Paradox and Perception: Measuring Quality of Life in Latin America
274
Paradox and Perception: Measuring Quality of Life in Latin America
274Paperback
-
PICK UP IN STORECheck Availability at Nearby Stores
Available within 2 business hours
Related collections and offers
Overview
Paradox and Perception greatly improves our understanding of the determinants of well-being in Latin America based on a broad "quality of life" concept that challenges some standard assumptions in economics, including those about the relationship between happiness and income.
The authors' analysis builds upon a number of new approaches in economics, particularly those related to the study of happiness and finds a number of paradoxes as the region's respondents evaluate their well-being. These include the paradox of unhappy growth at the macroeconomic level, happy peasants and frustrated achievers at the microlevel, and surprisingly high levels of satisfaction with public services among the region's poorest. They also have important substantive links with several of the region's realities, such as high levels of income inequality, volatile macroeconomic performance, and low expectations of public institutions and faith in the capacity of the state to deliver. Identifying these perceptions, paradoxes, and their causes will contribute to the crafting of better public policies, as well as to our understanding of why "populist" politics still pervade in much of the region.
"
Product Details
| ISBN-13: | 9780815703266 |
|---|---|
| Publisher: | Rowman & Littlefield Publishers, Inc. |
| Publication date: | 09/29/2009 |
| Pages: | 274 |
| Product dimensions: | 6.10(w) x 9.10(h) x 0.80(d) |
About the Author
Read an Excerpt
PARADOX AND PERCEPTION
Measuring Quality of Life in Latin AmericaBROOKINGS INSTITUTION PRESS
Copyright © 2009 THE BROOKINGS INSTITUTIONAll right reserved.
ISBN: 978-0-8157-0326-6
Chapter One
How Latin Americans Assess Their Quality of Life: Insights and Puzzles from Novel Metrics of Well-BeingCAROL GRAHAM AND JERE R. BEHRMAN
This book is an attempt to better our understanding of the determinants of welfare in Latin America and the Caribbean, a region that is diverse in terms of culture and levels of development. Some countries in the region are approaching developed country standards of living, while others approximate the per capita income levels of sub-Saharan Africa. Further, as is well-known, the region has relatively high intra-country variations in per capita income—indeed with the highest inequalities of any of the major world regions. These per capita income differences, in addition to the region's cultural diversity, provide a laboratory for studying how quality of life varies with a number of important objective and subjective measures, including per capita income but also including others such as job insecurity, job satisfaction, schooling attainment, educational quality, nutritional insecurity, personal insecurity, mortality, and self-assessed health.
The concept of quality of life is a broad one, which incorporates basic needs but extends beyond them to include capabilities, as typically measured by the United Nations Development Program's human development index (UNDP's HDI); the "livability" of the environment, as measured by income per capita and growth; and life appreciation and happiness, as measured by well-being surveys. In this book we focus on reported or subjective well-being ("happiness" or "satisfaction"), a concept that differs from but complements other indicators of the quality of life. Our analysis builds upon a number of new approaches in economics, particularly those related to the economics of happiness.
In addition to standard "objective" data such as on income, consumption, schooling attainment, mortality and job insecurity, our study relies heavily on surveys of reported well-being, both at a general level and in specific domains. This is a fundamental departure from traditional economics. Most economists historically have shied away from the use of survey data on expressed well-being and instead relied on revealed preferences—via consumption and investment choices—as a basis for analysis. The rationale is that answers to questions about well-being are not good signals of the preferences and constraints underlying actual behavior; there is no consequence to answering surveys, thus they do not clearly identify underlying preferences and/or constraints. In contrast, consumption and investment choices usually reflect a conscious choice about expenditure trade-offs. If the constraints are known, then preferences can be identified more accurately. In addition, unobserved heterogeneities in factors such as personality traits can bias the manner in which people answer surveys.
Yet in recent years the use of survey data on well-being has become more accepted. The increase in the number of economists working with such survey data over the past decade has been nothing short of remarkable. In part this is because econometric techniques have been developed that allow control for some of the unobserved error/personality traits that might significantly bias the interpretation of responses on well-being as reflecting underlying preferences. And the increasing coverage of large samples across countries and regions with varying levels of development and other conditions gives survey data on well-being tremendous analytical potential. Research by psychologists, meanwhile, demonstrates that answers to surveys are in large part validated by psychological and neurological measures of happiness and well-being.
While the traditional reliance on revealed preferences has informed a number of important questions, there are some limitations. One is that there is increasing evidence that factors other than rational choice may drive important consumption and investment choices, including addiction or self control problems on the one hand and low expectations norms on the other. These factors may help explain a number of consumption choices that would otherwise appear perverse from a welfare perspective. A second is that there are a number of choices that revealed preferences approaches cannot answer, precisely because individuals are constrained in their abilities to make those choices, due to lack of agency, lack of information, and other factors related to poverty or discrimination, among other reasons.
The use of survey data on well-being is not without its problems, however, some of which are relevant to all kinds of data and some of which are more important for survey questions on well-being. A particularly important question in terms of well-being data is that of question framing. How questions are asked, and where they are placed in the survey can make a big difference in the responses. Subjective well-being has both cognitive and affective components, and different questions capture more or fewer of these elements. Open-ended happiness questions, for example, capture more of the affective components than do more framed questions, such as the best possible life question or domain satisfaction questions. Which questions are used can have significant effects on the relationship that is found between happiness and critical correlates.
An important example is the relationship between happiness and income. There is a major debate in the literature on the relationship between happiness and income across countries and over time. This began with Richard Easterlin's (1974) seminal work demonstrating that there was not a relationship between happiness and income over time as countries got wealthier—the so-called Easterlin paradox. While Easterlin found that countries that were desperately poor were less happy than those above a minimum per capita income threshold, his work also showed that above that threshold happiness does not seem to increase with additional income, at least not in a linear manner. There is still a great deal of debate about whether there is an Easterlin paradox or not.
Some recent work based on a new and broader cross-country dataset from the Gallup World Poll has challenged the Easterlin paradox and highlights a strong and consistent relationship between per capita income levels and average country-level happiness. The findings are surely provocative, but it is not clear that they are fully comparable to Easterlin's, for two reasons. The first is question framing. The "life satisfaction" variable they rely on in the Gallup Poll is based on Cantril's 1965 ladder of life question, which asks respondents to compare their lives to the best possible life they can imagine. That frames the question much more than does an open-ended life satisfaction question (as Easterlin used). Surely when asked to compare, respondents in very poor countries are aware that life is likely better in wealthier ones, not least because of widespread access to the media and the Internet.
Affect questions typically (and not surprisingly) have less of a relationship with income than do more framed questions, such as the best possible life. Graham, Chattopadhyay, and Picon (2008), using the Gallup World Poll, find that the income and happiness relationship holds across countries for the best possible life, economic satisfaction, and the economic ladder questions, but not for life purpose and freedom to choose in life questions. Howell and Howell (2008), using a meta-analysis of 111 samples from 54 developing countries, find that the income-subjective well-being relationship is stronger when subjective well-being is measured as life satisfaction—a cognitive assessment (instead of happiness, an emotional assessment).
Secondly, there were far fewer developing countries in the sample when Easterlin did his original work. Since then, the Gallup Poll has included a large number of new countries: small, poor countries in sub-Saharan Africa and the transition economies, which have seen the dismantling of existing social welfare systems and dramatic falls in happiness. So the more linear relationship between income and happiness may not be driven by rising incomes and happiness in the rich countries, but by low or falling incomes and happiness in a large number of small countries at the bottom of the distribution.
Yet even the opposite extremes of the debate recognize that there is some relationship, albeit a modest one, between income and happiness across countries, with wealthier countries showing higher levels of happiness than much poorer ones. Graham and Pettinato (2002), based on a large sample of Latin American (and OECD) countries and an open-ended life satisfaction question, also demonstrate this trend, although within each of the poor and rich country sets there is not a clear income-happiness relationship (see figure 1-1). Howell and Howell find that the income-subjective well-being relationship is weakest among the wealthiest developing countries in their 54-country sample, and strongest among the poorest and least educated samples. They also find that the effect is strongest when economic status is defined as a stock variable (wealth or assets) rather than as a flow variable (income). Thus conclusions about the critical relationship between happiness and income are very much affected by the questions that are used and by the sample of countries that is available.
Another issue is where in the survey well-being questions are placed. Best practice suggests that open-ended life satisfaction or happiness questions should be placed first in the survey so that they are not biased or framed by other questions, such as about the nature of employment, financial situation, or marital status. Yet not all surveys employ this practice, and scholars have no choice but to compare happiness questions across countries or regions that are in different places in surveys, adjusting to the extent possible the expected bias caused by question placement.
Like all data, working with survey data on well-being requires accepting such limitations and potential biases. Regardless, the progress that economists and psychologists have made in recent years in the collection and analysis of survey data on well-being is providing new insights for assigning relative weights to income and non-income variables in determining human well-being or welfare.
The variables used in this study to measure well-being and quality of life are by definition multi-faceted. We use individual and national level variables, as well as objective and subjective indicators. For objective indicators at the individual level, we use the standard socioeconomic and demographic variables, such as age, gender, employment status, housing characteristics, income, and schooling levels. We also include assessments of interpersonal conditions, such as friendships and community engagement, and relative measures of economic status, such as reference group income. These objective indicators are matched with respondents' assessments of their happiness, life satisfaction, satisfaction with domains such as security, health, education, and jobs, as well as assessments of the general situation of the country and the state of public services.
At the national level, our variables include economic indicators such as GDP, inflation, and levels of poverty and inequality, as well as political and institutional variables, such as rule of law and the quality of public institutions. These are then benchmarked against national averages of individual responses to happiness and life satisfaction questions, as well as those about the general situation of the country, economy, and public services.
We rely primarily on the Gallup World Poll for our measures of well-being. The world poll has been conducted annually for years, but since 2006 has expanded its coverage to over 130 countries, of which 23 to 25 are from Latin America, depending on the year, and also includes several questions about life satisfaction. Approximately 1,000 individuals are interviewed in each country, and the surveys are nationally representative. There are over 100 quality-of-life-related questions in the Gallup World Poll, and 25 of them are asked in Latin America only. We complement the Gallup World Poll data with various other data sources, such as standard household surveys on income and employment. We also make some use of the Latinobarometro public opinion survey, which has been polling approximately 18,000 respondents across 18 countries in the region for over a decade. The survey also covers approximately 1,000 respondents per country and achieves national representation in most.
In departing from a strict reliance on traditional income indicators, we accept their importance, but also allow for the possibility that they are deficient along a number of dimensions. Gross Domestic Product (GDP), for example, excludes non-market activities that may affect welfare (for example, leisure time, social interactions) and includes market activities that have negative externalities for quality of life (such as pollution-increasing economic activity). We attempt to complement these traditional data with various indicators of quality of life. Happiness is meant to be a comprehensive evaluation of people's lives, and therefore an essential part of our analysis of quality of life. But aggregate measures of happiness are also deficient, as they are influenced by psychological and cultural biases, as well as inter-temporal problems: what makes people happy in the short run (say lower taxes) may not make them better off in the long run.
Because any one measure of well-being alone is likely to be insufficient to accurately and fully assess welfare or quality of life, we rely on a variety of measures and comparison of the results that they yield to get deeper insights into quality of life in the region than any one measure alone would yield. Indeed, we believe that some of the most important information is in the gaps or inconsistencies in the results based on different measures.
In the studies presented in this book, as well as earlier ones, we find a number of paradoxes in which positive results as assessed by one indicator are negative as assessed by another. While the authors discuss each of these in detail in chapter 3, they are worth noting in this chapter for two reasons. Firstly, they challenge some of the standard assumptions in traditional economics, such as the relationship between income and happiness, and provide a conceptual frame for our analysis. Secondly, they have important substantive links with several of the region's realities, such as high levels of income inequality, volatile macroeconomic performance and high levels of insecurity, and generally low expectations of public institutions and faith in the capacity of the state to deliver.
Main Findings
Happiness in Latin America
The simplest and perhaps most fundamental exercise in the book is to assess the determinants of happiness across the region and over time. Building from earlier work on happiness in the region by Graham and Pettinato (2002), based on the Latinobarometro, we ran similar happiness equations for the 18 countries in the region based on more recent Gallup data. While there are some modest differences, including in the phrasing of the happiness question (an open-ended happiness question in the Latinobarometro and the best possible life question in Gallup), the baseline results are very similar. This serves, among other things, as a robustness check.
As in virtually every context where happiness has been studied, there is a positive relationship between income and happiness within countries in Latin America, although it is not necessarily a linear one. Also as in other contexts, the relationship between age and happiness has a U-shape, with the low point being somewhere between 50 years (in the Latinobarometro) and 61 years of age (based on Gallup and the best possible life question). Employment, health, and marriage are all positively and significantly correlated with happiness, while being unemployed is negatively correlated. Women are happier than men in the Gallup dataset, while there is no gender difference in the Latinobarometro. Again, question phrasing may make a difference here, as the Gallup question is the best possible life ladder, while the Latinobarometro is an open-ended happiness question, and men and women may respond differently to the frames.
The Gallup Poll has some additional variables that correlate with happiness, some of them to a surprising degree. Friendships matter to the well-being of the average Latin American more than health, employment or personal assets, and only slightly less than food security. This varies according to income levels, meanwhile, with the rich valuing work and health more, and the poor valuing friendships (see figure 4-3 in chapter 4). These friendships most likely provide important coping mechanisms for the poor in the absence of publicly provided safety nets. Whether they serve as strong or weak ties in the Granovetter (1973) sense, with weak ties being more important for upward mobility and for acquisition of information, is an open question. Reporting religion to be important and having access to a telephone are also positively correlated with happiness in Latin America.
The life domains that are most relevant to happiness in the region are economic satisfaction, the importance of friends, work, health, and housing satisfaction (in that order of importance). Trust in education and satisfaction with one's city, meanwhile, are negatively correlated with happiness. Having a high optimism score, defined as higher levels of happiness than socio-demographic traits would predict (based on principal components analysis across domains), a variable that proxies for innate character traits, is strongly and positively correlated with life satisfaction. We use this residual or optimism score as a proxy or control for inherent character traits in many of our regressions.
When individual perceptions are compared to national ones, individuals in the region are typically more positive about their own individual situation than they are about their country's situation. This holds for both general satisfaction and satisfaction with one's economic situation. These findings reflect a general optimism bias among individuals in the region. In both cases, there are gaps between perceived and objective indicators, with a downward bias in the perceived indicators that is much stronger for richer than for poorer countries. This likely reflects rising aspirations in the wealthier countries, among other factors.
(Continues...)
Excerpted from PARADOX AND PERCEPTION Copyright © 2009 by THE BROOKINGS INSTITUTION. Excerpted by permission of BROOKINGS INSTITUTION PRESS. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
Contents
Foreword....................xiAcknowledgments....................xiii
1 How Latin Americans Assess Their Quality of Life: Insights and Puzzles from Novel Metrics of Well-Being Carol Graham and Jere R. Behrman....................1
2 Objective and Subjective Deprivation Leonardo Gasparini, Walter Sosa Escudero, Mariana Marchionni, and Sergio Olivieri....................22
3 The Conflictive Relationship between Satisfaction and Income Eduardo Lora and Juan Camilo Chaparro....................57
4 Satisfaction beyond Income Eduardo Lora, Juan Camilo Chaparro, and María Victoria Rodríguez....................96
5 Vulnerabilities and Subjective Well-Being Mauricio Cárdenas, Carolina Mejía, and Vicenzo di Maro....................118
6 Health Perceptions and Quality of Life in Latin America Carol Graham and Eduardo Lora....................158
7 Education and Life Satisfaction: Perception or Reality? Mauricio Cárdenas, Carolina Mejía, and Vincenzo Di Maro....................192
8 Job Insecurity and Life Satisfaction Naercio Aquino Menezes-Filho, Raphael Botura Corbi, and Andrea Zaitune Curi....................227
Contributors....................249
Index....................251