A Review of the Literature on Well-Being in Italy: A Human Development Perspective

Abstract

In recent years, a large literature on indicators of well-being and quality of life has emerged. While all these indicators are an important step toward the recognition of well-being as a multidimensional phenomenon, they are often rooted in very different approaches—when we can identify a relevant “theoretical” framework—such as basic needs, happiness, or capability approach, and vary significantly in terms of statistical quality. This paper has a twofold objective: (1) to analyze the state of the art of the literature on well-being in Italy; (2) to examine this literature from a human development perspective. Thus, we investigate the pros and cons of the existing approaches/indicators and assess whether they are rooted in the human development approach. This is deemed necessary for the final goal of constructing a context-based indicator of human development for Italy and its territorial units.

Keywords:

• well-being
• human development
• indicators
• territorial disparities
• Italy

JEL classifications:

• O150
• R580
• D630
• C430

1. Introduction

Over the last decade, we have witnessed astonishing developments worldwide in the debate on the meaning of well-being and quality of life, and on how to measure these phenomena. While in the past debate was primarily focused on developing countries and relegated to the academic world, in recent years its focus has moved toward high-income countries and involved national and international institutions. Many statistical offices, as well as NGOs, think tanks and research centers have proposed new indicators that overcome the traditional economistic view of well-being.

While most of these new initiatives depart from the view of gross domestic product (GDP) as an adequate measure of well-being, it is important not to put them all in the same category. The theoretical approach on which indicators are founded is often different, as well as the objective and the statistical methodology. This substantially affects policies: whether we focus on happiness as a measure of quality of life or Amartya Sen's notion of capabilities makes a substantial difference to what type of objectives are pursued and which tools are used.

This paper has a twofold objective: (1) to analyze the state of the art of the literature on well-being and quality of life in Italy; (2) to examine this literature from the perspective of the human development and capability approach (HDCA). In our view, the HDCA is the most suitable conceptual framework for measuring and analyzing well-being in high-income as well as low-income countries as it focuses on people's achievements in different life domains, rather on instrumental factors such as the income or commodities people possess (see Section 4). We perceive these as essential steps toward the construction of a rigorous context-based indicator of well-being rooted in the human development approach which can help to portray life conditions in the Italian territorial units.

In order to conduct this review, we need to compare different accounts of quality of life measurement. In particular, we need to compare measures of well-being, quality of life and human development. Can indicators that refer to different concepts be compared? We think so. We argue that, for example, well-being and quality of life are almost analogous concepts that are both multidimensional and focus on life conditions of people in given geographical areas.

While in the past Smith (1973) used the term well-being to refer to objective life conditions of a population and quality of life to refer to people's subjective assessments of their lives, the following literature has contributed to a substantial convergence (Langlois & Anderson, 2002). Sen (1985) argues that “the quality of life a person enjoys is not merely a matter of what he or she achieves, but also of what options the person has had the opportunity to choose from” (pp. 69–70) whereas well-being is simply made of people's achievements (i.e., functionings). This shows that there is a difference between the two concepts: however, when it comes to measuring them, the choices are exactly the same, i.e. relying on achievements given the lack of information on people's opportunity set.

Similarly, the human development index (HDI), elaborated by the United Nations Development Programme (UNDP, 1990), is not a real indicator of development, as a process of “enlargement of people's choices,” but rather an indicator of quality of life and a good proxy indicator of well-being.

The paper is structured as follows. In the next section, we review the international literature on measures of well-being. In Section 3, we provide an extensive review of the most important indicators used to measure well-being in Italy. In Section 4, we critically re-assess this literature, identifying the pros and cons of each indicator and verifying whether they are consistent with the human development approach, and finally, Section 5 includes our concluding remarks.

2. Review of the International Literature

For several decades, GDP has been considered by far the most relevant measure of well-being and development and it is still the most widespread. It is the result of an “economistic” view of both well-being and development, and not just of the strictly neoclassical economic theories. International comparisons, therefore, were done on the basis of GDP [or, the gross national income (GNI)]: consequently, economic growth was the single objective of economic policy to enhance the well-being levels in a country.

Kuznets (1934) is considered the father of the modern national income accounting systems, being the first one who proposed it for the USA. We should, however, stress that Kuznets himself considered GDP only a rough measure of the monetary flow of goods and services produced by a country within a given time span, not an indicator of well-being. The main interest of the Nobel Prize economist was to measure the levels of industrial and agricultural production and to understand how much of the national income was due to consumption and investment. As Kuznets (1934) argues, “The welfare of a nation can scarcely be inferred from a measurement of national income” (p. 7). This crucial point was soon forgotten, and since that moment on economists, policy-makers and governments have often used the GDP to measure also what it was not supposed to measure.

Over time, some streams of thought have highlighted the drawbacks of GDP as an indicator of quality of life. In the field of development, a particular contribution came from theoretical paradigms such as the basic needs (International Labour Organization, 1976; Stewart, 1985; Streeten, Burki, ul Haq, Hicks, & Stewart, 1981) and the human development approach (UNDP, 1990), and from the early 1960s work on social development indicators of the United Nations Research Institute for Social Development (UNRISD). In the economic field, we need to mention heterodox approaches such as the capability approach proposed by Sen (1984, 1985) and the paradigm of economic development, which, under different facets, looks at objectives of economic policies that go beyond GDP such as economic and institutional transformation and the linkages between GDP and inequality, poverty and unemployment (Meier & Seers, 1984; Myrdal, 1973). The Scandinavian school on the quality of life, developed during the 1970s and 1980s, has significantly influenced the debate on the real goals of social and economic policies in these countries, ensuring a gradual shift toward noneconomic aspects of life (Morris, 1976). Last but not least, we must acknowledge the contribution of the studies in environmental economics, which recognized the potential trade-offs between economic growth and environmental conditions as well as the role of natural resources (Daly & Cobb, 1989; Nordhaus & Tobin, 1972).

At the risk of oversimplifying the debate, we can identify four main critiques addressed to GDP:

• It uses a money metric to define the weights of goods and services. As argued by ul Haq (2003, p. 103), are we sure that the value of a gun is several hundred times more than a bottle of milk?

• As a direct consequence of point (1), GDP does not consider the commodities without a market value such as the care work, the domestic work, the environmental services, and often education.

• It is an aggregate measure, obtained using data on the production of goods and services, thus not able to indicate the real-life conditions of the population. This type of critic, therefore, concerns the way the indicator is constructed—which does not take into account income distribution—and is not a general critic to the exclusive use of economic variables and monetary parameters in the evaluation of people's quality of life.

• It does not account for human diversity. Personal characteristics (e.g., gender, age, and health status), environmental characteristics (e.g., climate), as well as social/institutional characteristics (e.g., law, social norms) determine the “conversion” of resources into well-being (Sen, 1985).

As a consequence of the above critiques, national and international institutions, research centers, and various researchers have proposed a series of well-being indicators that differ, in some cases more than others, from GDP. We can divide them into three categories: (1) those adjusting GDP; (2) those integrating GDP; and (3) those replacing GDP.

2.1. Adjusted GDP Indicators

These are indicators that take the standard GDP and correct it in order to reflect people's well-being. This literature has proliferated after the seminal works of Nordhaus and Tobin (1972) and Daly and Cobb (1989) who have proposed different measures of “economic welfare.” The most recent and advanced indicators are the Index of Sustainable Economic Welfare (ISEW) (Jackson & Marks, 2002; Jackson, McBride, Marks, & Abdallah, 2007) and the Genuine Progress Indicator (Hamilton, 1999).

Although they differ in some methodological aspects, both indices detract social and environmental costs and add social and environmental benefits to the GDP. They are constructed by taking the personal consumption expenditures, correcting it for income inequality, adding public expenditures for sectors such as education and health, the value of domestic labor and volunteering, and other economic benefits, and finally subtracting “defensive” private expenditures, costs of environmental degradation, and depreciation of natural capital. The construction of these indicators is done by assigning a monetary value to complex social and environmental benefits and costs. The ISEW has been computed for the USA, Thailand, Chile, and many European countries. Recently, a Regional ISEW was calculated for all English regions (Jackson et al., 2007).

2.2. Indicators Integrating GDP

All the indicators that include both economic and social elements belong to this group. The most famous example is the HDI (UNDP, 1990, 2010). It combines three “functionings”: being knowledgeable, having a long and healthy life, and having a decent standard of living. In the new version of the indicator, proposed in 2010, the three dimensions are measured by the following indicators: (1) a geometric mean of the mean years of schooling and the expected years of schooling; (2) life expectancy at birth; (3) purchasing-power-parity adjusted per capita GNI. First, the variables are standardized and then aggregated through a geometric mean.

In theory, the human development approach (see Section 4), which provides the theoretical foundation to this index, would require the use of variables reflecting the development goals and not its means. As GDP and national income are means to expand education, health, nutrition, and other development goals, the HDI should replace GDP rather than integrating it. However, the main reason for this choice lies with data availability (Sen, 2000) for the purpose of comparing all countries adhering to the United Nations system. Given that the paradigm of growth has been the dominant one for decades, governments and statistical offices have mainly collected economic data: as a direct consequence, GDP was selected as a proxy for all the other relevant functionings not appearing in the indicator (Anand & Sen, 2000). To be honest, according to Haq (2003, p. 104), the founder of the Human Development Division within the UNDP, “The merging of economic and social indicators is one of the distinctive features and chief strengths of the HDI.”

2.3. Indicators Replacing GDP

Other indicators go even further “beyond GDP,” identifying other dimensions and variables to portray the levels of well-being. Most of them are multidimensional indicators and, in some cases, composite indicators.1

After 10 years of work on the measurement of progress, the OECD has recently elaborated the Better Life Index (BLI). It is a tool used to compare the quality of life in 34 countries through a large set of indicators for 11 domains in the areas of material living conditions and quality of life: income, community, education, environment, civic engagement, health, housing, jobs, life satisfaction, safety, and work–life balance. Whereas in the “How is life?” report the OECD uses equal weight for all the dimensions, each user can utilize some interactive tools available in the website in order to construct a different BLI, according to personal preferences given to each of the 11 topics that make for a better life.

An important example for high-income countries is the Canadian Index of Wellbeing (CIW). This initiative of the Atkinson Charitable Foundation, started in 2004 and led, after a long process of discussion among many experts in different fields over the well-being dimensions, to the construction of the CIW (Institute of Wellbeing, 2009). The latest version of the index incorporates 64 indicators pertaining to eight dimensions: living standards, healthy populations, community vitality, time use, education, environment, leisure and culture, and democratic engagement (Institute of Wellbeing, 2011). After a long period of consultation, it was decided to aggregate the eight-dimensional indices through a simple arithmetic mean. Both the BLI and CIW do not contain GDP; however, income still plays an important role because, in the former, there is disposable income and wealth whereas, in the latter, median disposable income and income distribution.

Within the sociological literature and the broader field of “happiness,” some composite indicators of quality of life or development have been proposed. Within this school of thought, happiness, usually measured by life satisfaction, is considered a measure of quality of life: the happier residents are, the higher the quality of life in that place. The spread of this literature has been facilitated by the existence of large data-sets which allow cross-country comparisons. Most of these data are included in the World Database of Happiness.

Among the indicators proposed in this literature, an important role is played by the Happy Life Expectancy Index (Veenhoven, 1996). This indicator is constructed by taking life expectancy at birth and adjusting it for a variable—ranging from zero to one—reflecting people's average life satisfactions. The characteristic of this index is that it combines information present in demographic statistics (life expectancy) with information obtained through ad hoc surveys (life satisfaction). This way, it measures the degree to which citizens live long and happily.

3. Measuring Well-Being in Italy

In recent years, there has been an increasing proliferation of initiatives focusing on the concept of well-being and quality of life in Italy too. Whereas until the last decade, interest in these issues was confined to the research and academic world, today public institutions and local authorities, as well as civil society, are launching a series of initiatives to promote a shared measurement of well-being.

This section includes a brief overview of the most significant contributions to the measurement of well-being and quality of life carried out in Italy from academics who have focused on the correction of the HDI to the most famous rankings on the quality of life yearly published, together with the recent contribution of the Italian National Statistical Institute to the measurement of an Equitable and Sustainable Well-being. The most relevant information about these indicators is reported in Table 1.

Table 1 Quality of Life/Well-Being Indicators Elaborated in Italy

Indicator	Authors and year	Phenomenon	Dimensions	Composite index?	Territorial unit
Indicators elaborated in the academia
HDI	Conte et al. (2007)	Human development	Education, life expectancy, and income	Composite indicator	Country; macro areas
HDI	Costantini and Monni (2008)	Human development	Education, life expectancy, and income	Composite indicator	Regions
HDIM	Costantini and Monni (2008)	Human development	Education, income, and employment	Composite indicator	Regions
Regional HPI-2	Costantini and Monni (2008)	Human poverty	Life expectancy, decent standard of living, and social exclusion	Composite indicator	Regions
HDIM-M and HDIM-W	Costantini and Monni (2008)	Gender-specific human development	Income, education, and health	Composite indicator	Regions
Adjusted HDI	Monni (2002)	Human development	Education, job market, and income	Composite indicator	Provinces
Quality of life	Colombo et al. (2012)	Quality of life	Climate, environment, services, sociodemographic features, and economic features	Composite indicator	Provinces
Life satisfaction	D'Andrea (1998)	Life satisfaction	Income, health, family, friendships, leisure, and work	Composite indicator	Country
Indicators elaborated by the media and NGOs
Quality of life	Il Sole 24 Ore (yearly elaboration)	Quality of life (livability)	Standard of living, business and labor market, services, environment and health, population and public order, and leisure	Composite indicator	Provinces
Quality of life	Italia Oggi (yearly elaboration)	Quality of life (livability)	Standard of living, business and labor market, crime, social unrest, financial services, health, population, and leisure	Composite indicators	Provinces
Regional Quality of Development Index—QUARS	Sbilanciamoci! (yearly elaboration)	Quality of life	Environment, economy and employment, health, education and culture, rights and citizenship, equal opportunities, and participation	Composite indicator	Regions
Indicators elaborated by public institutes
Equitable and Sustainable Well-Being (BES)	ISTAT and CNEL (2013)	Well-being	Environment, health, wealth, education and training, reconciliation of work and life, social relations, security, subjective well-being, landscape and cultural heritage, research and innovations, quality of services, and policy and institutions	Suite of indicators	Regions

A first group of studies has compared well-being levels in Italy with those of other industrialized countries by using the UNDP HDI (Conte, Della Torre, and Vasta, 2007; Costantini & Monni, 2008). Conte et al. (2007) analyzed the trends in the HDI for 17 developed countries for more than a century (1870–1990). Italy's considerable progress in life expectancy at birth was not accompanied by an equal progress in education (measured by literacy and participation in primary, secondary, and tertiary education). A similar analysis has been conducted for four macro geographical areas of Italy, revealing, among other things, that the North-East and the Centre have much better results in converting economic growth into the well-being of the population than the North-West (Conte et al., 2007).

Costantini and Monni (2008) have examined the historical trend in human development in Italy too, with particular reference to gender disparities. In order to assess regional disparities both with and without gender disparities, they also employed a modified HDI (HDIM) for the Italian regions, which includes for the education dimension only the tertiary gross enrolment ratio, and substitutes the life expectancy index with the employment rate. A second set of studies focuses on HDIs adjusted for the Italian case (Monni, 2002; Passacantilli, 2003). Monni (2002), for example, has elaborated a specific HDI for the Italian provinces. Given the huge intra-regional differences from many perspectives, this has led to a focus on a lower territorial scale, the provinces. At this territorial level, official data of GDP are not produced so the author used an estimated value adjusted with Atkinson (1970) formulation. Moreover, this index adjusts the official HDI in order to reflect the socioeconomic specificities of the territories: the knowledge dimension is measured by the participation in higher education and university education whereas the health component has been replaced by a labor market indicator (i.e., the inverse of the unemployment rate). Passacantilli (2003) has proposed an HDIM to assess the quality of life in the 20 municipalities in the city of Rome. This index has been recently used by De Muro, Monni, and Tridico (2011) to evaluate the dynamics that occur in the center and on the edge of the city of Rome, and to investigate the relationship with socioeconomic policies implemented over past years.

Other studies propose different methodologies for the measurement of well-being. Colombo, Michelangeli, and Stanca (2012) apply the “hedonic price” method to the evaluation of quality of life in 103 Italian provinces. They consider five domains and their relative impact on well-being is estimated by the market prices of housing, as an expression of willingness of individuals to pay to live in an environment with certain characteristics. The results are expressed in terms of prices. The quality of services and the economic conditions are the domains that affect the final ranking most: the price effect strongly influences the definition of the model and the results that emerge from it.

On a yearly basis, Il Sole 24 Ore and Italia Oggi publish the two most famous surveys on the quality of life in Italy. The one carried out by Il Sole 24 Ore has been measuring the quality of life in 107 Italian provinces since 1988 through a set of statistical indicators. In its dossier, the quality of life in the provinces is measured through six domains considered equally relevant: standard of living, business and labor market, services, environment and health, population, public order, and leisure. Each of these domains is measured by more than one indicator.

Alongside the Il Sole 24 Ore, every year Italia Oggi, in partnership with the University of Rome “Sapienza,” elaborates a ranking of the quality of life in 103 Italian provinces. The composite index includes nine dimensions: the variables used to measure the different dimensions are similar to those used in the survey made by Il Sole 24 Ore.

A very important contribution to the measurement of well-being in Italy is the one promoted by the Italian campaign “Sbilanciamoci!” that since more than 10 years has computed an index of quality of life for the Italian regions—the QUARS (Regional Quality of Development Index)—through a process of consultation with 46 civil society organizations. The final aim is to develop a shared definition of qualitative development, which may grant legitimacy, stimulate an open, public discussion, and affect political decisions (Segre, Rondinella, & Mascherini, 2011). The consultation process has led to the definition of a framework for the quality of life, consisting of seven dimensions, all considered equally important. Various indicators were selected in each dimension through a deliberative approach, driven by the desire to reward those elements of well-being that can be directly obtained through the implementation of public policies at the various levels of government. The QUARS is a synthetic index obtained by aggregating the seven-dimensional indices through a simple mean.

A recent important contribution in the field comes from the Italian National Institute of Statistics (ISTAT), which follows the experience of an international organization such as the OECD and some institutes of statistics in Europe. Among them, in the UK, the Office for National Statistics launched the program “Measuring National Well-being,” with the aim of producing a shared set of indicators to measure the country's well-being, at the end of 2010.

In Italy, in the past 3 years, ISTAT has shown significant interest in the themes of well-being in various ways. The first is the inclusion of specific questions on well-being dimensions and life satisfaction in the traditional Multipurpose Survey system. A clear signal of a new direction in the work of the Institute, however, came in 2010 with the launch, jointly with the National Council for Economy and Labor (CNEL), of a national research program aimed at creating a “Steering Committee on measuring the progress of Italian society.” The final objective of this initiative is to build a dashboard of indicators of Equitable and Sustainable Well-being (BES).

At the centre of this project, there is a conceptualization of well-being as a phenomenon composed of two key attributes: equity, within and between generations; and sustainability, from an environmental, economic, and social point of view. In the wake of the influence of the famous Stiglitz Commission (Stiglitz, Sen, & Fitoussi, 2009), this initiative stands out for its ambitious aim of reaching a shared definition of the concept of well-being which is the result of a deliberative process involving all the most important stakeholders (Giovannini & Rondinella, 2011). The premise behind this approach is that the concept of well-being is closely linked to time and space and can therefore usefully be defined only through a process of democratic consultation with all sectors of society and, in particular, with civil society. To achieve this objective, CNEL and ISTAT have formed a Steering Committee in which the social partners and civil society work together to achieve a common understanding of the phenomenon. The objectives of the working group include the development of high-quality statistical indicators for what is considered relevant by the country, and the communication and dissemination of their results over time.

The work of the Steering Committee has given rise to 12 dimensions that are at the basis of the framework of BES: among the selected dimensions, 8 represent the outcomes for individuals whereas 4 are the drivers for well-being (CNEL and ISTAT, 2011). In the selection process, citizens have also been involved through the publication of a blog and an online questionnaire in which they can express their preferences for the identified dimensions.2 Then, the Scientific Commission for the measurement of well-being selected 134 high-quality statistical indicators considered as appropriate in order to measure the 12 domains identified by the Committee.3 At present, the first phase of the project has concluded with the publication of the final report (CNEL and ISTAT, 2013)4 that provides a multidimensional picture of quality of life in Italy and its regions over time.

Finally, other studies focus on happiness, expressed by the degree of individual satisfaction, as the main thermometer of quality of life (D'Andrea, 1998; Rampichini & D'Andrea, 1997). D'Andrea found an increase in happiness in Italy during the period 1975–1994; in his work with Rampichini he discovered that, in spite of strong inter-regional differentiation, an increase in the level of satisfaction from 1973 to 1991 had occurred. D'Andrea, in particular, took into account the satisfaction of individuals over the age of 14 years, compared with satisfaction expressed toward work and life in general: the domains that the population in analysis was called upon to evaluate, on a scale of four modes, were work, income, health, family, and affections and pleasure.

4. A Review Through the Lens of Human Development

In this paper, we endorse the HDCA as a conceptual framework for measuring and analyzing well-being in Italy. Therefore, in this section, we reassess the status of the existing literature on the measurement of well-being/quality of life in the country from a human development perspective. In particular, we discuss the pros and cons of each indicator and, above all, discuss whether the indicators are theoretically rooted in the HDCA. This implies investigating the dimensions and the variables used to create the indicators. We do not investigate the technicalities of the indicators such as the variables standardization procedure and whether and eventually how to aggregate dimensions in composite indices because they reflect the specific researchers' objectives such as building only rankings versus obtaining values to be compared across time, or having a powerful political tool (the one single number of composite indicators) versus having an immediate focus on separate dimensions.

The capability approach has been pioneered by the Nobel prize economist Amartya Sen since the 1980s as a new approach to development, well-being, inequality, poverty, and justice (Burchi & Passacantilli, 2013; Rippin, 2012; Sen, 1995, 1999). This approach centers on two core concepts, functionings and capabilities. Functionings are the set of things people are and do such as being a literate, being adequately nourished, and being in a good state of health. People's capabilities, instead, are all their potential functionings, i.e., what they can be and do in their life (Sen, 1985, 1995, 1999). This approach focuses on people's life conditions in terms of capabilities and functionings rather than on their income or commodities: the latter are one of the means to enhance people's capabilities. Well-being is a multidimensional phenomenon consisting of several functionings.

Similarly, the UNDP elaborated the human development approach as a process aimed at expanding “people's choices” (UNDP, 1990). The HDA is rooted in the capability approach which is a broader framework for analysis of different socioeconomic phenomena. Since the first Human Development Report, issued in 1990, this international organization has published country data (also for Italy) on the HDI, a multidimensional indicator of development. However, there is a substantial difference between a standard economic indicator such as GDP growth and the HDI: the first is calculated as a percentage variation of a variable (GDP or GNI) whereas the second indicates the performance of a country in three dimensions at a given moment. Therefore, it indicates the human development levels and not a dynamic process, which is very difficult to capture in an index. As highlighted in Section 1, the HDI is an indicator of quality of life and a good proxy for well-being (UNDP, 1990).

In the previous sections, we have mentioned some problems related to the official HDI. The most important is that it is incoherent with the conceptual framework it derives from since it includes GNI (before 2010, it was GDP). Another critique—which does not concern the technicalities of the statistical methodology such as the standardization of the variables and the aggregation methods that fall outside the scope of this paper—concerns the fact that it includes only three dimensions. However, the ultimate goal of the indicator was to be of direct use to policy-makers, politicians, and development practitioners. It was therefore supposed to be relatively simple, with few dimensions (Haq, 2003; Jahan, 2003). This explains why the UNDP decided not to add new dimensions such as political freedom or social relations which have been extensively discussed in the Human Development Reports since 1990.

The problems analyzed above are not really relevant to Italy because more and better data are available and because the objective of scholars and organizations is usually to look at historical trends in different geographical areas of the country or, more often, to compare the well-being conditions of regions or provinces. For example, Monni and Costantini use the official (pre-2010) HDI as well as all the other UNDP indicators in order to compare well-being levels in Italian regions. While this work provides a comprehensive picture of the human development conditions by using the dimensions and variables adopted by the UNDP, it does not seem able to capture the differences existing in a developed country. As argued by Anand and Sen, the concept of human development incorporated in the official HDI is “concerned only with the enhancement of the very basic capabilities of people” (Anand & Sen, 2003, p. 122). Why use life expectancy as an indicator of health? It can be suitable for developing countries whereas, in high-income countries, life expectancy heavily depends also on other factors not related to health (car accidents, medical and technological progress, etc.). It could be replaced by healthy life expectancy at birth5—as suggested by the World Health Organization and the European Union—or by morbidity rates for some relevant diseases.

In another article, Monni (2002) offers one of the few analyses of human development for Italian provinces. In fact, while official and administrative data sources offer a plenty of information on several well-being dimensions at regional level, for lower territorial scale data are still scarce. Another strength of the paper is an in-depth discussion of the need to contextualize the indicators in order to capture territorial disparities: this is especially reflected in the selected indicator of knowledge. The composite index, however, suffers from three methodological problems: (1) the employment conditions replace the “health” component: while having a “decent job” is without doubt an important functioning, it is not clear what its relationship with health is; (2) it incorporates the economic dimension (measured by the estimated provincial GDP); (3) because the (reverse of) the unemployment rate is traditionally highly correlated with GDP, some problems of multicollinearity/double counting may appear when they are combined in one composite index.6 For example, the Pearson's correlation coefficient between these two variables in Italian provinces in 2009 is 0.86.

In their very interesting analysis of the trends in well-being in Italian regions from 1871 to 1990s, Conte et al. (2007) use the traditional HDI since this is able to highlight the different development pattern in such a long time span. When the authors conclude that catching-up of the South with the other regions is much clearer with the HDI than with GDP data, they highlight some problems with the official UNDP HDI: the combined gross enrolment ratio (one of the two variables used to measure “knowledge”) is a problematic variable because it looks at gross rather than net enrolment ratio.7 The authors also argue that the knowledge dimension does not properly reflect the “efficiency” of an education system; however, we do not think this is an appropriate critique because, with a functionings-based well-being index, we need information on how people are with regard to knowledge rather than information on the education system. The main critique, in our opinion, is that the knowledge dimension focuses only on very basic functionings such as “being literate” rather than pointing to higher skills/abilities, reflected, for example, in the percentage of population with a university degree. Using a human development indicator more suitable for high-income countries—still possible with poor data availability for the whole historical period considered—would have contributed more to capturing the disparities between the North/Center of the country and the South.

Let us now focus our attention on indicators of quality of life that are not explicitly defined as HDIs. The work of Colombo et al. (2012) is still anchored to a traditional view of quality of life which is measured in terms of market prices. This is obviously not in line with a human development framework according to which basic functionings are intrinsically, and not only instrumentally, relevant.

Since 2003, the QUARS elaborated by Sbilanciamoci! has been an important tool for advocacy/political pressure as well as an important attempt to build an index from a civil society perspective. The figures they provide have surely contributed to shifting the attention of policy-makers and local administrators toward noneconomic dimensions of quality of life. Moreover, their proposal has probably been the first attempt to use civil society to identify the relevant indicators for each domain. However, these positive results are counterbalanced by a series of weaknesses.

The first one consists of a lack of an adequate conceptual framework for the choice of indicators. The definition of quality of life—the authors sustain that most of the other well-being indicators suffer from a poor definition of the phenomenon being measured (Segre et al., 2011, p. 51)—is “limited by data availability.” This shows that the authors follow a strictly empirical approach: indicators define the concept (latent variable) and because indicators depend on the data available, the concept itself depends on the available data. The fact that within the set of indicators there are “input,” “output,” and “outcome” indicators combined together is problematical because they reflect different means (or drivers) and ends (or constitutive elements) of well-being (Burchi & De Muro, in press; Nardo et al., 2008; OECD, 2011). To give an example, we find a series of input variables such as the number of women's counseling centers, cinemas, and libraries, which are policy variables and are coherent with an asset-based approach to quality of life (e.g., Brandolini, Magri, & Smeeding, 2012; Chambers, 1995),8 and other variables such as the proportion of population with a university degree, which could be a good indicator of human development for Italy. A theoretical approach, such as the human development approach, should contribute to defining the phenomenon and provide a guide to the choice of dimensions and variables.

A second problem, related to the first one, concerns the extensive number of variables, 41, used to generate the QUARS. With so many diverse variables ranging from input to outcome variables, there is a serious risk of not being able to understand the final value of the dimensional and aggregate indices.

Finally, we need to discuss the massive consultation process the organization Sblanciamoci! has started in order to select the variables. Without a general conceptual framework for the selection of variables, these consultations can end up with too many heterogeneous variables (see previous point). Moreover, 46 Italian civil society organizations were contacted, everyone with a different field of specialization. Looking at the final outcome, it seems that the variable selection was biased toward the area of work of the organization. In some cases, these variables can hardly be considered indicators of a well-being dimension. The index of biological agriculture is a typical example not only because it is an input indicator, but also because its connection with nutrition, health, or the environment is not straightforward,9 and because in many areas of the country agriculture is not or little practiced while the residents of these areas can still consume organic products bought elsewhere. Finally, this consultation process differs from Sen's (2004) idea of a context-based list of basic capabilities which should theoretically involve people and not organizations because people are less likely to be biased when it comes to deciding what constitutes their well-being. Furthermore, Sen (2004) refers to the list of dimensions (which in turn reflect values); here it is mainly the variables that are decided by the civil society.

The Sole 24 Ore index is even weaker from a theoretical point of view. There is no comprehensive definition of quality of life or a reflection on the relevant dimensions and indicators.10 One weakness of this indicator is that it goes only partially beyond GDP or economic parameters: indeed, two out of six macro dimensions concern the economic sphere. Another drawback is that the choice of dimensional indicators reflects a relative weight assigned to the different elements of the quality of life: for example, equal opportunities for women is measured only with regard to employment opportunities whereas the integration of migrants is measured only in terms of regularization of their presence.11 Moreover, for variables like the divorce rate, it is not clear whether they should have a positive (more freedom) or negative (social breakdown) sign. Finally, there is no explanation of the method used to identify the dimensions.

In the Italia Oggi index, as well as in Il Sole 24 Ore one, we encounter the problem of a massive presence of economic variables, ranging from individual income to variables related to the managerial system and to a large set of services available in the territory. As for the other dimensions, the problem of social exclusion receives a great deal of importance given the presence of networks against social exclusion, the suicide rate, and the youth crime rate. On the other hand, a dimension present in almost all indicators of human development and quality of life—education/knowledge—is missing: the only exception is the number of professors per 100 students, which is, however, only an input indicator.

With respect to the ongoing process leading to the construction of the BES, it is worth noting its participatory nature: not only associations and NGOs, but also citizens have been involved in expressing their preferences for the selected dimensions, contributing to creating an Italian definition of well-being (three or more domains that describe the well-being levels of their country). Moreover, it represents the engagement of official statistics toward the measurement of well-being and the creation of a dashboard of quality of life indicators that can be considered policy goals.

Nevertheless, the methodology lacks rigor in the selection of indicators. In fact, the dashboard includes 134 indicators, consisting of input, output, and outcome indicators, as well as subjective and objective ones.12 Although the leading institutions wanted to strike a balance between representation of the corresponding domain and the availability of data, this massive number of indicators can hardly help to portray the quality of life in Italian territorial units. Moreover, the BES includes dimensions (e.g., environment, research, and innovation) which are not necessarily constitutive elements of a concept of quality of life, but also drivers: for example, it is argued that “The environment in which people live affects heavily the well-being of citizens”; this marks a difference with a human development-related indicator of well-being.13 Besides, it is now impossible to evaluate the aspect of sustainability any further as this has not been analyzed in the first report: at present, it evaluates how life is in Italy, but not how this life will be in future.

Finally, the framework on which the BES is built is similar to that used for the OECD BLI (Hall, Giovannini, and Morrone, 2011), which puts the environmental and the human sphere at the same level and considers that human well-being is composed of individual and social well-being. This framework is clearly different from that offered by the human development approach which ultimately puts people at the centre of well-being and considers a person as a social agent. This is why the capability/human development approach is said to be characterized by ethical individualism—the individual is the main unit of analysis—but not methodological individualism—the individual is assumed to interact with other people, care about others, and thus consider them when taking decisions (Robeyns, 2008).

Another school of thought tends to consider life satisfaction as an indicator of well-being (Kahneman, 1999).14 The work of D'Andrea (1998), for example, goes in that direction: life satisfaction is a subjective indicator which is often highly correlated to objective socioeconomic indicators. Indicators related to demography, participation, and environmental sustainability are not considered constitutive elements of the concept of well-being, but only determinants of individual satisfaction. A person's health condition is important only as long as it does have an impact on her/his life satisfaction: here is the striking difference with the human development framework.

Following the HDCA, one main critique can be addressed to this measure: life satisfaction is only a state of the mind and people tend to adapt their preferences (and answers) to the context and conditions in which they live (Sen, 1985). Moreover, the exclusive use of subjective variables makes it difficult to extend results to populations because of problems in the aggregation of individual preferences. Life satisfaction can be one of the well-being dimensions (and is itself multidimensional because it is related to work, family, social relations, etc.).

In conclusion, it is clear that the whole “beyond GDP” movement has led to the proliferation of indicators of well-being and quality of life. However, the fact that more or less all these indicators were based on a strong critique to GDP does not mean that they all belong to the same cluster. The conceptual frameworks—an element needed to avoid “measuring without theory”—standing behind the indicators proposed in Italy in the last 10–15 years are often different. In this section, we argued that even indicators defined as HDIs can only be loosely linked to the HDCA framework.

5. Conclusions

In recent years, we have observed the emergence of a large amount of literature on new indicators of well-being and quality of life, at both international and national levels. These indicators all share the same critique applied to GDP—at least in the way it is currently measured worldwide—as a single measure of well-being. For this reason, we often hear about an overall “beyond GDP” movement. However, there is a serious risk of generalization if we put all these initiatives in the same cluster. Significant differences exist among the indicators proposed in terms of theoretical approach—when a consistent theoretical framework can actually be inferred—which affects the choice of dimensions and variables, in terms of statistical rigor and, often, in the objectives for which they have been elaborated.

In this paper, we have concentrated on the literature on well-being and quality of life in Italy. This review, on the one hand, highlights the existence of a large number of efforts to shift attention toward non-GDP elements of quality of life. This shows a general interest in the topic which is gradually involving national and local institutions in the country. On the other hand, it reveals a series of limits and weaknesses in these proposals in addition to a general low consistency with the human development approach which is endorsed in this paper.

We argue that many indicators seem to be the products of a strictly empirical approach, based on the selection of a large number of available indicators that intuitively seem to be connected with a broad concept of well-being or quality of life. A reflection on the relevant dimensions and the indicators to use—for example, whether they should be input, output, or outcome indicators—is often missing. Some of these proposals are implicitly or explicitly rooted in other approaches such as the basic needs and happiness approaches. Moreover, even those indicators that are specifically defined HDIs are based on a narrow view of human development as an expansion of very basic capabilities, which is suitable for low-income but not for high-income countries such as Italy. Generally speaking, we notice that researchers do not adequately exploit the increasing statistical information available in the national statistical offices.

In conclusion, this paper argues for a more rigorous approach to the identification of well-being indicators. There is an urgent need to elaborate context-based HDIs for Italy, which can reveal territorial differences and assist the work of policy-makers.

Acknowledgements

The authors thank the participants in the PRIN Workshop and in the EAPE conference, held in October 2012 in Modena and Krakow, respectively.

Additional information

Funding

This paper is an outcome of the National Research Project (PRIN) “Measuring Human Development and Capabilities in Italy: methodological and empirical issues” co-funded by the Italian Ministry of Education, University, and Research (MIUR).

Notes

¹ Multidimensional phenomena such as poverty and well-being can be measured with a set of dimensional indicators and/or with “composite” indicators, where the different dimensional indicators are finally aggregated. In this paper, we do not engage in the debate on whether it is better to have one single number or multiple numbers as this choice depends on the specific objectives of the researchers/institutions.

² The questionnaire can be found at http://www.misuredelbenessere.it/fileadmin/upload/docPdf/Questionario_per_il_sondaggio_on_line_sul_sito_del_BES.pdf (consulted: 14 July 2014).

³ The full list is available at http://www.misuredelbenessere.it/fileadmin/upload/docPdf/LISTA_INDICATORI_ENG.pdf (consulted: 14 July 2014).

⁴ The report is available (only in Italian) at http://www.istat.it/it/files/2013/03/bes_2013.pdf (consulted: 14 July, 2014).

⁵ A detailed description of this indicator can be found at http://www.who.int/whosis/indicators/2007HALE0/en/.

⁶ The authors of this paper do not follow a traditional “statistical” perspective on the basis of which two or more highly correlated variables can never be inserted in a multidimensional indicator. When these variables reflect different dimensions or aspects of people's lives, their simultaneous presence is not problematical. It can cause substantial bias in cases—like the one presented here—when the two variables indicate more or less the same (economic) dimension. The employment rate, in fact, is a variable traditionally used in standard economic analysis.

⁷ The gross enrollment ratio is the share of children of any age that are enrolled in, say, primary school, while the net enrollment ratio is the share of children of official primary school age that are enrolled in primary school. Where there are frequent grade repetitions, the gross enrolment can exceed 100%.

⁸ This is also referred to as livelihood approach in the context of developing countries.

⁹ In particular, the fact that in a given region there are more people producing biological food does not mean that there are more people consuming it.

¹⁰ We could use the famous expression “measuring without theory” (Koopman, 1947).

¹¹ The dimensions in which these phenomena are studied reflect only to a minimal degree of the whole phenomena. While this could be due to data constraints, the authors of the index do not explain it. Moreover, for example, equal opportunities could be analyzed with regard to education as gender-differentiated data on this domain are available.

¹² In the existing literature, there is no agreement on whether well-being and quality of life measures should include subjective indicators in addition to objective ones. For example, the Stiglitz report (Stiglitz et al., 2009) suggests to use both. However, the report does not provide an answer on how to integrate one with the other and considers subjective measures important mostly as predictors of people's behavior. How a person perceives her/his health status, for example, affects where she/he will use more or less the health services, whether she/he will take more or less medicines, and in turn this can affect current and future health status. Therefore, in our view, to measure—more than to assess—well-being dimensions such as health, nutrition, or education, we should rely mostly on objective measures, and use subjective ones only when the former are missing. Different considerations are made with regard to potential dimensions such as personal security or life satisfaction, where subjective indicators are clearly indispensable.

¹³ For a comparison of well-being levels in Italian regions calculated on the basis of BES dimensions and an alternative approach to select dimensions (the “Constitutional Approach”), see Burchi, De Muro, and Kollar (2014).

¹⁴ For a comprehensive review of this approach to quality of life, see Stiglitz et al. (2009).