The Receiving Side

Essay · economics · October 2026

This series has looked mostly at what Portuguese households pay to the State. This essay looks at the other side: what the better-off receive from it, in pensions and through the tax code. It then sets out the strongest case against taxing capital more, and asks how much tax escapes altogether. The answers are less one-sided than either the critics or the defenders of the system usually claim.

Reading time ≈ 11 min Pensions · tax breaks · investment · evasion

The question

All In showed that taxes take a similar share of income from almost every tenth of the Portuguese, and that the richest tenth stands apart mainly through its capital income, which is lightly taxed. That essay, and the ones before it, looked at what people pay. A fair account also needs what the better-off get back, the arguments for keeping capital taxes low, and the tax that is never paid. This essay takes them in turn.

28%of all pensions and cash benefits go to the richest tenth, about the same as to the poorer half

€1.9bnof income tax forgone in 2024 through the non-habitual resident regime, 58% of all income-tax breaks

€1.89of extra investment per euro of a targeted investment tax credit; much less from rate cuts

3.6%of VAT due that was not collected in 2023, against 9.5% in the EU

Pensions

Pensions follow past pay

The State’s largest cash payment is pensions, and they go up the income scale as well as down it (Figure 1). In 2024 the richest tenth of the population, about 1.06 million people, received about €14,700 per person in pensions and other cash benefits, against €1,900 in the poorest tenth and €3,000 to €5,000 in the tenths between. The richest tenth received 28 per cent of all such benefits, about as much as the poorer half of the population together, 31 per cent.[1]

Fig. 1 — Portugal, 2024: pensions and other cash social benefits received per person a year, by tenth of the population ranked by household income per adult equivalent (about 1.06 million people per tenth). Data: Eurostat, experimental household distributional accounts; author’s calculation (see All In).

This is how a contributory pension system is meant to work. A Portuguese pension replaces roughly the same share of past wages at every level, about 72 per cent for a full career on the average wage, so higher earners get higher pensions. Many retired civil servants, under older and more generous rules, receive more still: the average pension paid by the civil servants’ fund, closed to new members since 2006, was €1,649 a month in 2025, against €674 for an old-age pension in the general scheme.[2][3] Those pensions were paid for by contributions, as What Contributions Buy describes, and the rules lean against the largest: in years of weak growth, pensions above twice the social index rise by less than inflation, and in 2026 those above twelve times it, €6,446 a month, were not raised at all.[4] But it means that a large part of what the State hands out in cash goes to households that are not poor, which is easy to forget when its spending is described as going to the poor.

Portugal is unusual in this. Of 22 European countries with comparable estimates, only Cyprus gives a larger share of its cash benefits to the richest tenth; in Spain the share is 24 per cent, in Germany 17 and in Denmark 11.[1] The OECD found that the tenth of Portuguese pensioners with the highest pensions received more than four and a half times as much as the tenth with the lowest, against about twice in the OECD on average, because the pension formula passes on more than 90 per cent of differences in pay.[5] Large pensions are still few: in December 2023 about 65,000 pensions paid more than about €3,000 a month, 2.5 per cent of all old-age, invalidity and civil-service pensions, and 61 per cent of them were civil servants’ pensions; 7,487 civil-service pensions paid more than €5,000 in 2024.[6] And pensions remain the main tool against poverty in old age: before pensions, 84 per cent of Portuguese over 65 would be below the poverty line; after them, 21 per cent.[7] Portugal has taxed large pensions specially before: a solidarity levy on pensions above €1,350 a month, lowered to €1,000 in 2014, at rates rising with the pension, raised about €380 to €460 million a year from the civil servants’ fund in 2013 and 2014; the Constitutional Court did not strike it down, and it ended in 2017.[8]

Tax breaks

Spending through the tax code

The State also spends by not taxing. In 2024 the tax it forgave through tax breaks came to €20.4 billion, 7.2 per cent of GDP. Most of it, between €10.5 and €12.3 billion depending on the measure, is the cost of reduced VAT rates on food, restaurants, medicines and other goods. In euros, households from the sixth tenth up gain most from those rates, because they spend more; relative to income they help the poor, and without them income inequality would be slightly higher.[9][10]

In income tax the breaks came to €3.4 billion in 2024 (Figure 2), and one regime accounts for most of it. The non-habitual resident regime, created in 2009, taxed qualifying work income of new residents at a flat 20 per cent for ten years and exempted most of their foreign income, including foreign pensions, which paid nothing until 2020 and 10 per cent after. It cost €1.9 billion in 2024 by the tax authority’s measure, 58 per cent of all income-tax breaks.[11] Next came disability reliefs, €590 million; lower rates in the Azores and Madeira; IRS Jovem, the exemption for workers under 35 in their first years of earnings, whose cost rose from €30 million in 2022 to €230 million in 2024; and retirement savings plans, €110 million, where the average deduction is €77 in the poorest tenth of taxpayers and €248 in the richest.[11][10] Not counted as tax breaks at all, because the law treats them as part of the normal tax, are the deductions for health, education, housing and dependants, worth another €3.9 billion, and the exemption from stamp duty on inheritances between spouses, children and parents, about €745 million.[9][12]

Fig. 2 — Portugal: income tax forgone through tax breaks, by item, 2024, million euros (economic view, by tax year). Total €3,379 million. Family deductions for health, education, housing and dependants are not counted as tax breaks. Data: Autoridade Tributária, Dossier Estatístico IRS 2022–2024, Mapa 37.

The non-habitual resident regime is the clearest case of the State giving through the tax code to people with high incomes or wealth, and its record is mixed (Figure 3). Registrations rose from about 36,000 in 2019 to 129,000 in 2024, with a rush of 40,000 in 2023 after its end was announced. The benefit was extremely concentrated: in 2024, 50 tax returns, one in a thousand of the regime’s, accounted for 24 per cent of its cost, about €420 million. Its defenders point to what the newcomers pay: the tax paid by its beneficiaries rose from €275 million in 2019 to €789 million in 2023, and the official cost compares what they paid with what they would have paid as ordinary residents, including on foreign income they might never have brought to Portugal, which overstates the revenue truly lost; no official study has weighed the two. The regime was closed to new entrants in 2024, with existing beneficiaries keeping it until 2033, and replaced by a narrower one for researchers and qualified workers, which excludes pensioners.[12][13]

Fig. 3 — Portugal: income tax forgone through the non-habitual resident regime (Tribunal de Contas, central-government view) and tax paid by its beneficiaries (Inspeção-Geral de Finanças, 2019 and 2023 only), million euros, with cumulative registrations. The tax forgone is computed by re-taxing beneficiaries as ordinary residents, which overstates the revenue actually lost. Data: Tribunal de Contas, Parecer sobre a Conta Geral do Estado de 2024, Gráfico 74; IGF, Relatório 103/2025.

The other case

The case against taxing capital more

The essays on capital income in this series point to one conclusion: capital is taxed more lightly than work, and goes mostly to the richest tenth. The strongest objection is that Portugal needs more capital, not less. Investment fell from 28 per cent of GDP in 2000 to 14.8 per cent in 2013, against 19.6 per cent in the EU, and has only recently caught up (Figure 4); capital per worker fell from about €137,000 in 2013 to €117,000 in 2025, as the economy added workers faster than machines, which The Stalled Hour identified as a main reason productivity stalled.[14] Companies are not lightly taxed: corporate tax raised 3.8 per cent of GDP in 2024, more than the EU’s 3.15, and counting both the company and the shareholder, distributed profit pays about 50 per cent, close to France and Germany.[15]

Fig. 4 — Gross fixed capital formation, per cent of GDP, Portugal and the EU, 2000–2025. Dotted lines mark the corporate tax cuts of 2014 and 2015, from 25 to 21 per cent; no study has measured their causal effect on Portuguese investment. Data: European Commission, AMECO.

What the evidence says about the remedy is more specific than the objection. Cutting the corporate tax rate is an expensive way to raise investment. The Banco de Portugal’s model finds that a permanent one-point cut raises output by about 0.1 per cent in the long run, and only if firms reinvest what they save.[16] In the United States, the 2017 cut reduced corporate tax revenue by about 40 per cent and raised investment by about 11 per cent; allowing firms to deduct investment immediately bought about half the capital for a third of the cost.[17] Targeted credits do better in Portugal too: the investment tax credit for firms in eligible regions produced about €1.89 of extra investment for each euro of credit, €2.32 in small firms and only €0.28 in large ones, and the research credit up to €1.72 of research spending per euro.[10] Taxes on dividends, on the other hand, appear to matter little for investment: when the United States cut them in 2003, investment did not respond.[18]

Mobility is the second objection, and the evidence is that it depends on who. Highly paid foreigners and star earners respond strongly to tax: Denmark’s preferential regime for foreigners roughly doubled the number of highly paid foreigners, and similar estimates exist for inventors and scientists. Resident millionaires move much less: in the United States, a 10 per cent higher top rate is associated with about 1 per cent fewer of them.[19] Wealth taxes reduce the wealth people report, through how they hold and declare it more than through migration.[19] Economic theory once recommended no tax on capital at all; later work has shown that the conclusion rests on assumptions that do not hold, and most economists now accept some tax on capital, without agreeing how much.[20] The fair summary is that the case against higher capital taxes is strong for the corporate rate and for mobile foreigners, which is the logic of the non-habitual resident regime, and weak for the interest, dividends, gains and inheritances of residents.

Evasion

The tax that is never paid

A last part of the picture is tax that should be paid and is not. On this Portugal does better than its reputation (Figure 5). The share of VAT due that was not collected fell from 15 per cent in 2013 to 3.6 per cent in 2023, one of the four lowest in the EU and a third of the EU average; the European Commission credits electronic invoicing.[21] By comparison, the revenue the State chooses to forgo through reduced rates and exemptions is about 29 times larger. Estimates of the shadow economy vary enormously with the method, from about 7 per cent of GDP, which official GDP already includes, to 34 per cent in one Portuguese index; the most cited series puts Portugal at 15.7 per cent in 2022, slightly below the EU average.[22]

Fig. 5 — Portugal and the EU: estimated shadow economy as a share of official GDP (2022; model estimate, unweighted EU average; other methods give from about 7 to 34 per cent for Portugal), undeclared work as a share of private-sector labour input (2019), and the VAT compliance gap as a share of the VAT that should have been paid (2023). Data: Schneider and Asllani (2022), European Parliament; European Labour Authority (2023); European Commission, VAT gap in Europe, 2025.

Where undeclared work exists it is concentrated. About 7.5 per cent of private-sector labour was undeclared in 2019, against 11 per cent in the EU, but among the self-employed the share was about 21 per cent, against 5 per cent among employees; studies of spending suggest the Portuguese self-employed under-report about a tenth of their income, a low figure by European standards.[23] An audit found that up to 60 per cent of tenants had no rental contract registered with the tax authority.[24] Portugal publishes no estimate of the gap in income tax or corporate tax.

The balance

What the ledger leaves out

Seen from the receiving side, the picture of a State that takes from the rich and gives to the poor needs three corrections. The largest cash transfer, pensions, flows up the income scale as well as down it, because it follows past pay; that was paid for, but it is still where much of the money goes. The tax code gives most generously, per person, to a few: the non-habitual resident regime cost more than all other income-tax breaks together, and a quarter of it went to fifty returns. And the tax that escapes is smaller than often claimed, except among the self-employed and landlords.

The case against taxing capital more deserves its weight where the evidence supports it: Portugal does need more investment, corporate tax is already high by European standards, and the highly mobile respond to tax. But the evidence also suggests that investment responds better to targeted incentives than to lower rates, and that taxing the interest, dividends, gains and inheritances of residents more like work would cost little in investment. Which way to go is a political choice; the evidence narrows it, without making it.

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/receiving.py. Pensions and benefits by tenth come from Eurostat’s experimental household distributional accounts, as in All In. Tax-break figures come from the tax authority, the Tribunal de Contas and the Ministry of Finance, which measure them in different ways (by tax year or by cash year, for general or central government); the essay says which it uses. The cost of the non-habitual resident regime is a static estimate that overstates the revenue lost. Shadow-economy estimates depend heavily on method and are shown as a range. The evidence on capital taxation and investment comes from peer-reviewed studies and official evaluations; no study has measured the causal effect of Portugal’s 2014–15 corporate tax cut. Results are in docs/receiving-results.json; the downloaded sources, with page or table for every number, are kept with the script’s data.

The cover photograph is Praça do Comércio. Lisboa... by Miguel Ángel García; CC BY 2.0, via Wikimedia Commons, cropped.

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

Related essays on this blog

1. All In — everything each tenth receives and every major tax it pays.
2. What Contributions Buy — what social contributions buy back in pensions and insurance.
3. Earned and Unearned — how work and ownership share income, and how each is taxed.
4. The Stalled Hour — why output per hour stopped growing.
5. The Sustainability Story — how Portugal’s pensions are financed.
6. Work, Time and Money — the reading guide to the whole series.

Sources

1. Eurostat, Household distributional accounts, experimental statistics, July 2026 version, Portugal, 2024 (social benefits other than transfers in kind, received, by tenth); author’s per-person amounts.
2. OECD, Pensions at a Glance 2025, Portugal: gross replacement rates at 0.5, 1 and 2 times the average wage.
3. Caixa Geral de Aposentações, reports and accounts; Conselho das Finanças Públicas, Relatório 04/2026; GEP-MTSSS, Síntese de Informação Estatística da Segurança Social, May 2025, p. 5.
4. Lei 53-B/2006, art. 6; pension updates for 2026 by band (Portaria 480-B/2025/1).
5. OECD, OECD Reviews of Pension Systems: Portugal, 2019, pp. 33 and 70 (Figure 3.10).
6. Caixa Geral de Aposentações, Relatório e Contas 2023 and 2024, Quadro 16; IGFSS, Conta da Segurança Social 2023, Parte II, Gráfico 24, pp. 204–205; author’s sum (thresholds of €3,000 for civil-service pensions and six times the social index, €2,883, for the general scheme).
7. Eurostat, at-risk-of-poverty rate before social transfers, pensions included and excluded (ilc_li09, ilc_li10, ilc_li02), Portugal, survey 2025, persons aged 65 and over.
8. Tribunal Constitucional, Acórdãos 187/2013 and 572/2014; Caixa Geral de Aposentações, reports and accounts 2013–2016; Conselho das Finanças Públicas, Relatório 02/2015.
9. Autoridade Tributária e Aduaneira, Relatório da Despesa Fiscal 2024, June 2025, Quadros 1, 5 and 23; Tribunal de Contas, Parecer sobre a Conta Geral do Estado de 2024, pp. 196–203.
10. U-TAX (Autoridade Tributária), Relatório de Avaliação da Despesa Fiscal em Portugal, June 2025: reduced VAT rates by income (section III), retirement savings deduction by decile, and evaluations of RFAI (Table 81, p. 298) and SIFIDE (Tables 67–68, p. 246).
11. Autoridade Tributária e Aduaneira, Dossier Estatístico IRS 2022–2024, Mapa 37 (income tax expenditure by item, economic view).
12. Tribunal de Contas, Parecer sobre a Conta Geral do Estado de 2024, October 2025, pp. 196–201 and Gráfico 74.
13. Inspeção-Geral de Finanças, Auditoria aos Regimes Fiscais de Ex-Residente e Residente Não Habitual, Relatório 103/2025, synthesis, April 2026; Lei 2/2020, art. 329; Lei 82/2023 and Estatuto dos Benefícios Fiscais, art. 58-A.
14. European Commission, AMECO: gross fixed capital formation and net capital stock per person employed, Portugal and EU, 2000–2025.
15. European Commission, Data on Taxation Trends 2026 (corporate income tax revenue); OECD Tax Database, Table II.4.
16. Banco de Portugal, Boletim Económico, December 2024, box on the taxation of company income, pp. 56–61.
17. G. Chodorow-Reich, O. Zidar and E. Zwick, “Lessons from the Biggest Business Tax Cut in US History”, Journal of Economic Perspectives 38(3), 2024; E. Zwick and J. Mahon, “Tax Policy and Heterogeneous Investment Behavior”, American Economic Review 107(1), 2017.
18. D. Yagan, “Capital Tax Reform and the Real Economy: The Effects of the 2003 Dividend Tax Cut”, American Economic Review 105(12), 2015, p. 3531.
19. H. Kleven, C. Landais, E. Saez and E. Schultz, “Migration and Wage Effects of Taxing Top Earners”, Quarterly Journal of Economics 129(1), 2014; C. Young, C. Varner, I. Lurie and R. Prisinzano, “Millionaire Migration and Taxation of the Elite”, American Sociological Review 81(3), 2016; M. Brülhart et al., “Behavioral Responses to Wealth Taxes”, AEJ: Economic Policy 14(4), 2022.
20. L. Straub and I. Werning, “Positive Long-Run Capital Taxation: Chamley-Judd Revisited”, American Economic Review 110(1), 2020; E. Saez and S. Stantcheva, “A simpler theory of optimal capital taxation”, Journal of Public Economics 162, 2018.
21. European Commission, VAT gap in Europe: Report 2025, annex tables 204–205 and Portugal country report, p. 13; SWD(2025) 421, Mind the Gap: Portugal, pp. 3–4.
22. F. Schneider and A. Asllani, Taxation of the Informal Economy in the EU, European Parliament, 2022, Table 2.1; A. Fernandes, HIVA KU Leuven working paper, 2022 (non-observed economy in national accounts); University of Porto, OBEGEF index, June 2023.
23. J. Franić, A. Horodnic and C. Williams, Extent of Undeclared Work in the European Union, European Labour Authority, 2023; ELA, factsheet on undeclared work in Portugal, 2023; A. Kukk, A. Paulus and K. Staehr, “Cheating in Europe”, International Tax and Public Finance 27, 2020, Table 2.
24. European Commission, SWD(2025) 421, Mind the Gap: Portugal, pp. 7–8.