Growth of What

Essay · economics · September 2026

GDP, the PIB in Portuguese, measures how much an economy produces and sells, and it does that well. It was never meant to measure how well people live, and its own rulebook says so. Set Portugal beside other yardsticks, from household income to life expectancy, well-being and the wear on its natural capital, and the picture changes: sometimes better than GDP says, sometimes worse, and never quite the same.

Reading time ≈ 13 min What it counts → what it misses → the rivals Part II of two

The question

The first part of this essay, The Treadmill, asked why modern economies seem to need growth, and found that jobs, debts, pensions and politics were all built on it. It left open what exactly is meant to grow. In practice the answer is one number, gross domestic product: the produto interno bruto, or PIB, that the statistics office publishes every quarter and every government is judged by.

This part looks at what that number counts, what it leaves out, and what happens when Portugal is measured in other ways. None of the alternatives is free of problems, and the case for GDP is stronger than its critics usually admit. But the differences are large enough to matter for what a country thinks it is doing well.

18th vs 9thPortugal’s rank among the 27 EU countries on GDP per head, and on life expectancy

55%Ireland’s national income, adjusted for multinational accounting, as a share of its GDP in 2025

+15% vs +26%growth of an index of sustainable welfare and of GDP per head in Portugal, 1995–2006

−33%greenhouse gas emissions per person in Portugal since 2008, while real GDP per head rose 14%

The number

What GDP counts

GDP is the market value of all the final goods and services produced in a country in a year. It can be measured three ways, which should agree: add up the value each producer adds, add up what households, firms, the State and foreigners spend on final goods and services, or add up the incomes that production pays out as wages and profits.[1] The rulebook, the United Nations’ System of National Accounts, decides what counts as production, and its choices are practical rather than philosophical.

Some things are counted even though no one pays for them. People who own their homes are treated as paying rent to themselves, so that GDP does not fall when tenants become owners. Schools, hospitals and courts are valued at what they cost, since they have no price. Illegal activities that are bought and sold count too: since 2014 EU countries have included drugs, prostitution and smuggling in their accounts, which added about 1 per cent to Italy’s GDP.[1][2] Other things are left out although they matter a great deal. Cooking, cleaning, caring for children and for the old are excluded when done at home without pay, and counted when the same work is paid. Leisure is not counted, nor the wear on forests, fish stocks and the climate.

The rulebook says plainly that this is deliberate. Unpaid work at home, it notes, makes “an important contribution to economic welfare. However, national accounts serve a variety of analytical and policy purposes and are not compiled simply, or even primarily, to produce indicators of welfare.”[1] The economist who led the first official estimates of American national income said the same in 1934: “The welfare of a nation can, therefore, scarcely be inferred from a measurement of national income as defined above.” The popular version usually drops the last three words, and the same report also judged income per head “illuminating of movements in the nation’s economic welfare”.[3] Simon Kuznets did not oppose growth. In 1962 he endorsed “as high and sustained a growth rate as is compatible with the costs that society is willing to bear”, and asked that goals for more growth “specify more growth of what and for what.”[4]

The best-known attack came from Robert Kennedy, in 1968: the national product “counts air pollution and cigarette advertising, and ambulances to clear our highways of carnage”, yet “does not allow for the health of our children, the quality of their education or the joy of their play”; it measures “everything in short, except that which makes life worthwhile.”[5]

Gross and national

Two corrections inside the accounts

Two corrections are available without leaving the national accounts. The first is to subtract the capital used up in production, which turns gross domestic product into net domestic product. The 2008 rulebook admits that GDP is used even though it is “economically inferior” to the net figure, because gross figures are available sooner and compare better across countries. The new 2025 rulebook goes further: it names the growth of net domestic product “the conceptually preferred measure of economic growth, not replacing but to be used alongside” GDP, and for the first time subtracts the depletion of natural resources as a cost.[1]

The second is to count the income that belongs to a country’s residents rather than the production that happens on its soil: gross national income, GNI. For Portugal the difference is small; its national income has been between 96 and 101 per cent of its GDP for thirty years. For Ireland it is enormous (Figure 1). Multinationals that book profits and intellectual property there inflate its GDP; Ireland’s own statisticians publish a modified national income, GNI*, which strips those effects out. In 2025 it was 55 per cent of GDP.[6][7] On GDP per head in purchasing power Ireland ranks second in the EU, at more than twice the average; a large part of that output belongs to foreign owners.

Fig. 1 — Gross national income as a percentage of GDP, Portugal and Ireland, 1995–2025; dots: Ireland’s modified gross national income (GNI*), 2024 and 2025. Data: Eurostat, nasa_10_nf_tr; CSO Ireland, Annual National Accounts 2025.

Portugal, measured

Income, the median and the index

The Stiglitz–Sen–Fitoussi commission, set up by the French government in 2008, recommended looking at the household rather than the economy: “trends in citizens’ material living standards are better followed through measures of household income and consumption.”[8] In Portugal the two tell similar stories with different endings (Figure 2). Real disposable income per person, including the value of free public services, fell further than GDP in the crisis and has risen faster since 2013: by 30 per cent against 24 per cent. The income of the median person, the one in the middle, rose faster still, because inequality fell: the Gini coefficient of disposable income went from 34.5 in 2014 to 30.9 in 2025.[9] The Human Development Index, which combines income with health and education, rose steadily through the crisis, because life expectancy and schooling kept improving while incomes fell.[10]

Fig. 2 — Portugal, index 2005 = 100: real GDP per head; real adjusted gross disposable income of households per head (including public services in kind); real median equivalised disposable income (EU-SILC, by survey year, deflated by the consumer price index of the income year, to 2024); Human Development Index (to 2023). Data: Eurostat, nama_10_pc, nasa_10_ki, ilc_di03; UNDP, Human Development Report 2025; author’s calculation.

INE, the Portuguese statistics office, publishes its own well-being index, built from indicators in ten domains grouped into two halves: material living conditions and quality of life. It rose from 22.5 in 2004 to 47.0 in 2024. Its two halves parted company in the crisis (Figure 3). Material conditions, meaning income, jobs and economic vulnerability, fell from 29 to 21 between 2007 and 2013; quality of life, meaning health, education, security, the environment and social ties, rose from 30 to 38 over the same years.[11] A GDP chart for those years shows only the first half.

Fig. 3 — INE Índice de Bem-estar, Portugal, 2004–2024: material living conditions, quality of life and the overall index (dotted), on INE’s 0–100 scale, 2025 edition (earlier editions used a different base and should not be mixed with these values). Data: INE, via PORDATA.

Subtracting the costs

Welfare net of damage

A more radical approach starts from consumption and then adds what GDP leaves out and subtracts what it wrongly counts as a gain: add the value of housework and volunteering, subtract the cost of commuting, crime, pollution and the depletion of resources, and adjust for inequality. The result goes by the names of Index of Sustainable Economic Welfare (ISEW) or Genuine Progress Indicator. For seventeen countries with half the world’s population, one study found that the genuine progress indicator per head peaked in 1978, and that “GPI/capita does not increase beyond a GDP/capita of around $7000/capita.”[12]

A Portuguese estimate, a thesis at NOVA University, followed the index from 1950 to 2006 (Figure 4). Over the whole period it grew by about 2.9 per cent a year against 3.7 for GDP. Between 1995 and 2006 it rose by 15 per cent while GDP per head rose by 26, and it fell after 2000 while GDP kept rising slowly.[13] These indices are only as good as the prices they put on things that have none: an hour of housework, a tonne of carbon, a unit of inequality. Change those prices and the results change. Their value is in showing that the costs can be large, not in fixing their exact size.

Fig. 4 — Portugal: GDP per head and the Index of Sustainable Economic Welfare (scenario A) per head, euros at 2000 prices, 1950–2006. Data: S. Dias, Crescimento Económico, Sustentabilidade e Desenvolvimento: o caso de Portugal, MSc thesis, FCT-NOVA, 2009, Quadro A.20.

A related idea is to track wealth rather than income: not how much is produced this year, but whether the stock of assets that will produce future income is growing. The Dasgupta Review for the British Treasury reported that between 1992 and 2014 produced capital per head doubled worldwide while “the value of the stock of natural capital per head declined by nearly 40%.”[14] On the World Bank’s estimates, Portugal’s real wealth per head rose by 23 per cent between 1995 and 2020; produced capital rose by 65 per cent and human capital by 39, while renewable natural capital fell slightly and non-renewable resources, small to begin with, halved.[15] The World Bank’s adjusted net saving, which counts what a country saves after subtracting depreciation, resource depletion and pollution damage and adding spending on education, puts Portugal below Spain, Italy, France, Germany and the EU average in every year since 2000, and below zero in 2010.[16] By that measure Portugal has been consuming much of what it produced rather than building up the stock that future income depends on.

Living well

Health, happiness and rank

The case for GDP rests partly on how much else goes with it. Across 186 countries, the logarithm of GDP per head explains most of the variation in life expectancy (a correlation of 0.86); across 139, it correlates with life satisfaction almost as closely (0.79).[7] But the relationship flattens among rich countries (Figure 5). Portugal, with about 42,000 dollars of GDP per head in purchasing power, lives four years longer than the United States with about 74,000.

Fig. 5 — GDP per head in purchasing power parity (constant 2021 international dollars, log scale) and life expectancy at birth, 2023, 186 countries and territories. Data: World Bank, World Development Indicators; author’s calculation.

Charles Jones and Peter Klenow turned this into a single measure: how much consumption would make a person indifferent between living in the United States and living in another country, taking account of leisure, life expectancy and inequality. By that measure France, with GDP per head of 67 per cent of the American level, reaches 92 per cent; Western Europe as a whole about 85 per cent, against 67 on income alone. Longer lives, more leisure and less inequality each add about ten points for France.[17] The same study does not include Portugal, but its method suggests that Portugal’s long lives would narrow its income gap, and its long working hours would widen it.

Whether money buys happiness is still argued. Richard Easterlin found in 1974 that in the United States since 1946 “higher income was not systematically accompanied by greater happiness”, and maintains that over the long run the growth of happiness and of income are “not significantly related”. Betsey Stevenson and Justin Wolfers, using more countries and years, found a steady relationship between life satisfaction and the logarithm of income, with “no evidence of a satiation point”.[18] In Portugal, satisfaction and income both rose after 2013: the Gallup life evaluation went from about 5.1 in the middle of the last decade to 6.0 in 2022–2024, while in the United States it fell.[19]

Put Portugal among its EU partners and the ranking depends on the yardstick (Figure 6). It is 18th of 27 on GDP per head in purchasing power, 9th on life expectancy, 10th on healthy life years, 17th on life satisfaction in the EU’s survey, 23rd on Gallup’s, and 21st on the Human Development Index, held back by its education component, where older Portuguese still have much less schooling than their European peers.[9][10]

Fig. 6 — Portugal’s rank among the EU countries on six measures, latest year available (healthy life years: 26 countries; the Portuguese series has breaks, including in 2024). Data: Eurostat, prc_ppp_ind, demo_mlexpec, hlth_hlye, ilc_pw01; UNDP; World Happiness Report 2025; author’s ranking.

The rivals

Dashboards, indices and budgets

The debate among statisticians is no longer about whether GDP is enough, but about what to put next to it. The Stiglitz–Sen–Fitoussi report of 2009 argued that “what we measure affects what we do”, but that “changing emphasis does not mean dismissing GDP”; it recommended a dashboard, with household income, its distribution, quality of life and, separately, sustainability.[8] The OECD now publishes more than eighty well-being indicators for its members; INE has its index; the EU’s environment programme for 2030 commits it to “developing a summary dashboard and indicator set measuring ‘beyond GDP’”.[20] And the 2025 rulebook itself adds chapters on well-being and sustainability, household income by decile, and experimental accounts for unpaid work.[1]

A few governments tried to put well-being at the centre of the budget. Wales passed a law in 2015 requiring its public bodies to pursue seven well-being goals. New Zealand presented a “Wellbeing Budget” in 2019, with priorities such as mental health and child poverty. Six years later a new government proposed to repeal the legal requirement for well-being objectives, arguing that they “have not generated transparency or accountability benefits”.[21] Bhutan measures Gross National Happiness from a survey of nine domains.[20] The New Zealand experience shows the difficulty: a dashboard has no single number to hold a government to, and a single index hides the choice of weights.

Growth of what

Output, emissions and materials

The sharpest version of the question is environmental: can GDP keep growing while the use of energy and materials falls fast enough? In Portugal, since 2008, greenhouse gas emissions per person have fallen by a third while real GDP per head rose by 14 per cent (Figure 7). The material footprint, the raw materials used to meet Portuguese demand wherever they were extracted, fell by more than 40 per cent by 2013, largely because the construction industry collapsed, and has stayed near that level.[9]

Fig. 7 — Portugal, per person, index 2008 = 100: real GDP; greenhouse gas emissions excluding land use and forestry; raw material consumption (material footprint; 2024 estimated). Data: Eurostat, nama_10_pc, env_air_gge, env_ac_rme, demo_pjan; author’s calculation.

This is what economists call absolute decoupling, and the evidence on how common it is divides the field. A review of 835 studies found that “examples of absolute long-term decoupling are rare” and that observed rates are far too slow for the cuts needed.[22] A study of 36 high-income countries found that 11 achieved absolute decoupling of emissions from growth between 2013 and 2019, and that at those rates they would take more than 220 years to cut emissions by 95 per cent.[23] Other work stresses that 18 developed countries cut their carbon emissions between 2005 and 2015 while their economies grew.[24]

Three positions follow. Green growth, the view of the OECD and most governments, holds that the economy can keep growing while its environmental impact falls. Degrowth holds that rich countries cannot cut emissions and materials fast enough while growing, and should deliberately shrink the parts of the economy that use most of them. A third position, called “agrowth” by the economist Jeroen van den Bergh, holds that since GDP is “a very imperfect indicator of social welfare”, governments should stop targeting it altogether and pursue their social and environmental goals directly, “being indifferent about growth”.[25]

The balance

A good measure of one thing

GDP does what it was built to do. It measures market production consistently, quickly and in a way that is hard to manipulate, and it moves with employment, tax revenue and the capacity to pay debts, which is why the first part of this essay found so much depending on it. It correlates, across countries, with most things people value. What it does not do is tell a country how well its people live, who receives the income, what was used up to produce it, or what is left for later. Its own rulebook, its first builder and its latest revision all say so.

For Portugal the alternatives do not reverse the story, but they change its emphasis. By most measures the country is better off than GDP alone suggests: its people live long lives, inequality has fallen and the crisis years hurt incomes more than health or education. By others it is worse: it has saved and invested little of what it produced, it works long hours for its income, and its catch-up with Europe on income has stalled for thirty years. A country that reported median household income, healthy life years and emissions next to GDP every quarter would not have a different economy. It would be arguing about different numbers, which is where Kuznets’s question starts: more growth of what, and for what.

On method and tools

This piece was written collaboratively with Claude Opus 5.5 (Anthropic): human specification, editorial direction and critical review; machine data research, analysis and drafting. The figures and derived numbers are computed by scripts/growth.py, shared with the first part, from Eurostat, the UNDP, the World Bank, the World Happiness Report, INE (via PORDATA), CSO Ireland and a Portuguese ISEW thesis. Quotations were checked against the primary documents: the 2008 and 2025 System of National Accounts, Kuznets’s 1934 report and 1962 article, and the JFK Library transcript of Robert Kennedy’s speech; common misquotations of the first three are noted in the data files. The Irish figures were read from the CSO release through a summarising tool and should be re-checked; the Portuguese median income for the 2025 survey is not used because its jump needs confirming against INE; the INE well-being index is taken from PORDATA’s copy of the 2025 edition. The downloaded sources, with a table or page reference for every number, are kept with the script’s data; results are in docs/growth-results.json and the figures in docs/growth-figures.html.

The cover photograph is Contentores, Terminal XXI, the port of Sines, by Nuno Morão; CC BY-SA 2.0, via Wikimedia Commons, cropped.

Authored by: Luis Matos Ferreira — Physicist, Developer, Writer

Related essays on this blog

1. The Treadmill — Part I: why modern economies seem to need growth.
2. Where the Hours Went — income, consumption, unpaid work and the environment.
3. The Two-Thirds Country — Portugal’s productivity, hours, wages and housing.

Sources

1. United Nations et al., System of National Accounts 2008, paras. 2.138–2.142 (pp. 34–35), 6.27–6.29 (p. 98), 6.43–6.45 (p. 100) and 6.130; System of National Accounts 2025, pre-edit version, “Key changes” (PDF pp. 8–9) and para. 1.64.
2. Eurostat, Handbook on the compilation of statistics on illegal economic activities in national accounts, 2018, foreword; Istat, “ESA 2010: questions and answers”, 14 March 2014.
3. S. Kuznets, National Income, 1929–1932, Senate Document No. 124, 73rd Congress, 1934, pp. 6–7.
4. S. Kuznets, “How To Judge Quality”, The New Republic, 20 October 1962.
5. R. F. Kennedy, remarks at the University of Kansas, 18 March 1968, transcript, John F. Kennedy Presidential Library.
6. Central Statistics Office, Ireland, Annual National Accounts 2025, “GNI* and de-globalised results”, 2 July 2026.
7. Eurostat, national accounts (nasa_10_nf_tr, nama_10_pc); World Bank, World Development Indicators (NY.GDP.PCAP.PP.KD, SP.DYN.LE00.IN); World Happiness Report 2025, data for Figure 2.1; author’s calculations.
8. J. E. Stiglitz, A. Sen and J.-P. Fitoussi, Report by the Commission on the Measurement of Economic Performance and Social Progress, 2009, pp. 7, 12–18.
9. Eurostat: nama_10_pc, nasa_10_ki, ilc_di03, ilc_di12, demo_mlexpec, hlth_hlye, ilc_pw01, prc_ppp_ind, env_air_gge, env_ac_rme, demo_pjan; author’s calculations and ranks.
10. UNDP, Human Development Report 2025, composite indices time series, 1990–2023.
11. INE, Índice de Bem-estar 2004–2024 (release of November 2025), via PORDATA.
12. I. Kubiszewski et al., “Beyond GDP: Measuring and achieving global genuine progress”, Ecological Economics 93, 2013, pp. 57 and 66.
13. S. Dias, Crescimento Económico, Sustentabilidade e Desenvolvimento: o caso de Portugal, MSc thesis, Faculdade de Ciências e Tecnologia, Universidade NOVA de Lisboa, 2009, Quadro 4.2 (p. 65) and Quadro A.20 (pp. 127–130).
14. P. Dasgupta, The Economics of Biodiversity: The Dasgupta Review, HM Treasury, 2021, p. 114.
15. World Bank, The Changing Wealth of Nations 2024, and its wealth accounts database (Portugal, 1995–2020); author’s calculation.
16. World Bank, World Development Indicators, adjusted net savings including particulate emission damage (NY.ADJ.SVNG.GN.ZS), 1995–2021.
17. C. I. Jones and P. J. Klenow, “Beyond GDP? Welfare across Countries and Time”, American Economic Review 106(9), 2016; author version of February 2016, pp. 2–3 and Table 2.
18. R. A. Easterlin, “Does Economic Growth Improve the Human Lot?”, in David and Reder (eds.), Nations and Households in Economic Growth, 1974, p. 118; R. A. Easterlin and K. J. O’Connor, “The Easterlin Paradox”, IZA Discussion Paper 13923, 2020; B. Stevenson and J. Wolfers, “Economic Growth and Subjective Well-Being”, Brookings Papers on Economic Activity, Spring 2008.
19. World Happiness Report 2025 and 2026, Gallup World Poll life evaluations, three-year averages.
20. OECD, How’s Life? 2024; Decision (EU) 2022/591 on the 8th Environment Action Programme, art. 3(e); Centre for Bhutan and GNH Studies, GNH Survey Report 2022.
21. Well-being of Future Generations (Wales) Act 2015, ss. 2–4; New Zealand Treasury, Wellbeing Budget 2019; New Zealand Minister of Finance, cabinet paper on amendments to the Public Finance Act, released May 2025, paras. 35–36.
22. H. Haberl et al., “A systematic review of the evidence on decoupling of GDP, resource use and GHG emissions, part II”, Environmental Research Letters 15, 2020, abstract.
23. J. Vogel and J. Hickel, “Is green growth happening? An empirical analysis of achieved versus Paris-compliant CO2–GDP decoupling in high-income countries”, Lancet Planetary Health 7, 2023, pp. e759 and e762.
24. C. Le Quéré et al., “Drivers of declining CO2 emissions in 18 developed economies”, Nature Climate Change 9, 2019.
25. J. C. J. M. van den Bergh, “Environment versus growth — A criticism of ‘degrowth’ and a plea for ‘a-growth’”, Ecological Economics 70, 2011, abstract; G. Kallis, “In defence of degrowth”, Ecological Economics 70, 2011; OECD, Towards Green Growth, 2011.