In such cases, and depending on the availability of information, it could be convenient to adopt a metropolitan view where the urban agglomeration to be studied combines the central city and other surrounding localities that have a close association in terms of the relevant markets for locally produced goods (say, land, housing and labor).

5

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

In order to collect data showing variations within city, the appropriate sub-city areas should be identified. Clearly, the level of spatial disaggregation will be determined by the sources of

available information

and what

emphasis

the QoL methodology will place on

family/housing level variables vis-a-vis neighborhood or sub-city characteristics/amenities. Data gathered from population censuses supplies information on family and housing characteristics and access to public utilities by census radius. For example, the city of Rio de Janeiro has conducted various studies in which some socioeconomic indicators (education, size of household, water quality, sewage, etc.) were obtained for more than 8,000 sub-city areas (Prefeitura Da Cidade do Rio de Janeiro (2007)). On the other hand, other types of QoL indicators, indirectly associated with the supply of other public goods like crime, green spaces, infant mortality, environment quality, etc. are usually measured at a more aggregate level depending on how the provision of these services is organized across the city territory (say, crime statistics are collected by policy stations that in general cover an area higher than the census ratio; the same case may apply to

infant mortality, which

depends on the location of public hospitals). Thus, for example, in the case of the QoL reporting of the Urban Audit program of the European Union, a recommendation is made to divide cities into not less than 20 sub-city areas with a population average of 10,000 people but no one more than 100,000 or less than 2,000 (see European Commission (2000)). It is expected that the more indicators the QoL list includes, the lower the spatial disaggregation that can be reached. As a way of dealing with this issue, the following proposed set of QoL indicators are classified in terms of how geographically disaggregated the available information is. Thus, different types of QoL analyses can be implemented which will refer to different sub-city coverage. Table 1 defines a tentative grouping. There, variables are classified into two sets. Set 1 includes those obtained from censuses that are defined at the household level and are generally reported by census radius. Set 2 incorporates other more aggregate indicators referring to neighborhood/subcity/city level. Each case has different domains: those corresponding to Set 1 are housing infrastructure, access to basic services and education. Set 2’s domains are crime and personal safety, public heath, environment, formal educational facilities, public participation and social interaction, and cultural amenities. Each case provides examples of specific indicators.

6

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

Table 1. QoL indicators and grouping Set 1. Household level data (census)

Set 2: Neighborhoods/sub-city/city characteristics

1.1 Household characteristics

2.1 Crime and personal safety

Head of household with less than 4 years of school Numbers of children aged between 6 and 12 that are not at school Head of hosuehold with more than 15 years of schooling Size of the household Total household income

Total recorded crimes per 1,000 population per year Deads by car accidents per 1000 population per year

1.2 Housing infraestructure: Quality of the materials (roof, floor and walls) Nro of bedrooms Nro of bathrooms within the house Overcrowding 1.3 Access to basic services: Connection to water network within the house Sewage or appropiate sanitation facilities Electricity Natural Gas (connection to a Gas Network) Fixed telephone lines Apropiate waste disposal services Outdoor lightening (street lights) Number of public transport services within 300 mts of the house House located on a paved street House located in a shanty town ("Villas") Access to public transportation

2.2 Public Health Life expectancy at birth (years) Child mortality Number of hospital beds per 1000 inhabitants 2.3 Environmnet Days per year that NO2 concentrations reach 200mg/m³ Proportion of solid waste incinerated and/or recycled (%) Proportion of population exposed to outdoor noise level above 65 db Green space to which public has access (sq m per person) Presence of a informal waste disposal facility within the neigborhood 2.4 Educational Facilities Number of kindergarden educational facilities Primary educactional instritutions per 1000 inhabitants Secondary educactional instritutions per 1000 inhabitants Universities per 1000 inhabitants 2.5 Public participation and social interaction: Proportion of registered electorate voting in last municipal election (%) Participation in volunteering Number of Non Governmental Organizations (NGO) Number of local new papers 2.6 Cultural amenities: Local sports clubs or sport facilities (at city and subcity level) Local public or private libraries Teathers and Museums Number of restaurants in the city/neighborhood

The above classification allows for conducting the analysis of QoL indicators using a narrow Set 1-definition or a more comprehensive Set 1+ Set 2 measure. Of course, in the last case, producing sensible results demands consistency between the geographical boundaries defined by Set 1 and Set 2 variables. Once the relevant data has been collected, it must be interpreted so as to provide a diagnosis regarding the distribution of the QoL indicators. First, for each variable representative city level averages must be computed. Second, the goal is to illustrate the degree of disparity in QoL levels across city neighborhoods and also to investigate whether low and high QoL indexes are strongly correlated across space. To evaluate these issues the

7

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

corresponding distributions for the different variables have to be estimated. Territorial concentration of low or high indicators can be assessed by mapping the different variables and also by running correlation analyses. 4.2 Constructing Summary QoL indexes. Though information on the behavior of each QoL indicator separately will provide interesting results and constitute the key input for the proposed scheme of the QoL monitoring system, it may also be convenient to draw on the existing academic literature and policy practice to compute summary QoL indexes. As already mentioned, there are two approaches to this issue. The first approach is to build “a-theoretical” indicators where a-hoc grouping and weighting of different socioeconomic variables give rises to a QoL index. Once such index is the so called Human Development Index (HDI) which combines three indicators: (a) longevity as measured by life expectancy at birth; (b) educational attainment, measured as a weighted average of (i) adult literacy rate with a two-third weight, and (ii) combined primary and secondary gross enrollment rates with a one-third weight; (c) standard of living as measured by income per capita. An application of this methodology to an urban setting was done for Rio de Janeiro where census data was used to construct the HDI for 126 neighborhoods (Amorin and Blanco (2003)). Also, for Rio de Janeiro, Cavallieri et al (2007) presented the estimation of the so-called Social Development Index (SDI) covering 8,045 sub-city areas defined by census radii. The SDI is an equal-weighted average of 11 socioeconomic variables normalized between 0 and 1.2 For the case of Colombia, using the data provided by the National Survey on Quality of Life, Acosta et al (2003) construct a city level indicator based upon the methodology proposed by Cortes et al (1999) that includes sanitary and water services, garbage collection, schooling, overcrowding and house floors

2

The 11 indicators are: access to a water network within the house, access to sewage services, proper waste disposal collection, average size of household, number of bathrooms per house, % of illiteracy within household members older than 15 years, % of head of households with less than 4 years of school, % of head of households with 15 or more years of schooling; average income of head of household (in terms of minimum wages), percentage of head of household with income up to 2 minimum wages, percentage of head of household with income of 10 or more minimum wages.

8

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

and wall quality. They apply a principal component analysis to determine the weights for each indicator.3 A second alternative is to use land and housing markets to obtain implicit prices for the various components of the QoL index and then combine them using these prices as weights. It can be shown that this methodology is derived from microeconomic fundamentals describing the location decisions of families (and firms). Implicit prices for the various QoL attributes are obtained from a “spatial equilibrium” where a worker-resident, given equilibrium wages and prices for housing services, is indifferent to her city location. For this equilibrium to be sustainable, differences in urban amenities between alternative locations must be compensated for by differences in the price of local traded goods: housing prices and wages.4 Thus, the urban economic literature has usually assumed that city amenities affecting the QoL are capitalized not only in land or housing prices but also in wages. The key assumption here is that city borders could also place some limits on labor markets in the sense that the choice of residence also affects the access to job opportunities. An application of this conceptual framework for the case of an analysis that focuses on within-city variations in QoL would imply that most of the impact will be captured by housing prices and not as much in wages. This is because within-city location will not limit labor opportunities across the whole geography of the urban agglomeration if worker mobility is relatively high. To implement empirically this methodology, complementary data on real estate prices is needed. Ideally, for each sub-city area j, information on housing prices and characteristics need to be collected for a representative sample of housing

3

The index is calculated for nine regions of Colombia.

For a nice description of the microeconomic fundamentals behind hedonic pricing of quality of life indicators see Gyourko et al (1999).

4

9

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

units.5 Thus, the hedonic regression to be estimated would have the following form (Gyourko et al (1999)), Ln pij = constant + γ1 Hi + γ2 Zj + vij

,

vij=δj + ηi

(1)

Where pij is the rental price of house i localized in neighborhood j, Hi is a vector of individual house features (number of rooms, quality of construction, square meters, etc.), Zj is a vector of neighborhood j amenities (crime rate, green space, etc.), and vij is the composite error term which is a combination of a neighborhood-specific error component, δj, and a house-specific error component, ηi . The city specific error component is common to all houses in the neighborhood, and represents systematic uncontrolled differences in amenity characteristics across sub-city areas. But it also may capture systematic uncontrolled differences in housing quality across neighborhoods. Any of these two factors would imply that the composite error term across houses within the same sub-city area will be correlated, violating the OLS independence assumption.6 Estimation of equation (1) would produce estimates of the marginal price effects of both the k housing characteristics (the estimated γ11, γ12, …, γ1k) and the m neighborhood amenities (the estimated γ21, γ22, …, γ2m). Then, use these estimated coefficients to calculate the normalized price effect of each component, assuming a 1% change in the

There is a general feeling that the data on housing prices and rents is very scare and not of good quality in most Lastin American economies. On the one hand, cadastral valuations most of the time grossly underestimate the actual value of real estate property. Newspaper information for asking prices could be instead a good proxy for true prices but these prices tend to cluster in the formal and relatively wealthy areas of the cities. Thus, these neighborhoods tend to be overrepresented as compared to informal urban areas. The absence of reliable data on real estate prices could then preclude the possibility of applying this exercise in various Latin American cities. As mentioned below, this fact should be explicitly mentioned in the proposal as a justification to propose another approach to constructing QoL indexes. 5

In particular it will imply a downwards bias to the OLS-based standard errors (Moulton (1986)). Thus the potential problem of the presence of groups effects needs to be addressed by correcting the standard error by clustering or running a RE estimation (assuming city fixed effects are not correlated with any of the Z variables). Of course this problem will be minimized the better the data on individual housing characteristics and also the more data we have for neighborhood-level QoL attributes.

6

10

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

mean for each variable. Denote these normalized price effects by γ’11, γ´12, …, γ’1k and γ’21, γ’22, …, γ’2m . Then calculate the underlying quality of life index for neighborhood j as, QoLj = Σk γ’1k x Hkj + Σm γ’2m x Zmj

(2)

Where γ’1k and γ’2m are the normalized price effects of the k housing characteristics and m neighborhood amenities, respectively; Hkj is the coverage of the k housing characteristic in neighborhood j (i.e. % of residents without connection to a water network) and Zmj is the quantity of m amenity in neighborhood j (i.e. crime rate). Equation (2) introduces in the QoL estimate not only the impact of sub-city amenities, but also housing infrastructure features. Most of the empirical literature that uses this methodology ignores the housing component and concentrates on city/sub-city amenities. The reason for this is that household features are not inherent to a specific localization and also because, in most developed countries, there is not much variation in basic housing infrastructure across cities. This is not expected to happen in the case of a typical Latin American city so it could be useful to have both dimensions in the estimated QoL index. 4.3 Measuring QoL Satisfaction Using Direct Surveys. The methodology proposed in the previous subsection makes an effort to analyze QoL using objective, quantitative indicators as reported in census data and other sub-city/city level statistics. It is to be expected that some relevant dimensions of QoL would not be captured by this quantitative approach. First, it is important that the QoL indicator captures not only the actual supply or provision of city/neighborhood amenities (as it is the approach taken in Table 1) but also whether these amenities are actually enjoyed by people living in urban neighborhoods. So, for example, it could be interesting to know not only how many cultural or sport events the city organizes each year (and how these activities are distributed across the city geography) but also if people living in different city areas (both poor and rich) actually participate in those events. In other words, he aim is at unraveling whether beyond the

immediate

neighborhood

physical

characteristics

the

QoL

also

depends

on

amenities/externalities offered by the City and whether these events are more or less accessible to people living in different sub-city areas.

11

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

A second reason to develop direct QoL surveys is to ask for more subjective opinions about the level of satisfaction regarding the various indicators included in the QoL definition. There is a recent literature that has emphasized the utilization of more subjective satisfaction or happiness indicators in evaluating wellbeing in general and public sectors programs and policies in particular (Di Tella and McCulloch (2006), Kahneman and Krueger (2006)). This literature has shown that standard income or consumption variables are many times poorly related with subjective levels of QoL satisfaction. In this regard, Cattaneo et al (2007) provide evidence that certain basic housing characteristics generate significant improvement in health and self-reported levels of QoL satisfaction which are very poorly correlated with family income. In the same vein the results obtained from the Colombian Quality of Life Survey shows that though real income has declined between 1997 and 2003, access to some basic public services and the level of subjective satisfaction has increased during this period (Acosta et al (2005)).

Thus, it could be interesting to look at whether housing

features and other neighborhood/city amenities play any relevant role in self-reported or perceived QoL in the cities covered by the project. For the above reasons, we propose conducting direct surveys on QoL indicators. Ideally the survey should capture both meaningful city level averages and also variations across main sub-city areas. Still, we recognize that a survey with a sampling design that incorporates representative sub-samples of the main neighborhoods of a (large) urban agglomeration could exceed the budget proposed for this item in the project. The country research teams must make the decisions regarding the size of the survey and its sampling strategy. An alternative could be to apply the questionnaire in just two neighborhoods: one poor, lowQoL area and the other in a rich, high QoL zone (it could also be interesting to apply one to the “representative” average neighborhood). To select these sub-city territories it could be convenient to use the estimations performed in previous quantitative analysis to be sure that the two selected areas are representative sub-city regions (thus, they may not necessarily coincide with the extreme points of the QoL distribution). The surveys will include questions regarding QoL indicators such as those used in the quantitative analysis (census-type socioeconomic variables). It could also aim at obtaining information on housing prices and rents. It will look into a more detailed analysis of

12

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

neighborhood/city amenities emphasizing the actual access or consumption of these externalities by families living in different neighborhoods. Finally, as indicated, it could be interesting to include more subjective questions regarding the priorities people would assign to different indicators and their perceived impact on the quality of life and the overall degree of satisfaction. In part, these questions will be based upon questionnaires already applied by the Gallup Organization to measure Quality of Life at the national level. The IDB has reached an agreement with this institution to share this information so it could be interesting to use the urban QoL survey data to compute some variants of

QoL indices

(maybe more simplified ones) that could be compared with the nation-wide indicators obtained from the Gallup database. Appendix 1 shows a tentative questionnaire that can be taken as a starting point for a full survey to be applied in the country studies. Questions are classified into 3 types: type-A are those related to basic household and housing characteristics. Type-B are Gallup type questions associated with perceptions and subjective levels of satisfaction. Type-C questions are those that try to measure access to city and neighborhood amenities. Though we expect type A and B questions to be included in most country/city studies, we think there could be more variation in type-C questions reflecting the different particularities of the provision of city amenities in different countries.

V.

The Contents of the Country Studies. Each country study will be organized into 3 main parts. Part 1. The quantitative analysis of QoL indicators. In particular, it will cover the following issues: a. Definition of QoL indicators. Construction of relevant distributions for S-1 and S-2 variables across city neighborhoods Correlations across space. GIS-type maps describing the main results.

Measuring disparities in QoL indicators. Identification of

low QoL sub-city areas. Main determinants. b. Approaches to estimate QoL indexes: i) a-theoretical and ii) price weighted indexes. This last approach could be applied or not depending on the availability of reliable real estate price data. Results for intra-city variability in the QoL index. Mapping and identification of low QoL index sub-city areas.

13

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

Part 2. Analysis of the survey results. c. Selection of the representative neighborhoods and sampling. The survey-based level of QoL. Consistency with that obtained from census and city data. Disparity in QoL indicators between rich and poor neighborhoods. Its main determinants. d. Perceptions and degree of satisfaction about QoL dimensions in the rich and poor neighborhoods. Comparison with objective indicators.

Policy priorities as derived

from people’s perceptions vis-a-vis quantitative analysis. Part 3. Lesson from QoL estimation and a model for City-level QoL monitoring. e. Based on the analysis presented in Parts 1 and 2 and also considering current practices of QoL reporting,7 this part of the study will propose a model/system of QoL auditing. Here the aim is to develop a simple set of variables/questions that a local government can implement to monitor the evolution of QoL. There are already various institutions and organizations that report QoL indexes for a number of countries and cities. Nevertheless, many of these estimates do not report within-city variations and do not reflect local preferences, culture and tradition. This part of the study could be organized in the following way. f.

Lessons (from Parts 1 and 2) about measuring QoL in urban neighborhoods.

g. A proposed model to establish a system of QoL monitoring at the city level. •

The best indicators and questions: objective and subjective QoL variables.

•

Sources for the information and surveys.

•

Intergovernmental cooperation to gather QoL indicators.

•

Who should be in charge?

•

How to report the information.

•

Public participation

7

Interesting cases that can be looked at include: the City of London (London Sustainable Development Commission, 2005), Canadian Cities (Canadian Treasury Board, 2005), the Urban Audit Program of the European Union (European Communities, 2000), New Zealand Cities (http://www.bigcities.govt.nz) and the city of Porto (http://www.cm-porto.pt/ and Santos and Martins (2007)).

14

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

VI. Guidelines for the Research Proposals. To participate in the project, research institutions should submit a proposal detailing: The justification for the selection of a given city agglomeration including its boundaries and the level of within-city disaggregation. The proposed list of QoL indicators. It must include at least the following: indicators of housing infrastructure (quality) and basic services such as water, sewage, electricity, gas and fixed telephone lines. It must also contain variables related to neighborhood characteristics such as crime, the environment and access to local public goods (i.e. heath and education). The sources of the information for each QoL variable should be explicitly mentioned. A discussion regarding the availability of reliable data on real estate prices and whether this could be used or not for assessing QoL in urban neighborhoods. The type of statistical analysis to be used to perform the diagnosis regarding QoL analysis as described in section 3.

This must also include GIS-type (Geographical

Information System) methods to develop geo-referenced maps. A discussion of the methodology to be used to construct a QoL index by sub-city area. A

preliminary

questionnaire

to

be

applied

to

the

selected

representative

neighborhoods. In addition, the proposals must include, A list of the researchers that will be involved in the project. The center should present a research team justification of their capacity to meet the objectives of the project, including relevance of prior experience. Curricula vitae of the researchers may appear in a separate annex. Subsequent substitutions for researchers originally specified in the proposal may be made with prior approval from the IDB Network coordinator, but the project leader should lead the entire project until its completion. A budget (in a separate annex) indicating the time and resources that will be used within the context of the research work plan must be included. The budget proposed

15

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

by the research center should disaggregate items financed by the IDB contribution and those financed by the research center. The budget should distinguish among amounts assigned to professional honoraria, data collection (secondary sources), the cost of the QoL survey, “overhead” and other major categories of research expenditures. A brief description of the background (including previous work) of the firm or company to be hired for conducting the QoL survey. A proposal for the dissemination strategy of the final version of the paper and its policy implications. Eventually, some dissemination activities (non-academic and/or targeted to policymakers) could be supported by an amount up to 5 percent of the total cost of the project. Note: ALL proposals and research papers should be submitted in English

VII. Selection Criteria The research teams will be selected according to three main factors: Relevance. The research teams must spell out in detail the relevance of the cases chosen to the project’s objectives stated above. Data and Methodology. Data collection issues should be spelled out in detail (what sources of data will be used, what are the chances of success in obtaining such data, what channels are they planning to use to obtain data, the scope and quality of the QoL surveys to be implemented, etc.). Team Experience. The relevance of the team’s experience for the proposed project will be a very important criterion in the selection process. NOTE: Two institutions can apply with one proposal. The proposal should define the specific roles of the institutions and the distribution of financial resources between the institutions.

The IDB would make a contract with each institution.

16

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

Proposing research institutions should register as Research Network members (contact Raquel Gomez at [email protected]) and should be based in the Latin American and Caribbean region.

US and European institutions do not qualify as members of the

Research Network. However, researchers from the US and Europe can participate with research teams from proposing institutions.

VIII. Coordination and Tentative Schedule of Activities In addition to the individual country teams undertaking these studies, there will be a team of IDB coordinators (Eduardo Lora, Morgan Doyle and Andrew Powell) and one external academic coordinator (Pablo Sanguinetti, Professor, Department of Economics, Universidad Torcuato di Tella, Argentina). The tentative schedule of activities is as follows: June 15, 2007: Call for research proposals. July 13, 2007: Institutions interested in submitting proposals are invited to send an email stating their intention to participate in the bidding process to [email protected]. Institutions that do not send an email at this time may still submit proposals until the deadline. August 1, 2007: Due date for receiving proposals. Institutions should assure that the complete documentation is submitted to the evaluation committee. Complete documentation includes: “Registration form” with all information requested; the research proposal and CVs (up to three pages long max.) August 10, 2007: Announcement of selected research proposals. August 27, 2007: Due date for receiving a preliminary report with an annotated outline of the research paper, data sources and a very preliminary analysis of methodology and QoL indicators by sub-city level.

17

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

September 4-5, 2007: Kick off Seminar in Washington, DC with the directors of projects and the coordinating committee, or the purposes of presenting and discussing the preliminary report, data, a preliminary version of required statistics and the methodology to be used in the study. September 24, 2007: Due date for receiving the revised version of the annotated outline and methodology January 7, 2008: Due date for receiving a first draft of research papers. January 17-18, 2008: Discussion Seminar in Washington, DC, with project leaders and advisors for the purposes of discussing the first draft of the research paper. February 14, 2008: Due date for receiving a second draft of research papers. February 28-29, 2008: Videoconferences where the second draft of research papers will be discussed. April 1, 2008: Deadline for final version of the research papers, including a summary that discusses policy lessons. Deadline for presenting a list of the most relevant dissemination activities (e.g. events, seminars, workshops, etc.) to discuss the main policy lessons of the country study with local authorities, including the corresponding budget. Studies that are of good quality will be considered for publication as working papers. A selection of the best papers may be included in a special issue of an academic journal or in an edited volume on urban QoL in Latin America.

IX. Financial Aspects The IDB will provide US$40,000 as a contribution to the total budget of the study. Payments will be made according to the following schedule:

18

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

25 percent within 30 days of signing the formal agreement between the IDB and the respective research center. 10 percent within 30 days of presenting and approving the preliminary report with an annotated outline of the research paper, data sources and a very preliminary analysis of methodology and QoL indicators by sub-city level 30 percent within 30 days of presenting and approving the first draft of the research paper 20 percent within 30 days of presenting and approving the second draft of the research paper and upon delivery of the datasets utilized by the study to the IDB. 15 percent upon approval by the Bank of the final research paper.

Note: Fees for unjustified delays at any stage of the project will be charged in the form of a reduction in the corresponding payment of $200 per day of delay per subject. Failure to comply with any of the terms of the contract will imply a partial return of the funds paid up to the stage in which the failure occurs.

19

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

X. References.

Acosta, O., J. Guerra y D. Rivera (2005): “Acceso de los hogares a los principales servicios publicos y sociales y percepciones de calidad sobre estos servicios”. Serie Documentos nro 76, Economia, Universidad del Rosario. Amorin, E. and M. Blanco (2003): O Indice do Desenvolvimento Humano (IDH) na Cidade do Rio de Janeiro. Colecao Estudos Da Cidade, Prefeitura Da Cidade do Rio de Janeiro. Basu and Lopez Calva (2002): “Functionings and Capabilities” in K. Arrow, A. Sen and K. Suzumura (eds.), Handbook of Social Choice and Welfare. Elsevier, North Holland. Biagi, B., Lambiri, D and Royuela,V., 2006. "Quality of Life in the Economic and Urban Economic Literature," Working Paper CRENoS 200610, Centre for North South Economic Research, University of Cagliari and Sassari, Sardinia. Blomquist, G.C., Berger, M.C. and Hoehn, J.P. (1988) "New estimates of quality of life in urban areas", American Economic Review, 78, pp.89-107. Boyer, R. and Savageu, D. (1981) Places Rated Almanac: Your Guide to finding the best places to live in America, New York, Rand McNally. Cortes, D. Gamboa, LF. Gonzalez, J. (1999): “ICV: hacia un amedida de standard de vida”. Revista Coyuntura Social. Fedesarrollo, Noviembre. Di Tella, R. and R. MacCulloch (2006): Some Uses of Happiness Data in Economics. Journal of Economic Perspective, Winter, Number 1. Canadian Treasury Board (2005): Quality of Life. A Concept Paper Defining, Measuring and Reporting Quality of Life for Canadians. TBS Working Papers. Cattaneo, M, S. Galiani, P. Gertler, S. Martinez, R. Titiunik (2007): Hosing, Health and Happiness. World Bank Policy Research Working Paper Nro 4214. April. Cavalleri, F. A. Vial, G. Lopez and M. Abreu (2007): Diferenciais intra-urbanos no Rio de Janeiro: contribuicao ao cumplimento da Meta do Milenio. Prefeitura Da Cidade do Rio de Janeiro. European Commission (2000). The Urban Audit: Towards a Benchmarking of Quality of Life in 58 European Cities. ec.europa.eu/regional_policy/urban2/urban/audit/). Giannias, D. (1998) “A quality of life based ranking of Canadian cities”, Urban Studies, 35(12), pp. 2241-2251. Glaeser, E.L., Kolko, J., and Saiz, A. (2001) “Consumer city”, Journal of Economic Geography, 1(1), pp.27-51.

20

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

Gillingham, R. and Reece, W.S. (1979) “A new approach to quality of life measurement”, Urban Studies, 35(12), pp. 2241-2251. Gyourko, J. and M. Kahn and J. Tracy (1999): “Quality of Life and Environmental Comparisons,” Handbook of Regional and Urban Economics, Volume 3, edited by Paul Cheshire and Edwin Mills, 1413-1454. New York, North-Holland, 1999. Gyourko, J. and Tracy, J. (1991) “The structure of local public finance and the quality of life”, Journal of Political Economy, 99(4), pp. 774-806. Kahneman, D. and A. Krueger (2006): Developments in the Measurement of Subjective Well-Being. Journal of Economic Perspective, Winter, Number 1. Liu, B.C. (1976) Quality of life indicators in U.S. Metropolitan areas: a Statistical Analysis, N.Y. Praeger Publichers. Moulton, B. (1986): Diagnosis for Group Effects in regression Analysis. Journal of Business and Economic Statistics, 5: 275-282. London Sustainable Development Commission (2005): 2005 report on London’s Quality of Life indicators. May. Roback, J. (1982) “Wages, rents, and quality of life”, Journal of Political Economy, 90, pp. 1257-1278. Rogerson, R.J. Findlay, A., Morris, A. and Coombes, M.G. (1989) "Indicators of quality of life: some methodological issues", Environment and Planning A, 21, 1655-1666. Rosen, S. (1979) “Wage-based indexes of urban quality of life”, 74- 104, in P. Mieszkowsi and M. Stratzheim (eds.) Current issues in urban economics, John Hopkins Press, Baltimore. Santos, L.D. and Martins, I. (2007): “Monitoring urban quality of life: the Porto experience” Social Indicators Research, 80:411-425 Sen, A. (1987) The standard of living, University Press, Cambridge. Sen, A. (1993) Capability and well being, in M.C. Nussbaum and A. Sen (eds.), The quality of life, Wider studies in development economies, Clarendon Press, Oxford. Sen, A. (1997), On economic inequality, Expanded edition with substantial annexe by James Foster and Amartya Sen, Clarendon Press, Oxford. Shapiro, J.M (2006) “Smart Cities: Quality of Life, Productivity, and the Growth Effects of Human Capital.” Review of Economics and Statistics, 88(2), Forthcoming. Slottje, D. (1991) "Measuring the quality of life across countries", The Review of Economics and Statistics, 73, 4., pp. 684-693.

21

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

Stover, M.E. and Leven, C.L. (1992) "Methodological issues in the determination of the quality of life in urban areas", Urban Studies, 29(5), 737-754. Wingo, L. (1973) “The quality of life: toward a microeconomic definition”, Urban Studies, 10, pp. 3-18.

22

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

Appendix 1. Tentative Survey Questionnaire I. Household and Housing Characteristics 1. How many people live in this house/dwelling? (sleep in this house at least 6 months in a year) (type-A) 2. Basic characteristic of household members (type-A) Household member

Gender

Age

School Attendance (yes/No)

Maximum level achieved*

1. Head of household 2. 3. 4.

*1: primary incomplete, 2: primary complete, 3: secondary incomplete, 4: secondary complete, 5: university incomplete, 6: university complete.

3. What is the total household income per month? (type-A) 4. How many rooms does this house have? (type-A) 5. What is the material of the floors of the house?(type-A) a- Concrete b- Wood or tile c- Dirt or straw 6. How many square meters does this house have? (type-A) 7. How many bathrooms does this house have? (type-A) 8. The disposal of the toilet is connected to: (type-A) a. The sewage system b. A septic tank c. Hole or latrine d. The house does not have toilet (latrine outside the house) 9. Where does your drinking and cooking water come from? (type-A)

23

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

a. Public network (running water) b. Water well with electric pump c. Water well with manual pump d. They get it from a public faucet using buckets, bottles or cisterns e. River, canal, stream f. Bottled water or mineral water

10. Which of the following goods or services do you have in your home? (type-B Gallup)

Electricity connection (public network) Gas network/gas cylinder Refrigerator with freezer Refrigerator without freezer Washing machine Television Satellite / Cable T.V. VCR, DVD player Cellular phone Landline telephone Microwave oven Computer Internet Automobile Motorcycle Bicycle 11. Housing ownership (type-A) a. Do you own the house? Do you have an ownership certificate? b. Do you rent? How much do you pay per month? c. Legal occupation (with permission, lending or paying taxes) d. Illegal occupation (without permission)

12. Are you satisfied or dissatisfied with your current housing, dwelling, or place you live? (type-B Gallup)

24

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

II. Crime, personal safety and health care 13. How many times were you robbed at your house in the past 12 months? (type-C) 14. In this neighborhood (not in your house), how many armed robberies affected the members of this household over the past 12 months? (type-C) 15. Can you tell me which, if any, of the following issues has been a problem in your neighborhood over the past 12 months? (type-C) a. Vandalism, including broken windows in shops and public buildings b. Car theft, damage to cars or theft from cars c. Dangerous driving including drunk driving and speeding d. People who you feel unsafe around because of their behavior, attitude or appearance e. The presence of gangs f. Drug trafficking or drug sales 16. Do you feel safe walking alone at night in the neighborhood where you live? (type-B Gallup) 17. Do you feel satisfied with the police services in your neighborhood? (type-C) 18. Overall, do you feel safe and secure? Yes, No (type-C) 19. Would you say that in general your physical, mental, and emotional health is: (type-B Gallup) a. Excellent b. Very Good c. Good d. Fair e. Poor 20. Are there any health care facilities nearby? How much time does it take to get there? (type-C) 21. In the last 12 months did you need medical care for you or any of your children? (Yes/No) (type-C) 22. Did you receive the medical services you needed? (Yes/No) (type-C) 23. Medical services evaluation: (type-C) a. Did you find it difficult to reach the hospital or medical center? b. Did you have to wait long until you received the medical services? (yes/No) c. Did you find the proper physician or medical specialist? (Yes/No) d. Did you have to pay for the services? (Yes/No) How much? 24. In general, how would you rate the health care you currently receive? (type-B Gallup) a. Excellent b. Very Good c. Good d. Fair e. Poor

25

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

III.

Environment, transport and education facilities

25. What is the waste disposal system at home? (type-A) a. They are collected by waste disposal services (trucks) at the city level. b. You throw it to a nearby common open waste site. c. Burned or buried. d. You throw it to a river, public land or lake. 26. Does your household recycle any of the following? (Always(A)/Usually(U)/Sometimes(S)/Never(N)) (type-A) a. Glass (A/U/S/N) b. Newspapers / magazines / other paper (A/U/S/N) c. Cans (A/U/S/N) d. Cardboard (A/U/S/N) e. Clothes (Clothes banks or charity shops) (A/U/S/N) f. Plastic bottles (where possible) (A/U/S/N) g. Furniture (furniture re-use groups) (A/U/S/N) 27. How long does it take you to reach the nearby public park or square? (type-C) 28. Can you tell me which, if any, of the following issues has been a problem in your neighborhood over the last 12 months? (Type-C) a. Rubbish or litter lying in the streets b. Graffiti c. Air pollution d. Water pollution including floods, pollution in streams, rivers, lakes and in the sea. e. Noise pollution 29. How would you describe the present condition of your neighborhood’s green environment? (Type-C) a. Excellent b. Very Good c. Good d. Fair e. Poor 30. In the neighborhood where you live, are you satisfied or dissatisfied with? (Type-B Gallup)

a.

Thepublic transportation systems

1

2

Don’t know 3

b.

The beauty or physical setting

1

2

3

4

c.

The sidewalks or pedestrian ways

1

2

3

4

Satisfied

Dissatisfied

Refused 4

26

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

d.

The availability of public parks, squares and green areas

1

2

3

4

e.

The availability of public facilities where people can practice sports

1

2

3

4

31. From your home how much time does it take to get to work? (type-A) 32. In general, on a scale of one to five where one is very difficult and five is very easy, how easy or difficult is it for you to get to a public transportation facility such as a bus stop, train station in your neighborhood? (type-C) 33. Are there primary and secondary schools in the area where you live? (yes/no)(type-C) 34. Do your children attend these schools? (yes/no) 35. Are you satisfied with the quality of education your children are receiving? (type-C) 36. Overall, are you satisfied or dissatisfied with the neighborhood you live in? (Type-B Gallup)

IV.

Public Participation, Social interaction and diversity

37. In the past year, have you done any volunteer work in your neighborhood or community? (type-C) 38. Did you vote in the last municipal election? (type-C) 39. How much do you feel you can influence decisions affecting your local neighborhood? (type-C) a. A great deal b. A fair amount c. A little d. Not at all e. Don’t know

V.

Cultural amenities and leisure

40. How many times did you attend cultural events such as a movie, sport games, art festival, concerts, historical site, museums, art galleries, or theater performances in the past year? (type-C) 41. Overall how would you rate the arts and entertainment opportunities in: (type-C) i) Your neighborhood a. Excellent b. Very Good c. Good d. Fair e. Poor

27

THE LATIN AMERICAN AND CARIBBEAN RESEARCH NETWORK Call for Research Proposals

Quality of Life in Urban Neighborhoods in Latin America and The Caribbean

ii) Your city? a. Excellent b. Very Good c. Good d. Fair e. Poor

28