3. An Outline of Existing Monitoring Systems

There have been attempts, starting in the 1970s, to overcome the drawbacks of GDP by means of additions and subtractions.

One of the early attempts to create a better assessment of welfare was the Measure of Economic Welfare (MEW, 1972), developed by William Nordhaus and James Tobin. It subtracted environmental damage and defensive expenditures and added the value of unpaid work, of the informal economy and of leisure time, to the basic GDP measure.

Subsequently, the Index of Sustainable Economic Welfare (ISEW 1989) modified GDP by adding services produced by unpaid household activities, capital formation and the public sector, while subtracting defensive expenditures, the costs of environmental degradation and depreciation of natural capital. This was revised as the Genuine Progress Indicator (GPI 1995), which also starts with GDP, adding unpaid household labour, the increase in capital stock and the balance of international trade, and subtracting defensive costs and the cost of deterioration of nature. It contains 26 items, covering economic, environmental and social aspects. The ISEW and GPI have been estimated for many countries, provinces and cities (Kubiszewski 2018), but do not appear to have been much used in practice.

More recently, a “spectrum of measures” has been envisaged. Future GDP brings in missing capitals such as natural or environmental capital, human capital and some intangible productive assets; Welfare Minus approximates net national disposable income by also incorporating transfers; and Welfare also takes distribution into account (Heys et al. 2019).

Another reaction to the perceived inadequacy of GDP as a measure of wellbeing has been to shift the primary focus to capabilities, emphasising the importance of diverse abilities and activities in pursuing happiness. It involves a balance between materialistic and non-materialistic factors (Sen 2001; Nussbaum 2011). This has been developed into a philosophy of human welfare and development. In terms of practical indicators, it has informed the construction of the Human Development Index (HDI 1990), a composite indicator with equally weighted contributions from per capita GNI,¹ life expectancy at birth and years of schooling, aggregated by calculating their geometric mean. A revised method of calculation was introduced in 2010. As life expectancy and education coverage have increased in low- and middle-income countries, they contribute less of the variance of the HDI, which has therefore come to depend more on GDP alone.

Adjustment for inequality has also been carried out. The most widely used method multiplies per capita GDP by (1 – G), where G is the Gini coefficient, a standard measure of inequality; (1 – G) is therefore a measure of equality (Sen 1976). An inequality-adjusted version of the HDI has been developed (UNDP 2019), and features in the UNDP’s annual Human Development Report. However, it does not appear to be widely used for practical policy purposes. Other proposed measures are the Atkinson Index, which generates an adjustment based on the degree of inequality aversion (Atkinson 1970); and “the Vast Majority Income” (VMI), which measures the per capita GDP for the 80% of the population that has the lowest income, thereby excluding the richest 20% (Shaikh and Ragab 2008).

More recently, Jones and Klenow (2016) have extended per capita GDP as a measure of consumption, by adjusting it for mortality, inequality and leisure (based on the annual number of hours worked), using an expected utility framework. The adjusted measure is apparently designed for use in economic analysis, rather than as a practical policy indicator. It gives a more positive assessment of western Europe (e.g. France) than GDP alone, relative to the US, because of the higher life expectancy, lower inequality and fewer hours worked. With most developing countries, the opposite is seen. Trends in growth rates are typically revised upwards, mainly because of improving mortality rates over time.

It is now widely agreed that multiple measures are required for monitoring the economy, society and the environment. There is, however, no consensus on exactly which should be used, and how they relate to each other, despite important work by the United Nations, the OECD and many others. Progress has been and is being made on complementary metrics, particularly those related to the environment.

A widespread tendency in the”beyond GDP” movement has been to combine many different types of measures in the same index. More than 900 of such composite indicators now exist (Hoekstra 2019), some designed to be compatible between different countries, and some country-specific. The aim is often explicitly to move beyond an economic view of progress—different from, and broader than, the aim of this book which is to evaluate the success of the economy in human terms.

A few have contributed useful ideas to the discussion. However, it is generally unclear how beneficial change could take place as a result of the information included in the index. In many cases, composite indicators appear to be based on the notion that if something important is monitored, this in itself will shift its value in the favoured direction—whereas in reality, an indicator can only have a beneficial effect if it leads to appropriate action, implying that indicators need to be designed with this in mind. For whatever reason, composite indicators have had little impact on practical policy. Here I outline only a few of the major examples.

A notable instance of the effort to try and find a better indicator of societal—not only economic—success is the Better Life Index (2011), developed by the OECD. It consists of eleven areas, including environment, income, housing and life satisfaction (OECD n.d.; Van de Ven 2019). It therefore covers respectively assets, output, outcomes and impact. It also includes civic engagement, which is outside the scope of a specifically economic index. The Better Life Index can be used to compare countries, allowing the user to specify differential weighting of the various components, e.g. prioritising housing, or the environment (OECD n.d.). It has also been adapted for use in non-OECD countries that have lower levels of per capita income (Boarini et al. 2014).

Some countries have produced their own composite indicators, sometimes explicitly tailored to local conditions and concerns. A notable one is the New Zealand Living Standards Framework, released in 2018. It includes health, subjective wellbeing, time use, income/consumption, jobs/earnings, housing, cultural identity and the various types of capital asset, and is presented as a dashboard (New Zealand Treasury n.d.). It is, however, unclear whether it has altered the policy direction, and it is likely that 65 indicators are too many to provide clear guidance for policy development (McClure 2021). Similar initiatives are taking place elsewhere, e.g. in Wales and Scotland, and an alliance of “wellbeing economies” has been formed (WEGo n.d.).

While the presence of a variety of separate dimensions in one combined index may be appealing, there is no clear criterion of what should be included. Composite indicators may have some pragmatic merit by covering what is deemed to be important in various contexts, but any particular combination is arbitrary, with little agreement on what should take priority—as is clear from the number of different indices that have been developed; they are “ad hoc and too varied to build a consensus around a new global way of measuring progress” (Allin et al. 2022).

It is perhaps too soon to judge whether any of these indices will turn out to be useful in actual policy development and implementation. The OECD’s Better Life Index, or nationally-specific measures such as the New Zealand Living Standards Framework, may prove to be useful for certain types of policy initiatives in the future. But it is difficult to imagine any such indicator being taken as a serious criterion in the development of specifically economic policy, or of mainstream government policy with major economic implications. It is therefore unlikely to displace GDP as the overwhelmingly dominant economic measure.

In contrast, the proposed IEO and its component indicators provide a simpler and more streamlined methodology that is specific to the economy, and complementary to GDP. Its embeddedness in the rigorous structure shown in Fig. 1 (Chapter 1), i.e. assets, output, outcomes and impact, is a key strength that enables separate types of indicator to be used that are appropriate for monitoring the distinct aspects of the economy, society and the environment. This allows them to be combined in a transparent way when necessary for particular purposes, not only providing conceptual clarity but also allowing the calculation of the efficiency ratio and the sustainability ratio (see Chapter 4), which is impossible with composite indices.

There are some important developments in the “beyond GDP” work that do not involve composite indicators, which are highly compatible with the IEO approach. In particular, an initiative is being developed by the OECD and the European Union to move towards an “Economy of Wellbeing”. Its content closely resembles the perspective argued for in this book, and is almost identical to that of table 1 in Chapter 2. The proposal is to boost improvements in education and skills, ensure access to high-quality healthcare for the whole population, promote health including mental health, pursue social protection and redistribution as well as active labour market policies, and promote gender equality including access to good-quality care and preschool programmes for children (OECD 2019). It would be intended to improve people’s lives and promote upward social mobility, with special attention being given to inequalities and to those at the bottom of the distribution, as well as fostering environmental and social sustainability. This has been endorsed by the European Union, with the addition of access to social services and long-term care, safe and decent working conditions plus fair pay, and access to affordable housing; social inclusion and non-discrimination are also emphasised (Council of the European Union 2019).

The OECD case for an Economy of Wellbeing is partly framed not as an end in itself, but rather as “a “virtuous circle” in which individual wellbeing and long-term economic growth are mutually reinforcing” (i.e. involving what I have called “consequential gain”)—justifying the pursuit of wellbeing as a means to the end of economic growth. However, a parallel OECD initiative, New Approaches to Economic Challenges (NAEC) emphasises that “we need … to stop seeing growth as an end in itself, but rather as a means to achieving societal goals including environmental sustainability, reduced inequality, greater wellbeing and improved resilience” (OECD 2020). This latter position meshes very well with the IEO perspective.

There has also been an initiative to monitor the extent to which basic needs are met, in a wide range of countries, using whatever data are already available derived from a variety of sources (Social Progress Initiative 2022). The Social Progress Imperative has published information on what they term Components of Social Progress. These are “Basic Human Needs”—adequate nourishment and basic medical care, clean water, sanitation, adequate shelter and personal safety; “Foundations of Wellbeing”—access to a basic education, information and communication, healthcare, and a healthy environment conducive to a long life; plus “Opportunity” which is concerned with personal rights, freedom of choice and inclusiveness as well as access to advanced education. The first two categories largely coincide with table 2.​1 in Chapter 2. Annual reports have been published since 2013, and a longitudinal analysis is now available for 170 countries (Harmacek and Krylova 2023a, 2023b; The Economist 2023a).

Considerable progress has been made since Mismeasuring our lives (Stiglitz et al. 2009). The OECD produced a report of progress on their collaborative work in 2018 (Stiglitz et al. 2018). More recently, the United Nations has coordinated activity on the beyond GDP agenda more broadly (UN High-Level Committee on Programmes 2022), and has organised a series of online “Sprints” showcasing the progress being made by national statistical offices, the OECD, divisions of the United Nations and others (UNNES n.d.). This is in the context of the UN Secretary-General’s report Our Common Agenda (UN 2021).

On the technical side, one valuable contribution has been in assessing household disposable income as a measure of “economic wellbeing”—material living conditions, which determine people’s consumption possibilities and their command over resources (Van de Ven 2019). This can be done using the same System of National Accounts that forms the basis for calculating GDP, an approach that can be extended to include household saving and indebtedness, and cash transfers from government to households. And there have been initiatives on measuring household production (e.g. childcare and cleaning) using time use data, which is outside the System of National Accounts (Van de Ven 2019).

Progress has also been made in relation to several practical measurement problems. One has been the difficulty of valuing public services that are free at the point of delivery, and therefore have no observable market or exchange value. Traditionally, they were allotted a value equal to their inputs (the “outputs equals inputs” method). Clearly, this measure would by definition not be able to reflect any change in productivity (Foxton et al. 2019). In the UK, the Atkinson Report examined this issue in great depth. It proposed that the appropriate measure was value added, which was equal to the improvement in outcomes directly attributable to the activities of the public services concerned (Atkinson 2005). Development work has been carried out in the UK to meet this recommendation, and will continue to be needed to update the estimates. Strong progress has been made, by working with subject-matter experts and practitioners, with the result that estimates are now available at low cost, that are well accepted by stakeholders, for approximately half of UK public service output (Foxton et al. 2019).

Another issue is the relationship of GDP to innovation in production. This has become topical in recent years, with the rise of the digital economy. There are two aspects to this: price reductions, and new goods or quality improvement in existing products.

The impact of price reductions, and the failure of conventional measures to reflect them, has become inescapable since more and more functions have become available that are free at the point of use on smartphones, tablets, etc.—or more accurately, that are available in covert exchange for “eyeballs” (attention that facilitates advertising), and for data. The same conclusion applies to the introduction of new goods and services, such as social media, and those of higher quality. A great deal of work, some of which is controversial, has been carried out to address these issues, using willingness-to-pay (e.g. employing incentive-compatible choice experiments) and other approaches (Hulten and Nakamura 2018; Nakamura et al. 2018; Aizcorbe et al. 2019; Brynjolfsson et al. 2019; Heys et al. 2019; Poquiz 2023).

It is less well recognised that the process of real price reduction has been an important feature of successful economies since the industrial revolution. This has been well documented for illumination (Nordhaus 1996), and for multiple consumer items in twentieth-century America (Cox and Alm 1997). The process started in early-nineteenth-century England, and has continued since then in successful economies, with widespread real price falls large enough to have made a major contribution to the increase in prosperity during this period. Real price falls of this magnitude imply that the monetary value, and therefore the economic presence, of each item has fallen relative to its physical quantity [Joffe submitted for publication]. This is therefore a deep and long-standing phenomenon, not just a recent occurrence. Nor is it confined to what Corrado et al. (2017) call the quaternary sector (knowledge production including schooling and R&D). One implication is that the rise in per capita GDP has systematically underestimated the improvement in the standard of living. The recent focus on “free” services only looks at the tip of the iceberg.

Other measurement problems with GDP include intangibles (Corrado et al. 2017), and unproductive financial investment (Coyle et al. 2019). Arguably, adjustments should be made for them, which could be included in an augmented version of GDP. They are not discussed further in the present book.

To summarise, a strong case can be made for retaining GDP as a measure of the amount of activity in the economy, as this is appropriate for economic management. This could be augmented to include unpaid labour, because of the substitutability between paid and unpaid labour, and possibly quality improvement and “free” goods (Hulten and Nakamura 2020), and further extension of the asset boundary to include intangibles (Corrado et al. 2017). There is currently an active debate on these topics, that will not be further discussed here.

The view that wellbeing, in some sense, should be a foundational value is an ancient idea. It goes back to the Vedic philosophers of ancient India, to Confucius in China and to Socrates in Europe (Austin 2020, Chapters 2 & 3). A similar idea applied specifically to the economy goes back at least to Scitovsky (1976; revised edition 1992), who contrasted the economists’ ideal of abundant consumption with psychological evidence on the actual roots of joy.

The idea of monitoring “Gross National Happiness” rather than GDP was proposed by the King of Bhutan in 1972, and adopted as the goal of government in 2008. It comprises sustainable and equitable socioeconomic development, environmental conservation, preservation and promotion of culture, and good governance (Wikipedia n.d.). Bhutanese Gross National Happiness surveys have been conducted periodically since 2008. This initiative has been quite influential, and has been emulated by several cities and regions worldwide. However, international comparisons indicate that Bhutan’s level of happiness is in fact quite average by global standards. The cultural aspect of the measure includes a strong religious (Buddhist) orientation, and the initiative has been criticised because its introduction coexisted with the expulsion of 100,000 non-Buddhist ethnic Nepalese people (Frelick 2008).

The rigorous study of subjective wellbeing, including methods of measuring it, started growing in the late twentieth century, with pioneering work by Easterlin and others. The impetus to monitor it as a guide to policy increased in the early twenty-first century, for example with the publication of Layard’s Happiness (2005). Several national statistical offices now monitor wellbeing, e.g. the UK Office for National Statistics, which started in 2011 (ONS 2019), and the United Nations has published the annual UN World Happiness Report since 2012 (UNSDSN n.d.).

This increased focus on subjective wellbeing is highly encouraging, but it is unclear what practical effect it has had. For example, it has been monitored for over ten years in the UK, yet it is hard to discern any policy initiative that has resulted from the availability of the data. This is not surprising, because knowledge of the level of subjective wellbeing and/or of its trend provides no information on what policy initiative(s) would enhance it. One needs to know what the determinants are, and what policy levers are available. In addition, any such initiative would inevitably be in competition with other policy priorities.

There is also a burgeoning academic literature on wellbeing (e.g. Lee et al. 2021; Layard and De Neve 2023). For example, O’Donnell and Oswald (2015) discuss the relative weights that should be given to happiness, life satisfaction, perceived worthwhileness of life and anxiety in an overall index, and how a measure of its change could be developed. And the International Society for Quality-of-Life Studies is a forum for research in this area, which includes work on economic and material wellbeing among its topics (ISQOLS n.d.). In addition, wellbeing has formed the basis for the analysis of policy and the economy (Bache and Scott 2018; Dalziel et al. 2018; Bache 2020).

Few would disagree that wellbeing, in its various meanings, should influence policy. For many people, it would need to be balanced against other criteria, including protection of the natural world. Others regard it as the sole criterion, as in the extreme view that “the goal of government” should be wellbeing, as measured by reported life satisfaction (emphasis added) (Frijters et al. 2020). All policies would be subjected to a cost-effectiveness ranking in terms of the ratio of extra happiness to cost, based on an official list of “believed effects of various policies and circumstances”. The proponents justify this view partly on the basis that measured wellbeing is predictive of future earnings as well as marital stability and long-term survival, and that measures such as job satisfaction predict future job quitting. The initiative has the potential to promote programmes that increase life satisfaction, especially in the areas of mental health and social relations, such as emotional skills teaching and relationship coaching for high-risk groups. It also collates examples of best practice from which others can learn, and is stimulating important research into the societal determinants of life satisfaction (Frijters et al. 2020).

To that extent, its aim is in close alignment with the orientation proposed in this book. However, its scope is too diffuse for a measure of the success of the economy, and thus for a replacement of that function of GDP: one study found that the main causal factor influencing the degree of adult life satisfaction was diagnosed depression and/or anxiety (46%), with economic factors such as income, employment and education together accounting for less than 20% (Clark et al. 2018). And it is better suited to being applied “throughout the public services and by non-governmental organizations” (Frijters et al. 2020) than to the outcomes of the entire economy.

The monitoring of wellbeing faces measurement problems, because culture and language influence the response to survey questions (Kapteyn 2020). Responses are also sensitive to question order, and to the distorting effects of previous questions—a relationship that differs in different populations (Stone and Krueger 2018). And the reported level of wellbeing depends on mode, e.g. personal interviews, with or without show cards, as against telephone interviews (OECD 2013). In addition, the assessment of life satisfaction has the disadvantage that it is coarse grained because individuals can only answer with whole numbers, and its volatility means that large numbers of people are required in order to obtain stable estimates (Frijters et al. 2020).

Furthermore, a consistent observation is that its three aspects—life satisfaction, affective state (mood) and a sense of purpose—have different drivers, and different consequences for the person concerned (National Research Council 2016). The various measures assess essentially distinct concepts (Stone and Mackie 2015, 2018; Durand 2020). There are also several other methodological issues that require extensive research, such as causal attribution; rapid progress is being made on this research agenda (Stone and Krueger 2018).

A more fundamental feature of subjective wellbeing is that it responds strongly to one’s position relative to that of others. There is clearly an aggregation (fallacy of composition) problem here: it is impossible to raise everyone’s relative position (Kapteyn 2020). At the population level, the strategies that people use to boost their own relative position, and the expenditure involved in doing so, become irrelevant and should therefore be valued at zero.

Another fundamental feature of subjective wellbeing is that it is subject to adaptation. Events that make people better or worse off tend to have only a short-lived effect; the wellbeing score returns to its previous level, or close to it, after the passage of time. This reduces the sensitivity of subjective wellbeing to changes in life circumstances. It may be stronger for negative effects such as disability, entry into poverty or unemployment (Stone and Krueger 2018).

However, it has been noted that subjective indices are more resilient if they are tied to objective components of wellbeing (Corlet Walker and Jackson 2019). That accords with the approach taken in this book, to focus on the conditions that facilitate health and wellbeing.

Frijters et al. (2020) briefly consider combining measurement of subjective wellbeing with length of life, but drop the idea for technical reasons in measuring happiness [sic]. In general, health has had only a minor role in the “beyond GDP” discussions and proposals, although it has appeared in some of the composite indicators, notably as life expectancy in the HDI; and Deaton and Schreyer (2020) have advocated greater attention to health in the wake of the Covid-19 pandemic.

The neglect of health is odd. There is a strong empirical basis for the suggestion that health status is largely determined by living conditions, and specifically by the extent to which basic needs are met. It is well established—and widely known—that life expectancy depends on social conditions, with differences of up to ten years between rich and poor areas (e.g. Iacobucci 2019), and a gradient across areas of intermediate prosperity. This consistent observation cannot be wholly attributed to healthcare variations, or to “lifestyle” differences such as smoking rates. A similar social mortality gradient of Covid-19-related deaths was widely observed during the pandemic.

The World Health Organization states that the social determinants of health account for 30–55% of health outcomes, more than either healthcare or lifestyle choices. They “are the conditions in which people are born, grow, work, live, and age, and the wider set of forces and systems shaping the conditions of daily life”, and that “[i]n countries at all levels of income, health and illness follow a social gradient: the lower the socioeconomic position, the worse the health” (WHO n.d.; see also Marmot 2010; Braveman 2023). The observation of a gradient across the whole population is consistent and important: it is not a dichotomy of rich versus poor, or due to the existence of a marginalised subgroup.

The WHO lists factors that influence health outcomes and health inequities (“unfair and avoidable differences in health status”) (Table 3.1). The WHO list contains ten items, eight of which correspond closely to table 2.​1 in Chapter 2. The only exceptions are #8 and #9, which are not economic outcomes. Similarly, the US list comprises five items very similar to those in table 2.​1, plus racism, discrimination and violence (US Office of Disease Prevention & Health Promotion n.d.). Health status, including such “hard” endpoints as infant mortality risk and life expectancy, is structured by socioeconomic position, and related to the fulfilment of basic human needs. It is also responsive to policies across the different sectors of the economy, which has long been recognised by public health experts under the heading “Health in All Policies” (HiAP) (WHO 2014).

The neglect is also odd for another reason. When an earthquake, a hurricane or a train crash occur, it is routine for the death toll to be reported. And when a mining company pollutes the area around a mine, a primary concern is with chemicals that are toxic to humans and/or wildlife—i.e. a threat to their health. Yet the loss or gain in health resulting from economic policies is rarely reported in the same way, in spite of strong evidence of a link. For example, the austerity policies of the 2010s were followed by a large number of excess deaths. The precise number is unclear. In the UK, the estimates include 131,000 (IPPR 2019), more than 230,000 (Darlington-Pollock et al. 2021) or 250,000 (excluding deaths explained by Covid-19) (The Economist 2023b). The latter number is in addition to the 300,000 that would be expected on the basis of the life expectancy decline in similar European countries. This UK figure is associated with deprivation—and especially with particularly severe austerity policies that affected spending on social care, housing, etc., in the most deprived areas (The Economist 2023b). It has currently not been established that these observations can be taken at face value as representing cause and effect, but even the possibility prompts the suggestion that human health consequences should be a major criterion in judging economic policies.

Furthermore, most people would agree that health is at least as important as subjective wellbeing—having a broken heart is bad, but a heart attack is worse. The experience of the Covid-19 pandemic has probably raised awareness of the central importance of health. It should be included as a major criterion for the evaluation of the success of an economy (and more broadly, of a society).

In addition, omission of health as a primary outcome leads to an odd distortion, as with GDP and the man who married his housekeeper, as mentioned above. If a depressed person were to die, average happiness would increase (apart from the impact on family and friends).

Conversely, it would be equally misguided to omit subjective wellbeing and focus exclusively on health (even including mental health). In the evaluation of the economy, concern only with health would result in an indicator that is too insensitive. Many products and activities do not have a strong health impact, but they do enhance the pleasure of living in a way that is too subtle to be captured as mental health.

The strongest position is that health-plus-happiness should be the ultimate goal of policy. Wellbeing should be construed in a broad, inclusive sense that encompasses biological measurements as well as self-reports—what may be called inclusive wellbeing. However, the best way of achieving this goal is not to monitor these endpoints; rather, monitoring their economic determinants is a superior way of assessing the success of the economy in providing for individuals, households and society. There are several reasons for this.

One is that neither health nor wellbeing responds rapidly to changes in the economy at the aggregate level, unless there is a major shock that affects the whole population, as with the Covid-19 pandemic or a war. Relatedly, both health and happiness tend to be strongly influenced by factors throughout the life course, especially in childhood (Marmot 2010; Layard et al. 2014): homelessness, financial insecurity and unemployment cast a long shadow, and inadequate education leading to illiteracy is a lifelong burden. This is not a reason to ignore such causal factors, but rather, to monitor the causal factor itself rather than to wait for its impact—to focus on the intermediate outcome, as discussed in Chapter 2. A paradigm case is that access to good-quality early years education and childcare should itself be counted as a major contributor to economic wellbeing. Its benefits would be invisible if reliance were placed on monitoring health and wellbeing themselves, as many of these impacts take years or decades to become fully manifest, and causal attribution would be extremely challenging.

Second, each has specific causes that are inappropriate for assessing the success of the economy. In the case of health, some are unrelated to the economy, such as genetic disorders. Others, for example cigarette smoking, are themselves partly due to job insecurity and/or financial hardship. In assessing the outcome of economic activity, that is a reason to monitor the economic determinants of smoking (Marteau et al. 2021).

In the case of wellbeing, a dominant influence is the quality of interpersonal relationships, and there are also cultural factors that play an important part. These are best regarded as distinct from the economy. This does not mean that they are completely separate: employment is an important source of social relationships, and economic activity can affect relationships in other ways, e.g. if basic needs are not met, or are insecure, leading to anxiety. The economy may also have a causal role in another sense, if it requires a great deal of mobility and therefore disruption of social relationships, as with the large-scale migrant labour in the Gulf States from other parts of Asia.

The implication of all these problems is that health and subjective wellbeing do not provide rapidly responsive measures of the quality of life attributable to the economy, and wellbeing has additional issues, both methodological and substantive. Also, in practical terms, even if they can successfully assess “how well a country is doing”, they are not useful as a guide indicating how it might do better. This is because such indicators do not enable a link to be made with any specific sector or policy within the economy.² They therefore do not provide clear guidance for action. On the other hand, monitoring the economic determinants of health and subjective wellbeing is eminently practical and reliable. These economic outcome measures are responsive to changing conditions. And as argued above, monitoring such intermediate outcomes is more informative because it provides information on what measures could be taken to improve the situation.

Finally, health and wellbeing indicators produce an average score for the whole population. When they are used for monitoring, a separate measure is therefore required for the assessment of inequality, or at least, separate scores for subgroups of the population.