What We Found

Conclusions · economics · September 2026, revised October 2026

Thirty-seven essays on work, time and money, most of them about Portugal, reduced to what they found. A century of productivity went mostly into income, not time, and new technology has not changed that so far. Portugal’s core problem is an hour of work that stopped becoming more productive, despite the fastest rise in schooling in Europe; its taxes take a similar share from almost everyone, but treat work more heavily than capital; the heaviest bill on the young is the price of a home, and many of them leave. Growth is an arrangement rather than a law; and more free time, less pressure on the planet and more satisfied people are compatible, on conditions that Portugal does not yet meet.

Reading time ≈ 11 min Twelve findings → what is uncertain → what follows

The question

The series began with a prediction Keynes made in 1930: that his grandchildren would be four to eight times richer and would work fifteen hours a week. It asked where the time went, and followed the question into wages and profits, taxes and the State’s accounts, pensions, schooling, firms, housing, children, emigration and immigration, and what growth is for. Behind it was one question: can an economy and its institutions give people more free time, put less pressure on the planet and leave them more satisfied? This page sets out what the essays found, in twelve points, each linked to the essay that makes the case. The reading guide, Work, Time and Money, lists them all.

Ten findings

What the evidence says

1. The productivity went into income, not time. Keynes was right about wealth and wrong about hours: rich countries became as much richer as he predicted, and took most of the gain as more consumption, and in America since 1980 much of it went to the top. Portugal did the same: the household life of 1980 now costs about half the hours of work it did, and the normal week fell only from about 45 hours to 40. Artificial intelligence has not changed the pattern so far: workers who save time with it put almost all of it back into work, and recorded hours and pay have not moved. Technology makes shorter hours possible; it has never decided them by itself. (The Fifteen-Hour Week, Where the Hours Went, The Portuguese Basket, The Machine and the Hour)
2. The time that was taken came in two hidden forms. Longer retirements, as people lived longer and, until about 2000, left work earlier; and less housework, which fell for women in every country with comparable surveys, by more than their paid work rose, so their leisure held steady. Neither shows up in statistics of paid hours. (The Long Retirement, The Unpaid Ledger)
3. Who owns matters as much as how much there is. A quarter to two-fifths of national income goes to owners rather than workers, and ownership is concentrated: of the Portuguese profits that reach households, about five-sixths go to the richest tenth, and more go to owners abroad than to all Portuguese households. (Earned and Unearned, Who Gets the Profits)
4. Taxes take a similar share from almost everyone; what differs is what is taxed. The richest tenth pays about two-thirds of the income tax, but income tax is only a fifth of revenue: counting social contributions and VAT, it pays about 31 per cent of the total, and the salaried middle 40 per cent pays about 45, roughly its share of income. Counted against everything each tenth receives, taxes take a nearly constant 27 to 34 per cent. What sets the top apart is capital income, taxed far more lightly than work, while salaried workers face some of the OECD’s highest marginal rates. Social contributions are not simply a tax: for a middle earner they buy back, on average, about four-fifths of their value in pensions and insurance. On the receiving side, pensions follow past pay, so the richest tenth receives about as much in cash benefits as the poorer half, and the largest income-tax break went heavily to a few. Counting public services, the poorer half receives more than it pays. (The State’s Ledger, The Salaried Middle, What Contributions Buy, All In, The Receiving Side)
5. Portugal’s core problem is the hour of work, and schooling alone did not fix it. For thirty years an hour of Portuguese work has produced about two-thirds of the European average, and since 2013 its output has barely grown, 0.17 per cent a year. Over the same years Portugal had the fastest rise in schooling in Europe: adults with at most nine years of school fell from 81 to 36 per cent, and young adults are now as schooled as young Europeans and almost as skilled. On standard assumptions that alone should have added three times the growth that came. The stall happened inside industries, as the economy added workers faster than machines, in small firms often run by owners with little schooling. (The Two-Thirds Country, The Stalled Hour, What Schooling Bought)
6. What firms face is real, but it is not most of the gap. Portugal still has some of the OECD’s strictest dismissal rules, slow tax and administrative courts, protected professions, slow permits, and the EU’s highest share of firms that say success needs political connections. But Portuguese firms complain less than other Europeans about red tape and labour rules, ordinary business disputes now take about as long as in the median EU country, credit is no longer scarce, and the small size of firms explains only about a sixth of the productivity gap: large Portuguese firms are far behind large European ones too. The OECD’s own simulations give these reforms together about half of what a later retirement alone would add. (The Business Side)
7. The floor of pay rose; the middle did not, and many leave. The minimum wage rose about 40 per cent above inflation after 2015 while employment grew, though the evidence on what it did to jobs is not settled; it now sits closer to the typical wage than anywhere in the EU and one employee in five earns exactly that. Above it, pay has stagnated: graduates still earn twice as much as workers with secondary school, but the average graduate earns less in real terms than in 2010, and a German graduate earns 1.6 times as much after adjusting for prices. More than 600,000 people left between 2011 and 2016, and 65,000 to 80,000 still leave each year; one person born in Portugal in six lives abroad, and graduates leave about as often as everyone else. (The Floor, The Leavers)
8. The heaviest bill on the young is a home. Real house prices more than doubled after 2013 while young people’s pay did not, transferring wealth every month from those who must buy or rent to those who already own. The main cause was not foreign buyers, funds or tourist flats, which mattered in a few places, but many more households, mostly from immigration, meeting a building industry that had shrunk by half. Cheaper homes further out are paid for in time: one in six commuters into Lisbon spends more than an hour each way. (The Rent, The Bidders, Hours on the Road)
9. The State’s largest bills are not the debt. The public debt is falling and back near the euro-area average. The larger burdens left to the young are public capital allowed to wear out for a decade, and a pension system kept affordable by paying the next pensioners a much smaller share of wages, about a third of the average wage by 2070 against about a half now. That is a policy choice, and there are others. A bill already being paid, but not by the State, is care for the very old: their number doubles by 2060, families, mostly women, give most of the care, and households pay about half of what is spent on it. And more than half of school teachers are 50 or older, with few young ones behind them. (The Inheritance, The Sustainability Story, Who Cares, What Schooling Bought)
10. The Portuguese have fewer children than they want, and immigrants are carrying the numbers. Most want two children and have one and a half, because they start late and in insecure jobs. Immigrants, now about one in seven residents, are young, work as much as the Portuguese, pay far more into Social Security than they take out for now, and account for most of the recent growth in jobs, contributions and births, though they are paid less and more often poor. Portugal now trains graduates for other economies and fills many of its lower-paid jobs with newcomers. (Fewer Than Wanted, The Arrivals, The Leavers)
11. Growth is an arrangement, not a law, and GDP is not wellbeing. Economies do not need growth the way bodies need food, but jobs, debts, pensions and politics were built on it, and Portugal’s stagnant decade after 2000, with unemployment at 17 per cent and mass emigration, shows what happens when it stops by accident. Each dependence can be loosened, at a cost someone bears. GDP measures production: by most other measures Portugal lives better than its GDP suggests, and some of what is produced is waste, concentrated in food, low-value health care, products that do not last and credit that went into houses; it can be reduced by prices that include the costs, rights for buyers and rules on credit, without planning. (The Treadmill, Off the Treadmill, Growth of What, Worth Producing, The Manufacture of Wants)
12. More time, less pressure on the planet and more satisfaction are compatible, on conditions. Shorter hours raise satisfaction modestly, and the effect lasts; they cut emissions mainly by cutting income; the climate depends far more on how energy is made and priced. Many Portuguese want shorter hours, 39 per cent against 22 in the EU, but few can afford less pay, and the institutions that let people take time elsewhere are narrow or absent in Portugal. The goals fit together when productivity grows and carbon is priced; they conflict when productivity stalls. Portugal’s emissions have fallen by two-fifths, mostly by closing coal plants; the rest, in transport, farming and homes, has never fallen at the pace needed. And beyond income, what separates Portugal from the most satisfied countries is mostly mental health, trust and the lives of the old. (The Three Goals, The Fifth Day, The Duty to Work, The Hard Part, Beyond Money and Hours)

Uncertain

What the evidence cannot yet settle

Several conclusions rest on thin ground. Portugal has had no time-use diary survey since 1999, so how much unpaid work and leisure the Portuguese have now is unknown, and so are the hours of care that families give. Three surveys disagree on how satisfied the Portuguese are compared with other Europeans. Whether shorter hours raise life satisfaction, not just satisfaction with work and leisure, is not well measured, and the four-day-week results come from firms that chose to try. Whether rich countries can cut emissions fast enough while growing is disputed. Whether immigrants will stay, move up into jobs that match their qualifications and pay in more than they draw over a lifetime cannot be known for decades. Nobody counts emigrants at the border, and the three measures differ by a factor of two. The evidence on what artificial intelligence does to hours and pay covers only its first two years. Why Portuguese schooling has not shown up in productivity has three possible explanations that the data cannot separate. And the Portugal scenario to 2050 is arithmetic, not a forecast: it assumes productivity growth the country has not had since 2013.

What follows

For Portugal

Read together, the essays point to a short list, roughly in order of how much depends on it:

1. Raise productivity: invest in capital, let firms grow, and keep improving the education of those who run them. Almost every other goal is easier with it, including keeping the young who now leave.
2. Fix what holds firms back where the evidence is clearest: faster tax and administrative courts, unfair-dismissal awards nearer the European norm, open entry to the professions and retail, and permits with real deadlines.
3. Build homes where the jobs are, faster, and stop subsidising demand that ends up in prices.
4. Price and regulate carbon, in transport and heating as well as power, returning the revenue to the households that pay most, so that the climate does not depend on how many hours people work.
5. Give every worker a right to reduce hours and return, and consider a paid time credit and a partial pension, so that time is not only for those who can afford it.
6. Decide who pays for the care of the very old: protect family carers with pension credits and paid leave, and expand care at home.
7. Train and recruit the teachers who will replace the half now over 50, before they retire.
8. Treat mental health care and institutions people can trust as part of the standard of living.
9. Make pensions and public debt robust to slower growth: reduce debt while growth lasts, apply the pension rules as written, and decide in advance who adjusts.
10. Tax income from capital and what accumulates, wealth and inheritance, more like work, and steer credit away from bidding up existing assets.
11. Measure progress by more than GDP, including satisfaction with time and finances, which INE already collects.

A caveat

How the series leaned

Most essays were written to be neutral and cite their critics, but a few argued a position: The Duty to Work, Earned and Unearned, The Manufacture of Wants and, in part, The Sustainability Story. And the questions themselves were chosen mostly from one side: waste, the manufacture of wants, the need for growth. The Three Goals tried to correct for that by stating the case for growth at full strength: growth is how Portugal cut infant mortality from 78 to 2.5 per thousand, it pays for the clean-energy transition and for an ageing population’s care, and poorer countries have good reason to want income. The later essays on the business side, the minimum wage, emigration and artificial intelligence set out the case of employers and of the critics of each policy at full strength before testing it. The conclusions above are meant to stand after those corrections. Readers who disagree will find the data and the sources behind every number in each essay.

On method and tools

The essays were written collaboratively with Claude Opus 5.5 (Anthropic): human specification, editorial direction and critical review; machine data research, analysis and drafting. This page summarises their conclusions and adds no new data; every number here is sourced in the essay linked beside it.

The cover photograph is Lisbon Sunset by mhx; CC BY-SA 2.0, via Wikimedia Commons, cropped.

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