2
Introduction
Quality of life is a broader concept than economic production and living standards. It includes the full range of factors that influences what we value in living, reaching beyond its material side. While some extensions of economic accounting (discussed in chapter 1) allow including some of the elements that shape quality of life in conventional measures of economic well-being, every approach based on resources (or on people’s command over commodities) remains limited in important ways. First, resources are means that are transformed into well-being in ways that differ across people: individuals with greater capacities for enjoyment or greater abilities for achievement in valuable domains of life maybe better off even if they command fewer economic resources. Second, many resources are not marketed, and even when they are, prices will differ across individuals, making it problematic to compare real income across people. Finally, many of the determinants of human well-being are aspects of people’s life-circumstances: they cannot be described as resources with imputable prices, even if people do make trade-offs among them. These arguments by themselves are sufficient to suggest that resources are an insufficient metric for quality of life. Which other metric should be used instead for assessing quality of life depends on the philosophical perspective taken.
While a long tradition of philosophical thought has addressed the issues of what gives life its quality, recent advances in research have led to measures that are both new and credible. This research suggests that the need to move beyond measures of economic resources is not limited to developing countries (the traditional focus of much work on “human development” in the past) but is even more salient for rich industrialized countries. These measures, while not replacing conventional economic indicators, provide an opportunity to enrich policy discussions and to inform people’s view of the conditions of the communities where they live. More importantly, the new measures now have the potential to move from research to standard statistical practice. While some of them reflect structural conditions that are relatively invariant over time but that differ systematically across countries, others are more responsive to policies and more suitable for monitoring changes over shorter periods of time. Both types of indicator play an important role in evaluating quality of life.
Conceptual Approaches to Measuring Quality of Life
Three conceptual approaches have retained the attention of the Commission as useful in thinking about how to measure quality of life.
• The first approach, developed in close connection with psychological research, is based on the notion of subjective well-being . A long philosophical tradition views individuals as the best judges of their own conditions. This approach is closely linked to the utilitarian tradition but has a broader appeal due to the strong presumption in many streams of ancient and modern culture that enabling people to be “happy” and “satisfied” with their life is a universal goal of human existence.
• The second approach is rooted in the notion of capabilities. This approach conceives a person’s life as a combination of various “doings and beings” (functionings) and of his or her freedom to choose among these functionings (capabilities). Some of these capabilities may be quite elementary, such as being adequately nourished and escaping premature mortality, while others may be more complex, such as having the literacy required to participate actively in political life. The foundations of the capability approach, which has strong roots in philosophical notions of social justice, reflect a focus on human ends and on respecting the individual’s ability to pursue and realize the goals that he or she values; a rejection of the economic model of individuals acting to maximize their self-interest heedless of relationships and emotions; an emphasis on the complementarities between various capabilities; and a recognition of human diversity, which draws attention to the role played by ethical principles in the design of the “good” society.
• The third approach, developed within the economics tradition, is based on the notion of fair allocations. The basic idea, which is common to welfare economics, is that of weighting the various non-monetary dimensions of quality of life (beyond the goods and services that are traded in markets) in a way that respects people’s preferences. This approach requires choosing a particular reference point for each of the various non-monetary dimensions, and obtaining information on people’s current situations and on their preferences with respect to these points. This approach avoids the pitfall of basing evaluations on an “average” willingness-to-pay that may disproportionately reflect the preferences of those who are better-off in society and focuses instead on equality among all of its members.
These approaches have obvious differences, but also certain similarities. For example, subjective well-being is sometimes claimed to encompass all capabilities, insofar as these refer to attributes and freedoms that people value (implying that enhancing their capabilities will improve people’s subjective states). However, proponents of the capability approach also emphasize that subjective states are not the only things that matter, and that expanding people’s opportunities is important in itself, even if this does not show up in greater subjective well-being. Similarly, both the capability and the fair allocation approaches rely on information on the objective attributes of each person, while differing in the ways in which these are weighted and aggregated. While the choice between these approaches is ultimately a normative decision, they all point to the importance of a number of features that go beyond command over resources. Measuring these features requires the use of types of data (i.e., responses to questionnaires and non-market observations of personal states) that are not captured by market transactions.
Subjective Measures of Quality of Life
For a long time, economists have assumed that it was sufficient to look at people’s choices to derive information about their well-being, and that these choices would conform to a standard set of assumptions. In recent years, however, much research has focused on what people value and how they act in real life, and this has highlighted large discrepancies between standard assumptions of economic theory and real-world phenomena. A significant part of this research has been undertaken by psychologists and economists based on subjective data on people’s reported or experienced well-being.
Subjective measures have always been part of the traditional tool-kit of economists and statisticians, as many features of our economies and societies are measured through people’s responses to a standard set of questions (for example, “unemployment” is typically measured based on people’s answers as to whether they worked at all in a specific reference week, whether they actively looked for a job and whether they would be available to start working in the near future). The specific feature of the subjective measures of quality of life discussed here is that what people report about their own conditions has no obvious objective counterpart: we can compare “perceived” and “actual” inflation, for example, but only respondents can provide information on their own subjective states and values. Despite this feature, a rich literature on these subjective measures concludes that they help to predict people’s behavior (e.g., workers who report more dissatisfaction in their work are more likely to quit their job), and that they are valid with respect to a range of other information (e.g., people who report themselves as “happy” tend to smile more and to be rated as happy by people around them; these self-reports are also correlated with electrical readings of the brain).
Subjective approaches distinguish between the dimensions of quality of life and the objective factors shaping these dimensions. In turn, the subjective dimensions of quality of life encompass several aspects. The first is represented by people’s evaluations of their life as a whole or of its various domains, such as family, work and financial conditions. These evaluations imply a cognitive exercise by each person and an effort to take stock of and summarize the full range of elements that people value (e.g., their sense of purpose, the fulfilment of their goals and how they are perceived by others). The second aspect is represented by people’s actual feelings, such as pain, worry and anger, or pleasure, pride and respect. To the extent that these feelings are reported in real time, they are less affected by biases due to memory and to social pressure related to what is deemed to be “good” in society. Within this broad category of people’s feelings, the research on subjective well-being distinguishes between positive and negative affects, as both characterize the experience of each person.
All these aspects of subjective well-being (cognitive evaluations, positive affects and negative affects) should be measured separately to get a satisfactory appreciation of people’s lives. Which of these aspects matters more, and for what purpose, is still an open question. Much evidence suggests that people act to achieve satisfaction in their choices, and that choices are based on memories and evaluations. But memories and evaluations can also lead to bad choices, and some choices are made unconsciously rather than by weighing the pros and cons of various alternatives.
Subjective reports of people’s life-evaluations and affects provide measures of quality of life that can be monitored over time; some of these measures can also be compared across countries in reliable ways. Probably more importantly, however, is that these measures provide information about the determinants of quality of life at the level of each person. These determinants include both features of the environment where people live and their individual conditions, and they vary depending on the aspect considered. For example, activities (such as commuting, working or socializing) may be more important for affects, while conditions (such as being married, or having a rewarding job) may be more important for life-evaluations. In both cases, however, these measures provide information beyond that conveyed by income. For example, in most developed countries younger and older people report higher evaluations of their life than prime-age people, a pattern that contrasts sharply with levels of income for the same groups.
One area where various subjective measures of people’s well-being agree is in pointing to the high costs of unemployment for people’s quality of life. People who become unemployed report lower life-evaluations, even after controlling for their lower income, and with little adaptation over time; unemployed people also report a higher prevalence of various negative affects (sadness, stress and pain) and lower levels of positive ones (joy). These subjective measures suggest that the costs of unemployment exceed the income-loss suffered by those who lose their jobs, reflecting the existence of non-pecuniary effects among the unemployed and of fears and anxieties generated by unemployment in the rest of society.
While the initiatives of individual researchers and commercial data providers have led to important advances in the measurement of subjective well-being, the data remain limited in terms of the statistical inferences that they allow. National statistical systems need to build on these efforts and incorporate questions about various aspects of subjective well-being in their standard surveys. They should also develop longitudinal studies that could support more valid inferences about the relative importance of the various factors at work.
Objective Features Shaping Quality of Life
Both the capability and the fair allocation approaches give prominence to people’s objective conditions and the opportunities available to them, while differing in how these features are valued and ranked. While these objective features may also have an instrumental value for subjective well-being, both of these conceptual approaches regard an expansion of people’s opportunities in these domains as intrinsically important for people’s lives.
The range of objective features to be considered in any assessment of quality of life will depend on the purpose of the exercise: is the goal to assess changes in conditions within national jurisdictions, or to compare these conditions across countries at different levels of development? Some features may matter as descriptors of people’s states (e.g., health), while others may reflect the freedoms that people have to pursue the goals that they value (e.g., political voice). While the question of which elements should belong to a list of objective features inevitably depends on value judgments, in practice most of these themes are shared across countries and constituencies, and there is a large degree of consistency among the various exercises that focus on measuring “well-being” and related concepts.
3 In general, measures for all these objective features highlight that how societies are organized makes a difference for people’s lives, and that their influences are not all captured by conventional measures of economic resources.
Health
Health is a basic feature shaping both the length and the quality of people’s lives. Its assessment requires good measures of both mortality and morbidity. Data gaps remain significant in both fields. Mortality statistics by age and gender document the risk of death confronting people and are used to calculate the expected length of a person’s life. These indicators are today available in all developed countries but remain limited in large parts of the developing world, in particular for adults, which limits the possibility of monitoring progress in achieving the UN Millennium Development Goals. Further, age-specific mortality statistics are vectors: to obtain a scalar measure of people’s lifespan, they need to be aggregated in suitable ways and standardized for differences in age-structure across countries and over time. While different aggregation formulas and standardization methods exist, they lead to different results and rankings when comparing countries with survival curves (by age) that cross each other. This suggests that a variety of mortality measures should be compiled and regularly monitored. Nonetheless, it is significant that non-monetary measures of people’s health can diverge significantly from conventional economic measures. For example, although France has a lower GDP per capita than the United States, its life-expectancy at birth is higher, and this advantage has been widening (from less than 6 months in 1960 to almost 2 years in 2006) even while its GDP per capita relative to the U.S. was falling (
Figure 2.1).
Figure 2.1. Gaps in GDP per capita and life expectancy at birth between the United States and France
Source: OECD data.

The state of progress is far more limited for statistics on morbidity, a situation that has led to long-standing disagreements about whether declines in mortality have been matched by similar declines in morbidity. Existing measures of morbidity rest on a variety of sources: records of people’s height and weight; diagnoses by health professionals; registers for specific diseases; and self-reports drawn for censuses and surveys. Some of these measures relate to the prevalence of diseases or injuries, while others refer to their consequences in terms of the functioning of the person affected (which also depends on the quality of treatment). Variations in the measures and underlying data are inevitable given the many manifestations of poor health, but this also poses a real obstacle to comparing countries and monitoring changes in people’s morbidity over time. Measures are even sparser when moving from physical to mental disorders, despite evidence that these affect (at least in mild forms) a large share of people, that most of these disorders go untreated, and that their incidence has been increasing in some countries.
The variety of dimensions of people’s health has led to several attempts to define a summary measure that combines both mortality and morbidity. However, although several combined indices of people’s health exist, none currently commands universal agreement. Further, they all inevitably rest on ethical judgments that are controversial, and on weights for various medical conditions whose legitimacy is not always clear.
The challenges posed by this variety of health measures are not confined to cross-country comparisons but extend to within-country comparisons. Recent research on inequalities in health status has highlighted several patterns. First, people from lower occupational classes who have less education and income tend to die at younger ages and to suffer, within their shorter lifetimes, a higher prevalence of various health problems. Second, these differences in health conditions do not merely reflect worse outcomes for people at the very bottom of the socio-economic scale but extend to people throughout the socio-economic hierarchy, i.e., they display a “social gradient”: for example, life expectancy in the United Kingdom increases when moving from unskilled manual workers to skilled ones, from manual to non-manual workers, from lower-ranked office workers to higher-ranked staff. While these patterns in health inequalities have an obvious relevance for assessing quality of life, existing measures do not allow cross-country comparisons of their magnitude, due to differences in the measures of health outcomes used, in the individual characteristics considered (education, income, ethnicity), and in the reference population and geographic coverage of the various national studies.
4
Education
A long tradition of economic research has stressed the importance of education in providing the skills and competencies that underpin economic production. But education matters for quality of life independently of its effects on people’s earnings and productivity. Education is strongly associated with people’s life-evaluations, even after controlling for the higher income it brings. Further, better-educated people typically have better health status, lower unemployment, more social connections and greater engagement in civic and political life. While the available evidence does not always allow conclusions about the directionality of causation between education and these other dimensions of quality of life (e.g., less healthy children may miss school more often), there is a consensus that education brings a range of returns (monetary and non-monetary) that benefit both the person investing in the education and the community in which they live. Measuring the size of these wider benefits of education is an important research priority, where progress requires better measures of people’s characteristics in a range of domains and surveys that follow the same individual over time.
Available educational indicators cover a broad range of fields. Some refer to inputs (e.g., school enrolment, educational expenditures and school resources), while others refer to throughputs and outputs (e.g., graduation rates, completed years of schooling, standardized test-measures of people’s achievements in terms of literacy and numeracy). Which of these indicators is more relevant depends on the stage of a country’s development and on the goal of the evaluation exercise. The available indicators highlight large differences across countries, with various educational indicators sometimes highlighting contrasting patterns. Some countries, for example, may combine excellence for students that reach university education with widespread underachievement for a large number of youth, mainly from households at the bottom of the socio-economic ladder. These differences would cancel out in summary measures of education (e.g., mean years of schooling) but have significance for any assessment of quality of life. Within countries, measures of inequality in learning outcomes are especially important for youth at the bottom of the achievement scale who are at risk of poverty or exclusion from well-paid and rewarding jobs in adult life. As education is an important predictor of many dimensions of people’s lives, all social surveys should systematically include information on the learning experiences of respondents and of their parents, as well as information on other features shaping the quality of their lives.
Some of the most relevant indicators for assessing the effect of education on quality of life are measures of people’s competencies. Several tools have been developed in recent years to measure these in standardized ways, though the tools still have significant limitations. First, and most obviously, not all countries currently implement these surveys. Second, many of these tools were not developed from the perspective of measuring people’s capabilities in a broad sense, but for the purpose of assessing educational policies, which typically required focusing on a more narrow set of measurable competencies. Third, existing assessment tools often have a narrow coverage, as schooling is only one of the inputs that lead to knowledge, skills development and improvements in quality of life. Information about the experiences and “soft” competencies acquired by children in their early years remains limited, despite increasing evidence that early-childhood experiences matter for people’s learning and quality of life in later years. Measurement tools also remain limited when it comes to comparing the competencies of students in higher education and to assessing workers’ experiences in terms of adult education and training (although this will change as new surveys of adult competencies are developed and implemented). As for other features of quality of life, the main problem for indicators in this domain is not the lack of detailed information on education per se, but rather the lack of surveys measuring both education and other outcomes that matter for quality of life at the individual level.
Personal Activities
How people spend their time and the nature of their personal activities matters for quality of life, irrespectively of the income generated. The activities that people engage in have effects on their subjective well-being, in terms of both their hedonic experiences (
Figure 2.2) and their evaluative judgements. More generally, people do not always “choose” among these activities in the same way as they allocate their budget among various goods, due to a lack of effective alternatives. Further, these choices will generally affect other people within the family and community, with some of these personal activities effectively representing indirect costs to production (e.g., commuting) rather than consumption.
Because of both political demands and the feasibility of providing concrete and comparable measures, the main activities discussed by the Commission have been paid work, unpaid work, commuting and leisure time. Housing, although not representing an activity per se, was also discussed because it provides the setting for a number of personal activities.
Figure 2.2. Ranking of personal activities based on women’s hedonic experiences and time devoted to them in selected cities in the United States and France
Activities ranked in decreasing order of enjoyment in the United States.
Source: Krueger, A.B.,D. Kahneman, D. Schkade, N. Schwarz and A. Stone (2008). “National Time Accounting: The Currency of Life,” NBER, forthcoming in A.B. Kruger (ed.), Measuring the Subjective Well-Being of Nations; National Accounts of Time Use and Well Being, Unoversity of Chicago Press, Chicago.

• Paid work matters for quality of life partly because it provides identity to people and opportunities to socialize with others. However, not all jobs are equally valuable in this respect. This underscores the importance of collecting more systematic information on the quality of paid work, as a number of international organizations have been doing in the context of their ongoing studies of “decent work.” Some national surveys provide information on many aspects of decent work, such as non-standard employment, gender gaps in employment and wages, discrimination in the workplace, opportunities for lifelong learning, access to employment for disabled persons, working time and “unsocial hours,” the work-life balance, work accidents and physical risks, work intensity, social dialogue and workers’ autonomy. Their practical use is, however, limited by small sample sizes and survey differences across countries.
• Unpaid domestic work, such as shopping and the care of children and other household members, is important from the perspective of assessing both the total amount of household services produced and how family chores are distributed between men and women.
• Commuting time is also key to the quality of work, and monitoring it requires information on the number of hours spent travelling to and from work during a specified period, as well as on the accessibility and affordability of transport.
• A long tradition of research has emphasized the importance of leisure-time for quality of life. This research points to the importance of developing indicators of both leisure quantity (number of hours) and quality (number of episodes, where they took place, presence of other people), as well as of measures of participation in cultural events and of “poor leisure” (such as the share of children who did not take a holiday away from home in the previous year).
• Finally, despite the importance of housing for a variety of social outcomes (such as children’s education), no core set of housing indicators currently exists for international comparisons. Remedying this situation would require better information on the number of people who are homeless or living in emergency shelters and on housing quality (e.g., in terms of the local services available and overcrowding).
In several cases, suitable indicators in these various fields already exist, and the challenge is to improve upon what has been achieved in the past. In other areas, however, existing measures remain seriously deficient, and progress requires investment in new statistical capacity. A case in point, cutting across all the personal activities described above, is that of measuring how people spend their time. Time is the natural metric for comparing personal activities and (as argued in chapter 1) an essential input to the construction of satellite household accounts. One priority should be to develop measurement tools grounded on clear definitions and based on surveys with a consistent design that are representative of patterns over a full year and are undertaken with sufficient regularity—all requirements that are rarely met. Ideally, these surveys should cover both the amount of time spent in various activities and the feelings that they produce. This is important, as the same activity can generate different hedonic experiences depending on people’s own conditions (e.g., whether they are unemployed or not); this information also matters for assessing inequalities between different groups in society (e.g., by gender). While these investments in statistical capacity are costly, and compete with other priorities, their pay-off for quality-of-life research is potentially huge.
Political Voice and Governance
Political voice is an integral dimension of the quality of life. Intrinsically, the ability to participate as full citizens, to have a say in the framing of policies, to dissent without fear and to speak up against what one perceives to be wrong are essential freedoms. Instrumentally, political voice can provide a corrective to public policy: it can ensure the accountability of officials and public institutions, reveal what people need and value and call attention to significant deprivations. Political voice also reduces the potential for conflicts and enhances the prospect of building consensus on key issues, with pay-offs for economic efficiency, social equity and inclusiveness in public life.
The opportunities for political voice and the degree of responsiveness of the political system depend on the institutional features of each country, such as the presence of a functioning democracy, universal suffrage, free media and civil society organizations. This also depends on some key aspects of governance, such as legislative guarantees and the rule of law. Legislative guarantees include both constitutional rights and rights provided by general laws that enhance the quality of life of all residents and that reflect the social consensus prevailing in different countries and times. The structure of laws can also affect the investment climate in a country and thus have an impact on market functioning, economic growth, job creation and material welfare. However, to realize their potential, legal guarantees require effective implementation and substantive justice, which depend on how various institutions (e.g., the police, the judiciary and various administrative services) function, whether they are free from corruption, political interference and social prejudice, and whether they can be held accountable for their decisions.
Comparisons based on existing indicators of political voice and governance highlight vast differences between countries, especially between those with a long history of democratic functioning and those that have moved from authoritarian to democratic regimes only more recently and that have not yet established the full range of freedoms and rights. Even in the developed world, however, low trust in public institutions and declining political participation point to a growing gap between how citizens and political elites perceive the functioning of democratic institutions. There are also systematic differences in how different groups exercise political voice, and with respect to fundamental rights and opportunities for civic participation in these countries, especially between citizens and the growing numbers of immigrants.
Indicators of political voice and governance should help to evaluate the functioning of multiparty democracy and universal suffrage, the level of participation in government decisions at the local level and the presence of a free media and various freedoms (e.g., to form and join civil organizations, trade unions and professional bodies, or to participate in civic and social activities). Relevant indicators should cover the rights embedded in constitutions, laws (e.g., that promote civil and criminal justice, equality, inclusion, accountability and affirmative action), international covenants on human rights and basic freedoms, as well as the functioning of the judicial system (e.g., its independence from corruption and political influences, the speed with which it delivers justice and its accessibility to both citizens and residents). Many of these indicators are typically compiled by bodies outside the boundaries of national statistical systems and are based mainly on the opinion of experts. These indicators need to be complemented, and in some cases replaced, by surveys of citizens’ own perceptions of how well the political, legal and executive institutions are functioning, the difficulties they face in accessing them and the trust that they place in them. Such surveys also need to capture inequalities in access to these institutions across socio-economic groups.
Social Connections
Social connections improve quality of life in a variety of ways. People with more social connections report higher life-evaluations, as many of the most pleasurable personal activities involve socializing. The benefits of social connections extend to people’s health and to the probability of finding a job, as well as to several characteristics of the neighborhood where people live (e.g., the prevalence of crime and the performance of local schools). These social connections are sometimes described as “social capital” to highlight the benefits (direct and indirect) that they bring. As with other types of capital, the externalities stemming from social capital can sometimes be negative: for example, belonging to a group may strengthen a sense of unique personal identity that fuels a climate of violence and confrontation with other groups. This, however, underscores the importance of better analyzing the nature of these social connections and the breadth of their effects, rather than underestimating their significance. The available evidence suggests that social connections benefit people in the networks, with effects on non-participants that depend on both the nature of the group and the effects being considered.
The drivers of change in people’s social connections are not always well understood. Social connections provide services to people (e.g., insurance, security), and the development of both markets and government programs may have reduced the ties of individuals with their community thanks to the provision of alternative arrangements. What is clear is that a decline in these ties may negatively affect people’s lives, even when their functions are taken up by market or government alternatives that increase the level of economic activity (such as when the informal surveillance of neighbors is replaced by salaried security guards). To avoid a biased assessment of human well-being, measures of these social connections are therefore needed.
Research on social connections has traditionally relied on proxy measures, such as the number of individual memberships in associations, or the frequency of activities assumed to result from social connections (e.g., altruistic behavior and voter turn-out). However, it is by now accepted that these are not good measures of social connections, and that reliable measures require surveys of peoples’ behaviors and activities. In recent years, a number of statistical offices (in the United Kingdom, Australia, Canada, Ireland, the Netherlands and, most recently, the United States) have started surveys that measure various forms of social connections. For example, special modules of the labor-force survey in the United States ask people about their civic and political engagement, their membership and voluntary work in various organizations, their relationship with neighbors and family members and how they get information and news. Similar surveys should be implemented elsewhere, based on questions and protocols that allow valid comparisons across countries and over time. Progress should also be made in measuring additional dimensions of social connections (such as trust in others, social isolation, availability of informal support in case of need, engagement in the workplace and in religious activities, friendship across lines of race, religion and social class) by building on the experience accumulated by some countries in these fields.
Environmental Conditions
Environmental conditions are important not only for sustainability, but also because of their immediate impact on the quality of people’s lives. First, they affect human health both directly (through air and water pollution, hazardous substances and noise) and indirectly (through climate change, transformations in the carbon and water cycles, biodiversity loss and natural disasters that affect the health of ecosystems). Secondly, people benefit from environmental services, such as access to clean water and recreation areas, and their rights in this field (including rights to access environmental information) have been increasingly recognized. Third, people value environmental amenities or disamenities, and these valuations affect their actual choices (e.g., of where to live). Lastly, environmental conditions may lead to climatic variations and natural disasters, such as drought and flooding, which damage both the properties and the lives of the affected populations.
Measuring the effects of environmental conditions on people’s lives is, however, complex. These effects manifest themselves over different timescales, and their impacts vary depending on people’s characteristics (e.g., where they live and work, their metabolic intake). Further, the strength of these relations is often underestimated because of limits in current scientific understanding and in the extent to which various environmental factors have been subject to systemic investigations.
Much progress has been achieved in the last two decades in terms of measuring environmental conditions (through better environmental data, the regular monitoring of indicators and accounting tools), understanding their impacts (e.g., evaluation of related morbidity and mortality, labor productivity, the economic stakes associated with climate change, biodiversity change, damage from disasters) and establishing a right of access to environmental information. A range of environmental indicators can be used to measure human pressure on the environment, the responses from administrations, firms and households to environmental degradation and the actual state of environmental quality.
However, from a quality-of-life perspective, existing indicators remain limited in important respects. For example, emissions indicators refer mainly to the aggregate quantities of various pollutants, rather than to the share of people exposed to dangerous doses. Existing indicators should hence be supplemented in a number of ways, including the regular monitoring of the number of premature deaths from exposure to air pollution; the number of people who lack access to water services and nature, or who are exposed to dangerous levels of noise and pollution; and the damage inflicted by environmental disasters. Survey measures of people’s own feelings and evaluations of the environmental conditions of their neighborhood are also needed. Because many of the effects of environmental conditions on quality of life differ across people, these indicators should refer to people grouped according to various classification criteria.
Personal Insecurity
Personal insecurity includes external factors that put at risk the physical integrity of each person: crime, accidents, natural disasters and climate changes are some of the most obvious factors.
5 In extreme cases, these factors can lead to the death of the person involved. While these elements account for only a minority of all deaths, and they are captured by general mortality statistics, one rationale for having specific measures of their frequency is that their effect on people’s emotional lives is very different than that of deaths related to medical conditions, as shown by the large impact of bereavement on people’s subjective well-being.
Less extreme manifestations of personal insecurity such as crime affect quality of life for a significantly larger number of people, with even larger numbers reporting fear of being a victim of a physical aggression. One of the most remarkable features of reports on subjective fear of crime is how little the fear is related to experienced victimization: countries with a higher share of people reporting fear of crime do not experience a higher victimization while, within countries, older and richer people feel more unsafe than younger and poorer people, despite being less likely to be a victim of crime.
These patterns highlight the importance of developing more regular and reliable measures of personal security to orient public discussion. Victimization surveys are an essential tool to assess the frequency of crime and the fear it generates. Other tools need to be mobilized to assess other threats to personal security, such as domestic violence and violence in countries ravaged by conflict and war.
Economic Insecurity
Uncertainty about the material conditions that may prevail in the future reflects the existence of a variety of risks, in particular for unemployment, illness and old age. The realization of these risks has negative consequences for the quality of life, depending on the severity of the shock, its duration, the stigma associated with it, the risk aversion of each person and the financial implications.
Job loss can lead to economic insecurity when unemployment is recurrent or persistent, when unemployment benefits are low relative to previous earnings or when workers have to accept major cuts in pay, hours or both to find a new job. The consequences of job insecurity are both immediate (as replacement income is typically lower than the earnings on the previous job) and longer term (due to potential losses in wages when the person does find another job). While indicators of these consequences are available, cross-country comparisons are difficult, requiring special investments in this direction. Job insecurity can also be measured by asking workers either to evaluate the security of their present job or to rate their expectation of losing their job in the near future. The fear of job loss can have negative consequences for the quality of life of the workers (e.g., physical and mental illness, tensions in family life) as well as for firms (e.g., adverse impacts on workers’ motivation and productivity, lower identification with corporate objectives) and society as a whole.
Illness can cause economic insecurity both directly and indirectly. For people with no (or only partial) health insurance, medical costs can be devastating; forcing them into debt, to sell their home and assets or to forego treatment at the cost of worse health outcomes in the future. One indicator of illness-related economic insecurity is provided by the share of people without health insurance. However, health insurance can cover different contingencies, and even insured people may incur high out-of-pocket health expenses in the event of illness. To these out-of-pocket health expenses should be added the loss of income that occurs if the person has to stop working and the health (or other) insurance does not provide replacement income.
Old age is not a risk per se, but it can still imply economic insecurity due to uncertainty about needs and resources after withdrawal from the labor market. Two types of risk, in particular, are important. The first is the risk of inadequate resources during retirement, due to insufficient future pension payments or to greater needs associated with illness or disability. The second is the risk of volatility in pension payments: while all retirement-income systems are exposed to some types of risk, the greater role of the private sector in financing old-age pensions (in the form of both occupational pensions and personal savings) has made it possible to extend the coverage of pension systems in many countries but at the cost of shifting risk from governments and firms towards individuals, thereby increasing their insecurity.
The many factors shaping economic insecurity are reflected in the variety of approaches used to measure them. Some approaches try to quantify the frequency of specific risks, while others look at the consequences of a risk that materializes and at the means available to people to protect themselves from these risks (especially resources provided by social security programs). A comprehensive measure of economic insecurity would ideally account for both the frequency of each risk and its consequences, and some attempts in this direction have been made. A further problem is that of aggregating across the various risks that shape economic insecurity, as the indicators that describe these risks lack a common metric to assess their severity. A final, even more intractable problem is that of accounting for the long-term consequences for quality of life of the various policies used to limit economic insecurity (through their effects on unemployment and labor-force participation).
Cross-Cutting Issues
Most of the measurement challenges described above are specific to each dimension of quality of life, and the Commission has only hinted at some of the work required, leaving it to agencies with expertise in each field to detail concrete action plans. Other challenges, however, are cross-cutting and are unlikely to be picked up through initiatives undertaken separately in each field.
6 Three of these issues deserve special attention.
Inequalities in Quality of Life
The first cross-cutting challenge for quality-of-life indicators is to detail the inequalities in individual conditions in the various dimensions of life, rather than just the average conditions in each country. To some extent, the failure to account for these inequalities explains the “growing gap”—identified by the French Presidency when establishing the Commission—between the aggregate statistics that dominate policy discussions and people’s sentiments about their own conditions.
While established methodologies and data sources can be used to measure inequalities in the distribution of economic resources in a fairly reliable way, the situation is much less satisfactory with respect to the non-monetary dimensions of quality of life. This is especially true given that these inequalities cannot always be described through information on the size of the distribution of these features around their mean. For example, differences in the lifespan of people may reflect genetic differences that are randomly distributed in the population: in these circumstances, narrowing the overall distribution of life duration would not make society less “unequal” in any morally compelling way.
The problems, however, go deeper than developing suitable measures. There are many inequalities, and each is significant in itself: this suggests that we should avoid the presumption that one of them (e.g., income) will always encompass all others. At the same time, certain inequalities may be mutually reinforcing. Gender disparities, for example, while pervasive in most countries and groups, are typically much larger for households with lower socio-economic status: in many developing countries, the combined effect of gender and socio-economic status is often to exclude young women in poor households from attending school and getting rewarding jobs, denying them possibilities of self-expression and political voice and exposing them to hazards that put their health at risk. The measurement of some of these inequalities (such as those related to class and socio-economic status) has contributed, over the years, to a wide array of policies and institutions aimed at reducing their intensity and consequences. Other types of inequality, such as between ethnic groups, are more recent (at least in countries that have experienced large waves of immigration) and are set to become more politically salient in the future as immigration continues.
It is critical that these inequalities be assessed in a comprehensive way, by looking at differences in quality of life across people, groups and generations. Further, as people can be classified according to different criteria, each with some relevance for people’s lives, inequalities should be measured and documented for a plurality of groups. Appropriate surveys should be developed to assess the complementarities between the various types of inequality and to identify their underlying causes. It is up to the statistical community to regularly feed these analyses with suitable data.
Assessing Links Across Quality-of-Life Dimensions
The second cross-cutting challenge, already alluded to above, is to better assess the relationship between the various dimensions of quality of life. Some of the most important policy questions involved relate to how developments in one area (e.g., education) affect developments in others (e.g., health status, political voice and social connections), and how developments in all fields are related to those in income. While some of these relationships, in particular at the individual level, are poorly measured and inadequately understood, ignoring the cumulative effects of multiple disadvantages leads to sub-optimal policies. For example, the loss of quality of life due to being both poor and sick far exceeds the sum of the two separate effects, implying that governments may need to target their interventions more specifically at those who cumulate these disadvantages.
Assessing these links across the various dimensions of quality of life is not easy, as statistical systems continue to be highly segmented across disciplines, with measurement instruments in each field paying only scant attention to developments in other domains. But progress can be achieved by developing information about the “joint distribution” of the most salient features of quality of life (such as hedonic experiences, health status, education, political voice) across all people. While the full development of this information could be achieved only in the distant future, concrete steps in this direction could be accomplished by including in all surveys a few standard questions that allow classifying respondents based on a limited set of characteristics, and that describe their conditions in a broad range of fields. Investment should also be made in developing longitudinal surveys that could allow both controlling for people’s personal characteristics and better analyzing the directionality of causation between the different factors shaping life.
Aggregating Across Quality-of-Life Dimensions
The third cross-cutting challenge to quality-of-life research is to aggregate the rich array of measures in a parsimonious way. The issue of aggregation is both specific to each feature of quality of life (as in the case of measures that combine mortality and morbidity in the health field) and more general, requiring the valuation and aggregation of the achievements in various domains of life, both for each person and for society as a whole. The search for a scalar measure of quality of life is often perceived as the single most important challenge faced by quality-of-life research. While this emphasis is partly misplaced—the informational content of any aggregate index will always reflect the quality of the measures used in its construction—the demands in this field are strong, and statistical offices should play a role in answering them.
Traditionally, the most common response to this demand for parsimony in quality-of-life research has been to aggregate a number of indicators (suitably selected and scaled) of average performance in various fields at the country-level. The best known example of this approach is the Human Development Index. This measure has played (and continues to play) an important communication role, leading to country-rankings that differ significantly from those based on per-capita GDP, especially for some less-developed countries. However, the choices on the weights used to construct this (and other similar) indices reflect value judgments that have controversial implications: for example, adding the logarithm of per-capita GDP to the level of life expectancy (as done by the Human Development Index) implicitly values an additional year of life expectancy in the United States as worth 20 times an additional year of life in India. More fundamentally, being based on country-averages, these measures ignore the significant correlations between the various features of quality of life across people, and do not say anything about the distribution of these individual conditions within each country. For example, the scalar index will not change if average performance in each domain remains the same while the accumulation of advantages or disadvantages for the same person across various domains of life changes over time.
Several aggregate measures of quality of life are possible, depending on the philosophical perspective taken and the question addressed. Some of these measures are already being used sporadically (e.g., average levels of life-satisfaction for a country as a whole, and composite indices such as the Human Development Index, which is mainly focused on developing countries) and could be extended through questionnaire-based measures of people’s psychological health, feelings and evaluations, and through consideration of additional dimensions of quality of life. Others could be implemented if national statistical systems made the necessary investment to provide the type of data needed to allow their computation. For example, the U-index, i.e., the proportion of one’s time in which the strongest reported feeling is a negative one (see
Figure 2.2), requires collecting information on emotional experiences during specific episodes through time-use surveys. Similarly, methods based on counting the occurrences and severity of various objective features for each person (which is linked to the capability approach), before proceeding to construct country-averages, require information on the joint distribution of various objective features. Finally, the notion of “equivalent income” (which is linked to the fair allocations approach) requires information on people’s states in various dimensions of quality of life, and on their preferences with respect to these states (for a given reference level in each).
Figure 2.3. Characteristics of the most deprived people according to different measures of quality of life, Russia in 2000.
Source: Fleurbacy M., E. Schokkaert and K. Decancq (2009) “What Good Is Happiness?” CORE Discussion Paper, 2009/17, Université Catholique de Louvain, Belgium. Computations based on data from the Russia Longitudinal Monitoring Survey.
In general, different approaches will lead to distinct scalar measures of quality of life for each country, and to different characteristics of the people classified as “worse-off.” For example, in a sample of Russian respondents, people in the bottom quintile of the distribution of equivalent income report worse health and a higher incidence of unemployment compared to people identified as “worse-off” based on either their consumption expenditure or their subjective life-evaluations (
Figure 2.3). This suggests that, rather than focusing on constructing a single summary measure of quality of life, statistical systems should provide the data required for computing various aggregate measures according to the philosophic perspective of each user.
Main Messages and Recommendations
Quality of life includes the full range of factors that make life worth living, including those that are not traded in markets and not captured by monetary measures. While some extensions of economic accounting include some additional elements that shape quality of life in conventional money-based measures of economic well-being, there are limits on how much this approach can achieve. Other indicators have an important role to play in measuring social progress, and recent advances in research have led to new and credible measures for at least some aspects of quality of life. These measures, while not replacing conventional economic indicators, provide an opportunity to enrich policy discussions and to inform people’s view of the conditions of the communities in which they live; today, they have the potential to move from research to standard statistical practice. The Commission’s recommendations in this field can be summarized as follows:
Recommendation 1: Measures of subjective well-being provide key information about people’s quality of life. Statistical offices should incorporate questions to capture people’s life-evaluations, hedonic experiences and priorities in their own surveys.
Research has shown that it is possible to collect meaningful and reliable data on subjective well-being. Subjective well-being encompasses different aspects (cognitive evaluations of one’s life, positive emotions such as joy and pride and negative emotions such as pain and worry): each of them should be measured separately to derive a more comprehensive appreciation of people’s lives. Quantitative measures of these subjective aspects hold the promise of delivering not just a good measure of quality of life per se, but also a better understanding of its determinants, reaching beyond people’s income and material conditions. Despite the persistence of many unresolved issues, these subjective measures provide important information about quality of life. Because of this, the types of questions that have proved their value within small-scale, unofficial surveys should be included in larger-scale surveys undertaken by official statistical offices.
Recommendation 2: Quality of life also depends on people’s objective conditions and opportunities. Steps should be taken to improve measures of people’s health, education, personal activities, political voice, social connections, environmental conditions and security.
The information relevant to valuing quality of life goes beyond people’s self-reports and perceptions to include measures of their functionings and freedoms. While the precise list of these features inevitably rests on value judgments, there is a consensus that quality of life depends on people’s health and education, their everyday activities (which include the right to a decent job and housing), their participation in the political process, the social and natural environment in which they live and the factors shaping their personal and economic security. Measuring all these features requires both objective and subjective data. The challenge in all these fields is to improve upon what has already been achieved, to identify gaps in available information, and to invest in statistical capacity in areas (such as time-use) where available indicators remain deficient.
Recommendation 3: Quality-of-life indicators in all the dimensions they cover should assess inequalities in a comprehensive way.
Inequalities in human conditions are integral to any assessment of quality of life across countries and the way that it is developing over time. Each dimension of quality of life requires appropriate measures of inequality, with each of these measures being significant in itself and none claiming absolute priority over others. Inequalities should be assessed across people, socio-economic groups and generations, with special attention to inequalities that have arisen more recently, such as those linked to immigration.
Recommendation 4: Surveys should be designed to assess the links between various quality-of-life domains for each person, and this information should be used when designing policies in various fields.
It is critical to address questions about how developments in one domain of quality of life affect other domains, and how developments in all the various fields are related to income. This is important because the consequences for quality of life of having multiple disadvantages far exceed the sum of their individual effects. Developing measures of these cumulative effects requires information on the “joint distribution” of the most salient features of quality of life across everyone in a country through dedicated surveys. Steps in this direction could also be taken by including in all surveys some standard questions that allow classifying respondents based on a limited set of characteristics. When designing policies in specific fields, indicators pertaining to different quality-of-life dimensions should be considered jointly, to address the interactions between dimensions and the needs of people who are disadvantaged in several domains.
Recommendation 5: Statistical offices should provide the information needed to aggregate across quality-of-life dimensions, allowing the construction of different scalar indices.
While assessing quality of life requires a plurality of indicators, there are strong demands to develop a single scalar measure. Several scalar measures of quality of life are possible, depending on the question addressed and the approach taken. Some of these measures are already being used, such as average levels of life-satisfaction for a country as a whole, or composite indices that aggregate averages across domains, such as the Human Development Index. Others could be implemented if national statistical systems made the necessary investment to provide the data required for their computation. These include measures of the proportion of one’s time in which the strongest reported feeling is a negative one, measures based on counting the occurrence and severity of various objective features of people’s lives and (equivalent-income) measures based on people’s states and preferences.