The Business Side

Essay · economics · October 2026

Portuguese business is unusually small: more than two jobs in five are in firms with fewer than ten people. Employers point to rules that make firing expensive, slow tax courts, high taxes and a state that favours the well connected, and some of these complaints are borne out by the data. But firms complain less than other Europeans about red tape and labour rules, much has improved since 2012, and even large Portuguese firms produce far less per worker than large firms elsewhere. This essay looks at what firms face, and how much it explains.

Reading time ≈ 10 min Small firms · firing · courts · rules · money · the balance

The question

This series has looked at work mostly from the side of workers and the state. The Stalled Hour found that Portuguese output per hour has barely grown since 2013, with low investment and many small firms among the reasons, and Who Gets the Profits looked at the people who own firms. This essay takes the firms’ side: what they say holds them back, what the data show, and how much fixing it would add. The case made by business groups is given here at its strongest, and then tested.

42%of business jobs are in firms with fewer than ten people; EU 30%

20 monthsof pay: the average award for an unfair dismissal

861 daysto clear an administrative or tax case in 2024, second slowest in the EU

~15%of the productivity gap with the EU explained by the small size of firms

Small firms

Many small firms, and large ones that lag too

In 2023, the average Portuguese business had 3.3 people, against 4.9 in the EU. Firms with fewer than ten people held 42 per cent of business employment, against 30 per cent in the EU and 19 per cent in Germany; firms with 250 or more held 23 per cent, against 37 per cent (Figure 1).[1] Firms have also been shrinking for a long time: among firms with employees, the average fell from 17.7 workers in 1986 to 8.9 in 2009, a shift the authors could only half explain by better counting of small firms, the move to services and the end of state monopolies.[2]

Fig. 1 — Share of persons employed in the business economy in firms with fewer than ten persons and in firms with 250 or more, 2023, per cent. Data: Eurostat, structural business statistics by size class (sbs_sc_ovw).

Small firms produce less per worker everywhere, so a country of small firms is poorer. But size is not the main reason for Portugal’s gap. In 2023 a Portuguese micro firm produced €20,100 per person, less than half the EU’s €42,100; a Portuguese firm with 250 or more people produced €53,500, about three-fifths of the EU’s €86,800. With the EU’s mix of firm sizes, Portuguese output per worker would be about 12 per cent higher, closing only about 15 per cent of the gap with the EU; the rest is that Portuguese firms of every size produce less.[3]

Why do firms stay small? Some rules change at a size threshold: a lower corporate tax rate on the first €50,000 of profit for small firms, a statutory audit and an anti-corruption programme from about 50 employees. The OECD warns that the small-firm tax rate may discourage growth.[4] But Portuguese firms do not bunch below these thresholds, as French firms do below 50 employees; the study that tracked the shrinking concluded that no single threshold mattered and suggested instead an accumulation of small costs.[2] The education of owners matters too: firms founded by more educated entrepreneurs start larger and grow faster, and Portugal has one of the OECD’s highest shares of managers without secondary schooling, as What Schooling Bought described.[4]

Firing

Rules on dismissal: still strict, less than before

The business case is strongest here. On the OECD’s index of protection for permanent workers against dismissal, Portugal was the strictest of all its members every year from 1990 to 2013 (Figure 2). The reforms of 2011 to 2013, under the bailout, cut severance pay for collective and economic dismissals from 30 days of pay per year of service to 12, and made dismissal for unsuitability easier; Portugal is still the second or third strictest.[5][6] The Agenda do Trabalho Digno of 2023 raised that severance to 14 days and to 24 days at the end of fixed-term contracts, and barred firms from outsourcing work done by people dismissed in the previous year.[7] According to the OECD, courts award an average of 20 months’ pay for an unfair dismissal, against about 12 in France and 7 in the Netherlands, and often order reinstatement.[8]

Fig. 2 — OECD index of employment protection for regular workers (individual and collective dismissals), version 1, 0 to 6, Portugal, selected countries and the OECD average, 1990–2019; the OECD average is the simple mean of members with data. The index has not been updated for the 2023 changes. Data: OECD, Employment Protection Legislation database.

The evidence on what this costs is mixed. Across OECD countries, stricter dismissal rules slow productivity growth in the industries that most need to adjust their workforce.[9] A Portuguese natural experiment points the same way: when a 1989 law spared firms of 20 or fewer workers from most of the procedure for dismissal for cause, their performance improved, although hiring and firing did not change and wages fell.[10] But the same cross-country study finds that easing rules on temporary contracts, the route Portugal took for years, does nothing for productivity; it has left young Portuguese far more often on temporary contracts than other Europeans (23 against 17 per cent at ages 25 to 34 in 2025).[9] In the OECD’s simulations, cutting unfair-dismissal awards from 20 to about 12 months would raise GDP per person by 0.6 per cent in the long run.[11] A broader labour reform proposed by the government in 2025 had not become law by October 2026.

Courts

Slow where it matters for firms

Portuguese justice was notoriously slow. In 2012, more than 1.2 million civil enforcement cases, most of them firms trying to collect debts, were pending; by March 2025 there were 343,000, 72 per cent fewer, though part of the fall came from closing cases where no assets could be found.[12] Civil and commercial cases now take about as long as in the median EU country, 263 days at first instance in 2024 against 248, and Portugal’s appeal courts are among the fastest in the EU (Figure 3).[13]

Fig. 3 — Disposition time (cases pending at the end of the year divided by cases resolved, times 365) of first-instance litigious civil and commercial cases and of administrative cases (in Portugal including tax), Portugal and the median EU member state, 2016–2024, days. The 2024 rise in Portugal reflects a surge of residence-permit cases in Lisbon. Data: European Commission, EU Justice Scoreboard 2026, data workbook.

The weak spot is the administrative and tax courts, where firms dispute tax assessments and decisions of the state. They took 861 days to clear a first-instance case in 2024, second slowest in the EU, inflated that year by a wave of residence-permit cases; even in 2023 they took 597 days, nearly twice the EU median. A tax case resolved in 2022 had lasted on average 47 months.[13][14] Firms notice: 65 per cent of Portuguese firms call the difficulty of recovering debts a serious problem, against 43 per cent in the EU.[15] In the OECD’s simulations, more efficient justice is the business reform with the largest payoff, about 0.8 per cent of GDP per person in the long run.[11]

Rules

Good laws, slow practice, and connections

On the OECD’s index of product-market regulation, Portugal was 29th of 38 countries in 2023. The ranking is pulled down by barriers to entry into professions and retail, where Portugal is 37th, and by weak assessment of how new laws affect competition. On paper, its licensing regime is among the best in the OECD, fourth of 38.[16] In practice, a building permit took between 273 days in Funchal and 548 in Coimbra in 2023, and complying with taxes took a mid-sized firm 243 hours a year in 2018, against 161 in the EU.[17] A planning reform of 2024 introduced tacit approval when municipalities miss deadlines, and its revision, with much shorter deadlines, takes effect today; the OECD notes that earlier tacit-approval rules had “limited effectiveness”.[11]

What firms themselves say is revealing (Figure 4). Asked in 2024 what is a serious problem for doing business, Portuguese firms named tax rates far more than any other EU country (88 per cent against 66), and debt recovery, nepotism and corruption well above the EU average. Sixty-five per cent agreed that “the only way to succeed in business is to have political connections”, the highest share in the EU. But they complained less than the EU average about administrative complexity, fast-changing laws and labour regulations.[15] In the European Investment Bank’s survey, more Portuguese than EU firms call regulation an obstacle to investment (82 against 69 per cent), and the joint most common obstacle, with uncertainty, is a shortage of skilled staff (88 per cent).[18]

Fig. 4 — Firms answering that each issue is a very or quite serious problem for their company when doing business, Portugal and the EU27, 2024, per cent (505 Portuguese firms). Data: European Commission, Flash Eurobarometer 543, “Businesses’ attitudes towards corruption in the EU”.

Taxes deserve a note. Corporate income tax is being cut from 21 per cent to 17 per cent by 2028, with 15 per cent on the first €50,000 of profit of small firms, but with surcharges on large profits and the tax on dividends, the combined rate on distributed profits is still about 49 per cent, and corporate tax revenue is 3.8 per cent of GDP, against 3.2 in the EU.[19] The Banco de Portugal’s model suggests that each point of tax cut adds only about 0.1 per cent to output in the long run, and only if the money is reinvested; The Receiving Side looked at the case for and against taxing capital less.

Money

Less debt, little equity

In 2012, Portuguese companies owed loans and bonds worth 120 per cent of GDP, among the highest in Europe; micro firms as a group paid more in interest than their operating earnings. By 2025 the figure was 65 per cent, below the euro area’s 72 (Figure 5), and the share of equity in companies’ assets had risen from 28 to 44 per cent.[20] Small firms that paid nearly three points more than euro-area firms for a loan in 2012 now pay about half a point more, and access to finance is a smaller problem for Portuguese small firms than for those in the euro area.[21]

Fig. 5 — Debt of non-financial corporations (loans and debt securities, consolidated), per cent of GDP, Portugal, Spain, Italy and the euro area, 2000–2025. Data: Eurostat, financial balance sheets (nasa_10_f_bs).

Two problems remain. One firm in four still has negative equity, a share almost unchanged since 2006; it is a problem of micro firms (29 per cent of them, against 3 per cent of large firms).[20] In 2013, 44 per cent of bank credit went to firms with very low productivity, and about a fifth to “zombie” firms that could not cover their interest, keeping capital locked in weak firms.[22] And there is little risk capital: venture capital was 0.05 per cent of GDP in 2025, against 0.08 in Europe, and listed shares of Portuguese companies are worth 19 per cent of GDP, against 51 in the euro area.[23]

The balance

Real obstacles, but not the main story

The case made by business is partly right. Dismissal is still expensive, tax and administrative courts are slow, entry to professions and retail is protected, permits take long, and many firms believe that success depends on connections. Each of these has a cost, and fixing them would help. But the evidence does not support the idea that they are the main reason Portugal is poorer. Firms complain less than other Europeans about red tape and labour rules; courts for ordinary business disputes now work about as fast as the EU median; finance is no longer scarce; and firm size explains only a small part of the productivity gap.

The OECD’s own arithmetic makes the point. Its simulations give about 1.5 per cent more GDP per person in the long run from faster courts, cheaper dismissal and freer entry to professions and retail, together. Raising the effective retirement age gives 3.1 per cent and halving the gender gap in work 2.5.[11] What holds Portuguese firms back is mostly within them: little capital per worker, managers with less schooling, and slow adoption of new technology, as The Machine and the Hour found for AI. Better rules would make those easier to fix; they will not fix them on their own.

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/business.py. Firm-size shares use Eurostat’s new structural business statistics, which break with the series before 2021. The productivity counterfactual applies EU employment shares by size class to Portuguese output per person in each class. Disposition time is a ratio of pending to resolved cases, not a measured length of cases. The OECD dismissal index is version 1, which runs to 2019; later values repeat 2019. Survey answers are from firms’ own perceptions. Results are in docs/business-results.json; the downloaded sources, with page or table for each number, are kept with the script’s data.

The cover photograph is Porto de Leixões 2017 by Niels Johannes; CC BY-SA 4.0, via Wikimedia Commons, cropped.

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

Related essays on this blog

1. The Stalled Hour — why Portuguese output per hour stopped growing.
2. Who Gets the Profits — who owns Portuguese firms.
3. What Schooling Bought — the education boom and what it paid.
4. The Floor — the minimum wage and the flattened pay scale.
5. Work, Time and Money — the reading guide to the whole series.

Sources

1. Eurostat, enterprises by size class (sbs_sc_ovw), business economy excluding public administration, 2023.
2. S. Braguinsky, L. Branstetter and A. Regateiro, “The Incredible Shrinking Portuguese Firm”, NBER Working Paper 17265, 2011, abstract, pp. 11, 25 and Table 1.
3. Eurostat, structural business statistics, apparent labour productivity by size class, 2023; author’s counterfactual.
4. OECD, Economic Surveys: Portugal 2026, pp. 42–52; F. Queiró, “Entrepreneurial Human Capital and Firm Dynamics”, Review of Economic Studies 89(4), 2022.
5. OECD, Employment Protection Legislation database (DSD_EPL), version 1, 1985–2019, and version 4, 2013–2025.
6. Lei 23/2012, of 25 June, art. 366; Lei 69/2013, of 30 August, arts. 344–345 and 366.
7. Lei 13/2023, of 3 April (Agenda do Trabalho Digno), amending Labour Code arts. 338-A, 344–345 and 366.
8. OECD, Economic Surveys: Portugal 2026, p. 70.
9. A. Bassanini, L. Nunziata and D. Venn, “Job protection legislation and productivity growth in OECD countries”, Economic Policy 24(58), 2009 (IZA DP 3555, pp. 22–25); Eurostat, temporary employees by age (lfsa_etpgan), 2025.
10. P. S. Martins, “Dismissals for Cause: The Difference That Just Eight Paragraphs Can Make”, Journal of Labor Economics 27(2), 2009 (IZA DP 3112).
11. OECD, Economic Surveys: Portugal 2026, Table 1.8, p. 48, and pp. 52 and 115.
12. DGPJ, Destaque estatístico 139, July 2025, pp. 1–2; OECD, Justice Transformation in Portugal, 2020, pp. 34 and 63.
13. European Commission, 2026 EU Justice Scoreboard, COM(2026) 273, Figs. 5–8 and data workbook.
14. DGPJ, Destaque estatístico 87, 2023, pp. 2–4 (administrative and tax courts, 2022).
15. European Commission, Flash Eurobarometer 543, Businesses’ attitudes towards corruption in the EU, April 2024, Volume A.
16. OECD, Product Market Regulation indicators 2023 (DSD_PMR, version 1.3); OECD, “PMR indicators: How does Portugal compare?”, 2024, pp. 1–5.
17. World Bank, Subnational B-READY in Portugal 2024, pp. 20 and 28; World Bank, Doing Business 2020, Portugal profile, p. 4; PwC and World Bank, Paying Taxes 2020, EU and EFTA fact sheet, p. 3.
18. European Investment Bank, EIB Investment Survey 2025: Portugal overview, pp. 2 and 19.
19. Código do IRC, art. 87, as amended by Lei 64/2025; OECD Tax Database, Table II.4; Eurostat, tax revenue by type (gov_10a_taxag), 2024; Banco de Portugal, model simulation as cited in The Receiving Side.
20. Eurostat, financial balance sheets (nasa_10_f_bs); Banco de Portugal, Central Balance Sheet Database (BPstat), 2006–2024.
21. European Central Bank, MFI interest rate statistics (MIR) and Survey on the Access to Finance of Enterprises (SAFE), 2009–2026.
22. N. Azevedo, M. Mateus and Á. Pina, “Bank credit allocation and productivity”, Banco de Portugal Working Paper 25/2018, p. 11 and abstract.
23. Invest Europe, Investing in Europe: Private Equity Activity 2025, May 2026, p. 58; Eurostat, financial balance sheets, listed shares (F511), 2025.