Measuring Maori Wellbeing: Concepts and Content - Developing 2 Indices of Standard of Living and Quality of Life

World Development Vol. 35, No. 7, pp. 1259–1276, 2007
2007 Elsevier Ltd. All rights reserved

Multidimensional Measures of Well-Being: Standard of Living and Quality of Life Across Countries

VALE´ RIEBERENGER
Universite´ de Nice-Sophia Antipolis, CEMAFI
(Centre d’Etudes en Macroe´conomie et Finance Internationale), Nice, France
and
AUDREY VERDIER-CHOUCHANE *
African Development Bank, Tunis, Tunisia

International organizations now recognize that human development goes beyond economic growth and is a multidimensional phenomenon covering all aspects of well-being.

This partly dates from Sen’s work on social justice and inequalities (Sen, 1985, 1992), which inspired a new concept of development. Sen’s capability approach contributed to the design of the UNDP Human Development Index (HDI) in 1990, which was intended as a more comprehensive indicator than per capita income for comparing the well-being of countries.

However, the HDI’s critics say its indicators are too few and too arbitrarily chosen and that its definition is still inadequate and does not allow the capability approach to work.

This article defines two composite indices— SL (Standard of Living) and QL (Quality of Life)—across 170 countries, supporting Sen’s capability approach.

......

2. CONCEPTS FOR STANDARD OF LIVING AND QUALITY OF LIFE
GDP per capita is the most commonly used indicator to compare wealth among countries and is a measure of well-being and development exclusively based on material wealth. However, insufficient income is merely one dimension of under-development, so development cannot be understood by only taking into account economic performance.

Attempts were made in the 1970s to construct socio-economic indicators as an alternative to GDP per capita, which was criticized as capturing neither distributional aspects nor social and human welfare
dimensions (Desai, 1991). The standard human development concept dates from the 1990 Human Development Report (UNDP, 1990). It drew on Sen’s work and led to growing acknowledgement of the multidimensional nature of development and to different strategies for moving from promotion of growth to promotion of well-being. It also highlighted the need to construct alternative composite indices, including non-monetary indicators, to assess the achieved development levels. The HDI was launched in 1990 to represent the broad ideas included in the human development concept.

So the capability approach has helped enlarge the concept of human development. International organizations such as the World Bank (2006) have adopted notions of ‘‘quality of growth’’ and ‘‘pro-poor growth’’ that reflect greater concern about non-monetary dimensions of well-being.

(a) Sen’s capability approach and the need for going beyond the HDI

Sen’s capability approach (1985) proposes a normative framework to evaluate individual well-being, social relationships and changes in society. Its main components are the ‘‘commodities’’ or resources, the ‘‘functionings’’ and the ‘‘capabilities.’’ The ‘‘commodities are all goods and services, not just merchandise. They can include transfers in kind and make possible the ‘‘functionings,’’ which take into account achievements of individuals— what they ‘‘are’’ and what they ‘‘make’’ with their resources—and reflects life-style. The concept of ‘‘capabilities’’ is related to ‘‘functionings’’ but also includes notions of opportunity and freedom—the range of opportunities a person has and can choose from. ‘‘Capabilities’’ are various combinations of functionings (beings and doings) that the person can achieve. ‘‘Capability is, thus, a set of
vectors of functionings, reflecting the person’s freedom to lead one type of life or another (...) to choose from possible livings’’ (Sen, 1992, p. 40). A functioning is an achievement, while capability is about the ability to produce it. Functionings are thus more directly related to living conditions, while capability is a concept of freedom, in a positive sense. According to Sen’s definition, the UNDP (1997) defines human development as increasing people’s choices by expanding their human capabilities and opportunities. Under-development is thus not a deprivation of basic needs, but is a deprivation of basic capabilities or freedoms that would allow an individual to have the kind of life he/she wants.

Sen’s (1999) approach is qualitative and multidimensional. He puts human beings at the center of the development concept. The aim of development is to enhance human capabilities so as to lead full, productive
and satisfying lives. A higher income is necessary but not sufficient and thus calls for wider measures of well-being. The HDI was supposed to use Sen’s approach to make international comparisons. It has been improved, in particular, for GDP calculation and extremes fixing. It has also led to other indices such as the Human Poverty Index (HPI) which assesses human development not on average national achievement but on how many people live in deprivation. Several heterodox World Bank studies consider indicators of inequality for analysis of true development. But Easterly (2002), who examines inequality as a barrier to prosperity and growth, and Pritchett, Suryahadi, and Sumarto (2000), who look at vulnerability to poverty, use household data. After GDP per capita, the HDI is the most discussed measure of well-being. The literature seems to take two approaches that are not inevitably exclusive of each other.

In the first, the main criticism of the HDI relates to its very narrow definition of human well-being. New indices, sometimes excluding the income component, have been proposed without having content necessarily justified or based on an explicit theoretical approach of well-being.

• The Physical Quality of Life Index (PQLI) developed by Morris (1979) takes into account life expectancy, infant mortality, and literacy. 
• The Quality of Life Index of Dasgupta and Weale (1992) adds civil liberties and political rights to the HDI. The Index of Economic Well-Being proposed by Osberg and Sharpe (1998) is similar, though it also takes into account economic aspects of well-being neglected by GDP per capita (such as production stocks,
• unequal income distribution and uncertainty about future income). 
• Rahman, Mittelhammer, and Wandschneider (2003) propose a composite index of well-being based on eight social dimensions, each including indicators for social relationships, emotions, health, work, material well-being, civil, and political liberties, personal security, and environment quality. The index is applied to 43 countries using the Borda rule and the principal components analysis method

Several other studies (Ivanova et al., 1999; Ogwang & Abdou, 2003; Qizilbash, 2004) point out the difficulties and risks of creating indices, in view of the multidimensional aspects of wellbeing, the redundancy of variables and the measurement sensitivity to any weighting system. McGillivray (1991) and McGillivray and White (1993) highlight the redundancy between the HDI and its components and say the HDI is ‘‘yet another redundant index’’ and that its significantly high correlation with GDP per capita demolishes the argument that it would produce different rankings of countries. To overcome this drawback, McGillivray (2005) has recently applied principal components analysis and regression estimates to the HDI components to extract an aggregate measure of non-economic aspects of human well-being.

Cahill (2005) analyses the implications of such correlations for the choice of weighting. The main conclusion is that high correlations between HDI components would produce a composite index insensitive to the weighting.

In the second approach, the HDI’s reductionist nature is also criticized but bigger questions are raised, such as the relationship between the capability approach and the concept of human development (Gasper, 2002), or broader ones, about the content and empirical measurement of ‘‘capabilities.’’ Most studies supporting the capability approach use disaggregated data from household surveys and very few use aggregated data to allow international comparisons as our indices permit. Slottje (1991) uses 20 indicators to build a composite index of well-being for 126 countries. Baliamoune (2003) explicitly uses Sen’s capability approach and proposes classifying countries according to new indicators close to the concept of freedom
conveyed by ‘‘capability.

(b) Alternative applications of the capability approach
Studies using Sen’s approach have flaws. They do not measure ‘‘capability.’’ With limited data available, usually only ‘‘functionings’’ carried out are used as a proxy of ‘‘capabilities.’’ These attempts are also sometimes far from the conceptual framework they are supposed to be linked to because the composite
indices rely on a combination of indicators that are different by nature, some corresponding to ‘‘capabilities’’ (civil liberties and political rights), others to ‘‘functionings’’ (literacy) and others still to resources or assets (such as the number of telephones per capita).

Well-being, standard of living and quality of life are generally not differentiated in these studies and they apply different concepts and realities, raising two questions—which and how many indicators to use to move towards the concept of human development, and the matter of new and more complete indices to make the distinction between the various concepts of Sen’s approach.

Sen gives no list of ‘‘capabilities’’ to take into account for constructing well-being indices and allows multiple proposals (Alkire, 2002). The three HDI components could be justified conceptually as being universal, basic to life and measurable but they raise the non-inclusion of other dimensions. For example, Dasgupta
(1990, 1992) criticizes the HDI for neglecting human rights. The choice of indicators suggests that if ‘‘capabilities’’ were carried out in these three basic dimensions, it would be done in the other dimensions of human development. With the capability approach, while the education and life expectancy indicators refer to
‘‘functionings,’’ the income per capita component seems to be a ‘‘commodity.’’ As the human development concept has emerged from the GDP limitations, it seems inappropriate to include an income component
in an index of well-being. As Anand and Sen (2000) themselves argue, income can be an indirect indicator of some capabilities. However, well-being is not determined by possession of resources but by their transformation into ‘‘functionings’’ which depends on personal, social and environmental factors. GDP per capita is necessary but not sufficient for human development as shown by countries where high and growing GDP per capita has not led to enrichment of human lives (Sen, 1999). The income component affects the purity of the HDI as a capability-based measure. First, the level of GDP per capita is a poor indicator of the
means of a group of people and its usefulness for the expansion of social services and infrastructure development is not clear (Anand & Ravallion, 1993; Anand & Sen, 2000; Sen, 1981, 1999). Means indicators that are determinants of well-being and part of the standard of living index are needed. Second, GDP per capita is a bad proxy of freedoms and quality of  life.

Like our indices, the PQLI and the Capability Poverty Measure (CPM), used in the 1996 Human Development Report (UNDP, 1996) and replaced the following year by the HPI (with a variant for developing and industrialized countries), are examples of non-income measures. The HPI replaces the income component by several means indicators, while the CPM focuses on human capabilities, functionings and outcomes. Unlike the HDI, these indices are headcount proxies measuring the percentage of people in each country who lack basic human capabilities.

(c) Justification of Standard of Living and Quality of Life Indices
Standard of Living (SL) and Quality of Life (QL) are not new indices of well-being, but past indices differ from ours in that they capture different dimensions of well-being, use different methods of aggregating and have different theoretical foundations.

To take other aspects of development into account, while distinguishing between the concepts, we define the composite indices of SL and QL as based on ‘‘commodities’’ on the one hand and the ‘‘functionings’’ and ‘‘capabilities’’ on the other. SL corresponds to the quantity of goods and services and to the services the GDP produces. It includes several means indicators that correspond to ‘‘commodities’’ that could be called inputs. QL includes (unlike SL) more intangible or qualitative aspects such as quality of education, extent of child labor and quality of the environment. It is a combinnation of ‘‘functionings’’ and/or ‘‘capabilities’’
indicators within the meaning of freedoms. They are result indicators that refer to output within a transformation system of ‘‘commodities,’’ as Sen suggests.

SL involves nine indicators in three domains (education, health, and material well-being).

• Public expenditure as a percentage of GDP in education and health take into account the money allocated to these social services. 
• The number of doctors/physicians is an indicator of health facilities. 
• Access to safe water reflects public facilities available and a means to prevent illness and epidemics. 
• The age dependency ratio and net primary school enrollment indicators are more difficult to justify as means indicators and have been chosen on the basis of data availability. 
• The teacher/student ratio would have been a better means indicator but was not available for many countries. The age dependency ratio is the demographic pressure. It is a summary of ageing that includes both young and old in relation to the potentially active population, so is a rough indicator of education.

QL focuses on measures of well-being or human outcomes that include explicit references
to human freedoms. It tries to take into account the capability of participating in community life that was dropped from the HDI. The main indicators emphasize the ‘‘being and doings’’ of a population, their opportunities as well as their non-opportunities. Capability failure can stem from violation of personal
rights or absence of positive freedoms (Sen, 1992).

The QL composite index is a combination of nine indicators covering three domains: health, education, and environment in a broad sense.

• Education quality includes capabilities that education can provide. Adult literacy is accompanied by two other indicators measuring children’s and women’s capabilities. 
• Child labor indicates denied opportunities to acquire human capabilities needed in productive and social life. 
• Female labor captures the intensity of gender equality in productive activity. It may also depict to a lesser extent ‘‘outcomes’’ of education as capability of entering the labor market. 
• Life expectancy is an indicator of capacity to live a long life, maternal mortality accounts for capacity to give birth in good health conditions and the percentage of underweight or stunted children accounts for the potential for being well nourished. 
• Civil rights and political freedoms refer to the aim of development relating to the actual freedom enjoyed by the population involved. 
• Trade openness involves the capability to exchange goods with partners and contribute to better quality products. 
• Carbon dioxide (CO2 ) is not directly related to air pollution but to resource depletion. However, all fuels produce CO2 when burned but sometimes also other pollutants (nitrogen oxides, gaseous hydrocarbons, and carbon monoxide). The more CO2 emissions, the more likely are other dangerous emissions and worse quality of air. 

The choice of indicators in Table 1 is justified by arguments above but is also limited by availability of data to compile SL and QL indices for at least as many countries as there is data for the HDI and GDP per capita (we have 170 countries).

Division of indicators between SL and QL is not always easy to establish.

The concept of ‘‘capability’’ is difficult to discern from country data because it is initially de- fined in reference to individuals and their relationships with other members of society.

Several lists of ‘‘basic capabilities’’ are proposed in the literature (Alkire, 2002) to approach the concept of human development but their use in identifying indicators is mostly limited to the disaggregated household data. Several indicators can be interpreted at the same time as a ‘‘commodity,’’ a ‘‘functioning’’ or a ‘‘capability.’’ So it is hard to produce a division of indicators that is not contestable.

Table 1: Standard of Living Indicators for Analysis of African Countries
Standard of health

• Public health expenditure (% of GDP)
• Improved water source (% of population with access)
• Physicians (per 1,000 people)

Standard of education

• Age dependency ratio (dependents to working-age population)
• Public spending on education, total (% of GDP)
• Net primary enrolment (%)

Material well-being

• Vehicles (per 1,000 people)
• Roads paved (% of total roads)
• Television sets (per 1,000 people)

Quality of Life Indicators for Analysis of African Countries

Quality of health

• Under-weight or under-height children under age five (%)
• Life expectancy at birth (years)
• Maternal mortality reported (per 100,000 live births)

Quality of education

• Literacy rate, adult total (% of people aged 15 and above)
• Labor force, children 10–14 (% of age group)
• Labor force, female (% of total labor force)

Quality of environment

• Openness (trade, % of GDP)
• CO2 emissions (metric tons per capita)
• Political rights and civil liberties (index)

..................................