Worth Producing

Essay · economics · September 2026

Productivity counts what people pay for, not whether it is worth having. Some of what rich economies produce is plainly wasted: food thrown away, treatment that does not help, products replaced before they wear out, finance that grows faster than it serves. Much of their borrowing has gone into houses rather than things that produce. None of this needs a planned economy to fix. It needs prices that tell the truth, rights for buyers, and rules that stop credit from favouring the wrong things.

Reading time ≈ 10 min What we count → what we waste → what we borrow for Part IV of four

The question

Off the Treadmill found that the easiest way out of Portugal’s debt arithmetic is productivity: more output from each hour of work. But productivity depends on what the hours produce. If a large part of what an economy makes is wasted, or its borrowing finances things that do not pay their way, then more growth of the same kind is a poor answer, and producing better things may matter as much as producing more. This essay asks three questions: what productivity actually measures, how much of what rich economies produce and borrow is waste, and whether there are ways to reduce it that do not involve the State deciding what gets made.

161 kgof food wasted per person in Portugal in 2024, 105 kg of it at home; EU average 129 kg

~60%of bank lending in rich countries is now for real estate, against about 30% in 1900

€2.80 vs €8.00capital a bank must hold per €100 lent for a mortgage and for a loan to a firm

10–16%cut in emissions attributed to the EU’s carbon market in the firms it covers

What we count

Productivity measures price, not worth

Labour productivity is the value of what an hour of work produces, at market prices, after removing inflation. It counts what people pay for. A meal cooked and eaten counts the same as a meal cooked and thrown away; an unnecessary scan counts the same as a necessary one; a phone replaced after two years adds more to output than one that lasts five. Where there is no price, as in schools and hospitals, the accounts value output at what it costs, so productivity there can hardly rise however much better the service gets, as Growth of What described.[1]

Finance shows how far measured productivity can drift from usefulness. Andrew Haldane, then at the Bank of England, pointed out that the national accounts count the extra margin banks charge for taking more risk as extra output: British finance reached 9 per cent of output at the end of 2008, as the crisis broke. Adjusted for risk, the banks’ lending output would have been about 60 per cent of the official figure, and the sector’s measured productivity growth, about twice the economy’s before 2007, a “mirage”.[2] Thomas Philippon found that the cost of financial intermediation in the United States, 1.5 to 2 per cent of assets a year, is about as high today as around 1900, despite a century of computers.[3] The measure is not wrong; it answers a narrower question than the one people ask of it.

What we waste

How much, where

“Waste” is easy to assert and hard to add up; one person’s waste is another’s choice. The categories where the evidence is firm are narrower than the rhetoric, but real.

Food. Portugal wasted 161 kg of food per person in 2024, against an EU average of 129; 105 kg of it was thrown away in homes, and shops wasted twice as much per person as the EU average (Figure 1).[4] An EU directive of 2025 requires a 30 per cent cut in waste from shops, restaurants and homes by 2030; on the latest figures Portugal has made no progress towards it.[5]

Fig. 1 — Food waste per person by stage of the food chain, kilograms, Portugal and the EU, 2024. Countries’ measurement methods differ. Data: Eurostat, env_wasfw.

Health care. The OECD estimates that “up to one-fifth” of health spending could be put to better use, citing unnecessary tests and treatments, avoidable admissions and administrative waste; Portugal is among the countries where at least one emergency visit in five is judged inappropriate.[6] Direct counts of clearly low-value services are much smaller, 0.6 to 2.7 per cent of spending in one American study; the larger estimates include prices and administration.[7] Portugal spends about 10 per cent of its GDP on health.[4]

Products that do not last. A study for the German Environment Agency found that large appliances replaced because they broke within five years rose from 3.5 to 8.3 per cent of replacements between 2004 and 2012, and that 30 per cent were replaced while still working. It “could not confirm” deliberate planned obsolescence; shorter lives came from cheaper design, repair costs and changing tastes.[8] The history of the idea, and of advertising that encourages early replacement, is in The Manufacture of Wants.

Finance. More finance helps an economy up to a point. Studies at the IMF and the Bank for International Settlements find that the benefit turns negative once private credit reaches about 80 to 100 per cent of GDP, or the financial sector passes about 3.5 per cent of employment, partly because finance draws skilled people and favours projects with good collateral, such as property, over projects with high returns, such as research.[9][10]

Other estimates, of the share of workers employed to guard, supervise and enforce, or of spending that only repairs the damage done by other spending, are more speculative and depend heavily on definitions.[11] The honest summary is that waste is real, concentrated in a few areas, and not a fixed share of the economy that can be switched off.

Why

Not inevitable, but not accidental

Economics has names for most of these failures, and each suggests a remedy. When a cost falls on someone else, as pollution does, the market overproduces it; Pigou proposed taxing it. When buyers cannot judge quality, as with durability or medical treatment, sellers compete on price and appearance rather than on what cannot be seen. When goods are valued for their rank rather than their use, as with status goods, spending becomes an arms race no one wins. A seller with market power in durable goods has a reason to make them last less: Jeremy Bulow showed that such a firm will “desire uneconomically short useful lives”.[12] And where rules create rents, effort goes into capturing them rather than producing: Anne Krueger estimated the rents created by import licences and similar controls at about 7 per cent of India’s national income, and 15 per cent of Turkey’s.[13]

None of these is a law of nature, and none requires the State to decide what gets produced. Each is a case where the price, the information or the rules are wrong, and can be corrected while leaving buyers and sellers to decide.

What we borrow for

Debt that builds and debt that doesn’t

Debt has a clear economic purpose: to bring forward spending whose benefits come later. A firm borrows to build a factory whose output will pay the interest; a family borrows to spread the cost of a house over the years it lives there; a State borrows for roads and schools that serve future taxpayers who will share the cost. Britain’s old fiscal “golden rule” put it simply: “borrow only to invest and not to fund current spending”.[14] Debt becomes excessive when it finances things that do not generate the income to repay it, or when it inflates the price of assets that already exist.

Over the last century, bank lending in rich countries shifted from firms to houses. The share of mortgages in bank lending rose from about 30 per cent in 1900 to about 60 per cent today, and Jordà, Schularick and Taylor estimate that real-estate lending now makes up about two-thirds of banking; mortgage booms are followed by deeper recessions.[15] Across countries, a rise in household debt predicts slower growth over the following three years, by about 2 points of GDP for a one-standard-deviation rise, while “no such relation” exists for the debt of firms.[16]

Fig. 2 — Portugal: loans outstanding from banks and other monetary financial institutions to non-financial corporations and to households for house purchase, December of each year, per cent of GDP, 2003–2025. Data: ECB, balance-sheet statistics; Eurostat, GDP; author’s calculation.

Portugal followed the pattern closely (Figure 2). In the 2000s its banks, borrowing abroad, lent heavily to households for houses and to construction and real-estate firms, which took close to two-fifths of bank lending to firms in 2007; the country ran current-account deficits of about 10 per cent of GDP a year.[17] The Productivity Council later found that in 2013, 44 per cent of the stock of credit had gone to firms of very low productivity.[18] The public debt, meanwhile, rose while net public investment was negative every year from 2012 to 2022: the State was borrowing without building.[19] Portugal’s debt problem was, in large part, a problem of what the debt was for.

What can be done

Prices, rights and rules

The tools that work without planning fall into three groups.

Prices that include the costs. The strongest evidence is for carbon pricing. Studies of the EU’s emissions trading system find that it cut emissions in the firms it covers by about 10 to 16 per cent; a review of schemes worldwide finds cuts of 5 to 21 per cent.[20] The principle extends to any cost that falls on others: packaging, landfill, congestion.

Rights and information for buyers. Since July 2026 the EU requires manufacturers of many appliances to offer repairs at reasonable prices after the warranty expires, and bans the destruction of unsold clothing and shoes.[21] Evidence on softer measures is mixed: France’s repairability score on products, introduced in 2021, had no significant effect overall on what people bought, and only a small one online.[22] Portugal already taxes appliance repairs at the reduced VAT rate of 6 per cent.[23] In health care, telling doctors about low-value treatments does little on its own, while more active programmes, with feedback and changes to how care is ordered, succeed far more often: in about two-thirds of studies, against one in eight for guidelines alone.[7] France’s 2016 law obliging large supermarkets to give away unsold food was followed by a 23 per cent rise in donations, though without a comparison to show how much was due to the law.[24]

Rules that stop credit favouring property. Bank capital rules are a quiet form of credit steering already. Under the EU’s standard approach, a bank must hold about €2.80 of capital for every €100 of mortgage lending, against €8 for a loan to an unrated firm and about €6.10 for a small firm (Figure 3).[25] Mortgages are safer for the bank, but the difference also makes them cheaper to supply. Portugal has begun to lean against it: since October 2024 banks must hold an extra buffer of capital equal to 4 per cent of their residential mortgage exposures, and since 2018 the Banco de Portugal has capped loans relative to property values and to incomes.[26][19]

Fig. 3 — Minimum total capital (8 per cent of risk-weighted assets, before buffers) per €100 of lending under the EU standardised approach: residential mortgage (35 per cent risk weight), loan to a small or medium-sized firm (100 per cent with the SME supporting factor of 0.7619) and loan to an unrated firm (100 per cent). Banks using internal models hold different amounts. Data: Regulation (EU) 575/2013, arts. 92, 125 and 501; author’s calculation.

Taxes favour debt too. Most tax systems let firms deduct interest but not the cost of equity, which encourages borrowing over investment from owners’ own funds. An allowance for the cost of equity removes the bias; Belgium and Italy have used one, and the European Commission proposed one in 2022 before withdrawing the proposal. Portugal had a similar allowance for share capital, which was replaced in 2023 by a new incentive for capitalising companies.[27] For public debt, the old golden rule, borrowing only for net investment, has long been proposed for the EU’s fiscal rules.[28] More interventionist tools exist, such as Japan’s post-war “window guidance”, in which the central bank told banks where to lend, but they are closer to the planning the essay set out to avoid.

The balance

Producing things worth producing

The link to the rest of this series is direct. An economy that wasted less food, used health care better, kept its products longer and lent for factories rather than for bidding up the price of existing houses would need less growth, measured as GDP, to give its people the same standard of living, and would carry less debt for the same capital. Some of those changes would lower measured GDP, which is why productivity, as currently measured, is a poor guide to them.

None of it requires the State to decide what is produced. It requires markets whose prices include the costs they now ignore, buyers who have the right to repair and the information to choose, and financial rules that do not quietly favour one kind of borrowing over another. Those are the ordinary tools of a market economy, used deliberately. The evidence says they work unevenly, better for prices than for labels, and that the largest gains are in the few places where the present rules point the wrong way.

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 are computed by scripts/worth.py from ECB, Eurostat and EU regulatory data. The share of Portuguese bank lending to construction and real-estate firms is read from a Banco de Portugal chart and is approximate. The capital figures use the standardised approach and the minimum ratio, not the internal models or buffers most large banks apply. Several studies were read in working-paper versions. The details of Portugal’s 2023 capitalisation incentive and the withdrawal of the EU proposal were not checked in the official journals. Results are in docs/worth-results.json and the figures in docs/worth-figures.html; the downloaded sources, with a page reference for every number, are kept with the script’s data.

The cover photograph is Shoe repair shop, Rua do Poço, Ferreiras by Kolforn; CC BY-SA 4.0, 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. Growth of What — Part II: what GDP measures and the alternatives.
3. Off the Treadmill — Part III: policies that could loosen the dependence on growth.
4. The Manufacture of Wants — advertising, obsolescence and consumer credit.
5. The Stalled Hour — why Portuguese productivity stalled after 2013.

Sources

1. United Nations et al., System of National Accounts 2008, paras. 6.130 and 15.64–15.84.
2. A. Haldane, S. Brennan and V. Madouros, “What is the contribution of the financial sector: Miracle or mirage?”, in The Future of Finance: The LSE Report, 2010, pp. 3, 7, 9, 11–12 and 14.
3. T. Philippon, “Has the US Finance Industry Become Less Efficient? On the Theory and Measurement of Financial Intermediation”, American Economic Review 105(4), 2015; author’s summary, pp. 1 and 5.
4. Eurostat: food waste by stage of the food chain (env_wasfw), 2020–2024; health care expenditure (hlth_sha11_hf), 2024.
5. Directive (EU) 2025/1892 amending Directive 2008/98/EC on waste (food-waste reduction targets); author’s calculation from Eurostat.
6. OECD, Tackling Wasteful Spending on Health, 2017, Highlights, p. 3.
7. A. L. Schwartz et al., “Measuring low-value care in Medicare”, JAMA Internal Medicine, 2014; W. H. Shrank et al., “Waste in the US Health Care System”, JAMA, 2019; evidence on Choosing Wisely interventions as summarised in the research notes (secondary).
8. S. Prakash et al., Einfluss der Nutzungsdauer von Produkten auf ihre Umweltwirkung: Schaffung einer Informationsgrundlage und Entwicklung von Strategien gegen “Obsoleszenz”, Umweltbundesamt, Texte 11/2016, pp. 43 and 49.
9. J.-L. Arcand, E. Berkes and U. Panizza, “Too Much Finance?”, IMF Working Paper 12/161, p. 7 (published in Journal of Economic Growth, 2015).
10. S. G. Cecchetti and E. Kharroubi, “Reassessing the impact of finance on growth”, BIS Working Paper 381, 2012, p. 6; “Why does financial sector growth crowd out real economic growth?”, BIS Working Paper 490, 2015, pp. 3–4.
11. A. Jayadev and S. Bowles, “Guard labor”, Journal of Development Economics 79, 2006 (working-paper version); C. Leipert, work on defensive expenditures (secondary).
12. J. Bulow, “An Economic Theory of Planned Obsolescence”, Quarterly Journal of Economics 101(4), 1986, abstract.
13. A. O. Krueger, “The Political Economy of the Rent-Seeking Society”, American Economic Review 64(3), 1974, p. 294.
14. HM Treasury, Stability and Investment for the Long Term: Economic and Fiscal Strategy Report 1998, Cm 3978, pp. 5 and 19 (PDF); O. Blanchard and F. Giavazzi, “Improving the SGP through a proper accounting of public investment”, CEPR Discussion Paper 4220, 2004, pp. 4 and 8–9.
15. Ò. Jordà, M. Schularick and A. M. Taylor, “The great mortgaging: housing finance, crises and business cycles”, Economic Policy 31, 2016; NBER Working Paper 20501, pp. 2 and 38–40.
16. A. Mian, A. Sufi and E. Verner, “Household Debt and Business Cycles Worldwide”, Quarterly Journal of Economics 132(4), 2017; working paper, p. 2.
17. ECB, balance-sheet statistics (MFI loans by sector, Portugal); Banco de Portugal presentation (2018), p. 4 (shares read from a chart); IMF, current-account balance, Portugal.
18. Conselho para a Produtividade, First Report, 2019, p. 71, citing Azevedo et al. (2018).
19. Eurostat, government finance statistics (net public investment, Portugal); Banco de Portugal, macroprudential recommendation on new credit agreements (2018).
20. J. Colmer, R. Martin, M. Muuls and U. J. Wagner, study of the EU ETS and manufacturing firms (2024); A. Dechezleprêtre, D. Nachtigall and F. Venmans, study of the EU ETS (2023); meta-analysis of carbon pricing effects, as summarised in the research notes.
21. Directive (EU) 2024/1799 on common rules promoting the repair of goods; Regulation (EU) 2024/1781 on ecodesign for sustainable products, art. 25.
22. Direction interministérielle de la transformation publique (DITP), evaluation of the repairability index, 2023.
23. Código do IVA, List I, item 2.36, as amended by Lei 12/2022.
24. Assemblée nationale, information report on the application of Loi 2016-138 (Garot law), 2019.
25. Regulation (EU) 575/2013 (Capital Requirements Regulation), arts. 92, 125 and 501; author’s calculation.
26. Banco de Portugal, sectoral systemic risk buffer of 4 per cent on residential mortgage exposures, in force from 1 October 2024 (as disclosed by banks).
27. European Commission, proposal for a Council Directive on a debt-equity bias reduction allowance (DEBRA), COM(2022) 216; Estatuto dos Benefícios Fiscais, arts. 41-A and 43-D (details not checked in the official journal).
28. O. Blanchard and F. Giavazzi (2004), as source 14.