Why Are We Still Working This Hard? Productivity Gains, Wages, and Ownership

The central claim is that American business produces in sixteen hours what took forty hours in 1973, yet Americans work more hours per person than they did then. The essay “Why Are We Still Working This Hard?” argues that the missing time was not lost to technology but to an institutional failure: the wage is asked to be both a market price and the main ticket by which people receive a share of output, and those two jobs have been pulling apart for fifty years.

The productivity–workweek gap is real in the essay’s accounting

The essay states that an hour of work in American business now makes two and a half times what it made in 1973, based on Bureau of Economic Analysis input–output tables. For every forty hours once needed, twenty-four have been freed. The forty-hour week was written into U.S. law by the Fair Labor Standards Act of 1938 and has not changed. Measured per head of population, hours worked in American business in 2024 were 10 percent higher than in 1973.

The essay frames this against John Maynard Keynes’s 1930 prediction of “three-hour shifts or a fifteen-hour week.” Keynes is usually said to have been right about wealth and wrong about leisure. The essay argues he was right about both: U.S. output per person in 2024 was 7.8 times its 1930 level, so the fifteen-hour week arrived in terms of output, but not in the workweek.

The author says the numbers are American because the United States publishes the accounts needed for the analysis, and notes living in Sweden while expecting a similar story there. The essay links to a formal paper on SSRN and says the model economy is joint work with Johan Båge.

A wage is a price, a ticket, and a place

The essay’s analytical core is that a wage does three jobs. It is a price that tells businesses what an hour costs and tells people which work is wanted. It is a ticket that gives nearly everyone a share of what is made. And it is a place: a shape to the day, people to be among, a sign that somebody needs you.

One thing can do three jobs only while the jobs pull in the same direction. The essay argues that industrial-age machines were strong and stupid: they could lift, spin, stamp, and haul, but could not see, judge, or decide. Nearly everything worth making took both machine power and human judgment, so every new machine needed people beside it, and machine owners had to bid for those people. The price of an hour rose with everything else, and a ticket made of that price was a good ticket.

The essay claims that the price and the ticket have been pulling apart for fifty years, and that each is damaging the other. The driver, it argues, is technological change: machines now can see, read, write, and decide, making it harder each year to name work safely out of their reach.

Where the missing hours went: eighteen never reached paychecks, six arrived as goods

The essay reports that until 1973, output per hour and what an hour of ordinary pay could buy rose together. Between 1950 and 1973, output per hour rose 81 percent and ordinary pay rose 66 percent. “Ordinary pay” is defined as the hourly pay of private-payroll workers who are not managers or supervisors, measured against prices excluding housing.

After 1973 the two series parted. Since 1973, output per hour has risen two and a half times, while an hour of ordinary pay buys 19 percent more. To buy what forty hours of ordinary pay bought in 1973 now takes thirty-four hours; making it takes sixteen. The eighteen hours in between went somewhere else.

The essay says about a fifth of the gap went to owners of machines, and about two-fifths is pay spreading out among workers, with the top pulling away from the ordinary worker. The rest cannot be separated cleanly because it is a tangle of prices, benefits, and measures that do not line up.

The largest piece, the essay argues, is automation hitting routine work first. It cites Daron Acemoglu and Pascual Restrepo’s estimate that between 50 and 70 percent of the change in America’s wage structure over the four decades to 2016 came from “relative wage declines of worker groups specialized in routine tasks in industries experiencing rapid automation.” The essay’s interpretation is that machines learned to do what ordinary workers did before they learned to do what the best-paid did.

The second missing piece is the six hours that did reach paychecks but as goods rather than time. Ordinary pay rose by a fifth, which could have been six hours off the week at the same standard of living. Instead hours went up. The essay argues this happened because the wage is also the ticket: if selling hours is the only way to eat, a person will sell them for whatever they fetch, and the wage records how badly the person needed money rather than what the hour was worth to them.

“An auction in which one side can’t walk away tells you very little about what the thing was worth.”

The essay says this produces two failures. Some work gets done that is not worth doing: a job that produces a little more than the wage it pays and less than the hour is worth to the person doing it covers its costs, but consumes something worth more than it makes, with the loss falling on the worker. And the economy has no way of saying “enough” in the market for work, because for the seller a price that falls far enough means not eating.

The ownership split: two-thirds of household purchases pay for work, one-third pays owners

The essay traces a dollar spent in a shop through America’s national accounts. For 2023, of what is paid to Americans for making the things households buy, 66 percent is pay for work and 34 percent goes to owners. A third of the economy already pays people for owning things, with no work asked of them.

The distribution is not evenly shared. The poorest tenth of American households lives mostly on government benefits. From the second tenth to the ninth, about four-fifths of what a household earns in the market is pay for work. In the richest tenth, pay and ownership are about half and half, and that tenth receives two-thirds of all income from ownership. The essay also states that 68 cents of every tax dollar in America is raised from pay for work, so a benefit is mostly a claim on somebody else’s wage.

The model: at a hundredfold improvement, work is paid 17 percent of income

The essay builds a small economic model fitted to the United States. Businesses give each task to whoever does it more cheaply, a person or a machine. People are far better than machines at some tasks and only a little better at others. Machines must be built out of other machines, some human work, and some land. Whoever pays for a machine wants a 10 percent annual return, and land is finite. Fitted to the United States today, the model pays 60 percent of income for work, 31 percent to machine owners, and 9 percent to land owners.

The experiment lets machines get steadily better at everything until they are a hundred times better relative to humans, and lets them learn to build one another. For a long time little happens: when machines are ten times better, work is still paid 52 percent of all income because people move to what they are best at. At about 35 times, machines that could replace even the hardest hour of human work take less than an hour of human work to build, and from there no wage can stay above the replacement cost. By the hundredfold point work is paid 17 percent of all income; at ten thousand times it is paid less than 1 percent.

The essay says nobody in the model gets poorer in terms of “stuff.” At the hundredfold point total income is 32 times what it was, and an hour of work buys 14 times as many goods as today. But an hour stops buying land: an hour’s pay rents about a twelfth of the land it does now, and the income of machine and land owners is 66 times what it was.

The distributional result is stark. If ownership stayed as it is today and work were paid 17 percent of income, the richest tenth of American households would receive 62 percent of all market income, up from 46 percent. The poorer half would go from 14.5 percent to 10 percent. The essay notes that such an economy would not seize up for want of customers: the owners are the customers, and the economy would turn to face them.

The author flags three cautions. The dial measures technology, not time: a hundredfold improvement takes 117 years at 4 percent annual improvement, the post–Second World War pace for machinery, and 25 years at 20 percent, and nobody knows the pace. If people have abilities no machine can match at any price, the same dial stops with work at 48 percent. And the model has one kind of worker, no risk, and no politics.

Taxing machines is not the fix; owning them is

The essay argues that a second ticket cannot be made of wages. A benefit paid out of today’s taxes is, for 68 cents in the dollar, wage earners paying each other; it moves the first ticket around. It also cannot be made by taxing what machines earn, which the author says was his own first idea. In the model, a 20 percent tax on what machines earn their owners raises 6.3 percent of national income, and wages fall by 6.1 percent of national income. Investors build machines until the last one just pays the return they want; tax the return and they build fewer, until the remaining machines earn the old return after tax. The owners keep their rate of return on a smaller pile, and workers have fewer machines to work with.

What works, the essay argues, is owning the machines. Let a company deduct the full cost of every investment from taxable profit on the day it makes it. Then on every new machine the government is effectively putting up a fifth of the cost and collecting a fifth of what the machine earns. The investor’s return on their own four-fifths is untouched, so just as many machines get built. The public gets its fifth, which is 6.2 percent of national income in the model, and wages do not change. The catch is that owners of existing machines lose about a fifth, once.

Land is different because nobody makes it. Tax machines and there will be fewer machines; tax land and there is as much land as before. The essay cites Thomas Paine’s 1797 argument in Agrarian Justice that “Man did not make the earth,” and that every landowner therefore owed the public a ground-rent. The author says he reached the idea independently before hearing of Paine, and for a while thought it was most of the answer. The numbers talked him out of it: the rent of the ground under America’s homes came to between 4.3 and 7.1 percent of households’ disposable income in 2024, or between $2,776 and $4,627 a year for every person in the country. That is real money, but not what a country can live on. It grows, though: in the model, land’s share of income more than triples.

The policy package: a common fifth, personal ownership, and a tax shift

The essay’s second ticket is ownership. For the United States it comes to three things.

First, hold a share in common and pay everyone its income. A fifth of all machines and land in the country would have paid $4,493 a person in 2023, or $374 a month, for every man, woman, and child. Most of that fifth is within reach without buying anything: the tax America already levies on company profits, turned into the partnership described above, and the growth in land rent together come to about a seventh of all ownership income, and a public fund could buy the rest. The essay is explicit that this is not free money: the profits tax is already being spent, so paying it out means finding $608 billion a year elsewhere, and whatever a fund buys, somebody pays for.

Second, let people buy the rest out of wages. A common share of about a seventh would be enough to keep the poorer half of households whole even if wages went to nothing. It cannot do that for the comfortable middle, who earn nearly half of all pay in the country. With the common fifth in place, saving 5 percent of pay and reinvesting what it earns at 5 percent a year would do it in 34 years, which is a working life.

Third, move taxes off the wage and toward spending, so that the safety net stops depending on the same thing as the incomes it is there to protect.

The essay points to existing examples. Since 1982 Alaska has paid every resident a yearly dividend from a fund built from oil revenues: $1,312 a person in 2023 and $1,702 in 2024, to more than 600,000 people each year, with a fund worth $88 billion. Norway’s national fund owns, in its own words, “almost 1.5 percent of all shares in the world’s listed companies.” Sweden puts 2.5 of the 18.5 percent of pay that goes toward the public pension into invested accounts that belong to the worker.

The essay says the dividend starts small: $374 a month is 29 percent of the poverty line for a person living alone. But it is made of the thing that is growing. In the model, the income of a common fifth is 13 percent of all wages today and 96 percent of all wages at the hundredfold point. A floor paid out of something everyone owns is mostly people paying themselves; as it grows, refusing becomes possible, and the wage goes back to being a price that means something.

There is a deadline. Today, everything that pays ownership income in the model economy, every machine and every acre, is worth about 6.7 years of the whole country’s wages. That is the price of a second ticket in the only currency most households have. At the hundredfold point it is 48 years. The saving that takes a comfortable-middle household 34 years today takes 76 years there. The households that most need a second ticket are the ones who would have to buy it out of wages, and wages are the thing losing value.

The fairness argument: tax what is most clearly earned, least

The essay engages the standard that people are entitled to what they contribute and that pay measures what you add. It calls that a reasonable standard, but says it is reasonable only if it is wholly true. A wage is paid for work. The return on a machine is paid to someone who went without something in order to build it. The rent of land is paid for something nobody made.

America, the essay says, raises 68 percent of its revenue from pay for work, the most plainly earned income, while its tax on what machines earn lands largely on wages again in the model. By its own favorite standard, the country taxes most heavily what is most clearly earned. The proposal is that standard applied all the way through.

Push the standard further. Suppose machines can do nine-tenths of all work. The standard does not break: those who work should keep their pay, and those who saved should keep their return. But it has less and less to say because less and less income is anyone’s contribution. In the model with the program in place, pay for work goes from 60 percent of income today to 17 percent at the hundredfold point and less than 1 percent in the end. What is held in common, the public’s fifth of machines plus growth in land rent, goes from 6 percent today to 40 percent at the hundredfold point and 48 percent in the end. That is about half, and nobody chose the number; it is where the contribution standard arrives when work becomes scarce.

The essay’s case for hope

The essay offers four reasons for hope. The hard problem of making enough is nearly solved: the United States makes in sixteen hours what took forty within living memory. In 1955, paying every American a poverty-line income would have cost between 27 and 52 percent of everything the country produced; in 2024 it would cost between 9 and 18 percent. What remains is a problem of arrangement, and arrangements are made by people.

Most of the change can still be done as a trade in which nobody is made worse off. Today a household gets a second ticket by putting aside pay and buying ownership from someone willing to sell it, and each side is paid in full. After a large shift, the same move becomes a transfer from those who gained to those who lost, which is much harder to ask. The essay argues we are still in the before, and that no one knows which side of the line their own work will fall on.

Every piece of the proposal already exists somewhere and works. And the machines themselves offer two different worlds from the same hundredfold improvement; the only difference is who holds the tickets. In one world, a tenth of households receive three-fifths of income and everyone else lives on a wage that has stopped meaning much. In the other, people own a share of what makes everything, the wage is one of two incomes, and work is something you choose. That world is 32 times richer, and your time in it is your own.

What the Hacker News discussion added

The Hacker News thread did not dispute the productivity–pay gap so much as contest its diagnosis, its novelty, and its remedy. Commenters raised competition, measurement, labor supply, and collective action as forces the essay underweights.

The AI disclosure drew criticism

The essay discloses that it was written with Claude Fable 5.2, Anthropic’s AI model, and that the author did not try to force the model to agree with him. One commenter objected to the framing:

“Condescending attitude aside, if I wanted to hear what Fable says about this, I'd ask Fable myself. I want to hear what YOU have to say about this.”

That criticism is about authorship and trust, not the arithmetic. It matters for how the essay’s claims are read: the author presents the work as his own argument, with AI used for measurement, figures, and drafting, but the disclosure’s tone became part of the discussion.

Some commenters said the argument is not new

One commenter argued that the basic observation, that productivity rises while the rich get richer, is not novel, and that the explanations rehash 250-year-old classical economics. The same commenter said the apparently novel part, a macro wage model described in “Pinning the Wage to Scarcity and Technology,” is not discussed in enough detail to judge its merits or historical accuracy. That is a fair boundary on the essay’s contribution: the policy proposal and the model are the new material, and the model’s assumptions carry much of the weight.

Measurement is harder than the headline numbers suggest

A commenter focused on productivity and inflation warned that comparing output over time is complicated by changing product mixes. If you look only at goods that are nearly identical across periods, intuitions hold: food once took 10 percent of hours and now takes 0.6 percent. But statistical agencies adjust for “features,” and a car or television today is treated as “more stuff.” The commenter’s point is that the old car might be much cheaper to make now with modern inputs, so the productivity story depends on how quality change is handled.

Another commenter raised the Baumol effect: people have become much more productive at some things and not at all more productive at others, and the less productive things become a larger fraction of what people want. From roughly 1980 to 2020, prices stayed stable for most things except housing, health care, education, and child care. That is a counterpoint to a single aggregate productivity number: the composition of spending matters.

Competition and “Red Queen” dynamics explain persistence

Several commenters argued that even if the essay’s accounting is right, competition explains why hours do not fall. One wrote that a company that slows to 30 hours loses to competitors and loses its livelihood, so it keeps 40 hours or disappears. Another described capitalism as a competitive environment in which efficiency gains raise the opportunity cost of not working harder: “We aren't competing with workers from 1973, we're competing with other workers leveraging the same accelerated efficiency.”

A related comment called tech and business a “Red Queen’s Race”: if everyone else runs faster, you have to run faster just to stay in place. Software’s goal is to make things faster, and then the edge is eaten by making them faster for everyone. Another commenter framed the same dynamic as a tragedy of the commons: if everyone worked 20-hour weeks, they would too, but others hustle to get rich, so they adapt to that culture.

One commenter tied the persistence of long hours to labor-market competition among the hardest-working participants: the hardest workers set the tone for competition for high-paying jobs, and a typical tech job requires far more than 40 hours. That is a supply-side explanation the essay’s model, with one kind of worker and no politics, does not capture.

Housing, rank, and relative status

One commenter argued that competition for housing drives prices up as wages increase, and that happiness correlates more with economic rank in a society than with absolute wealth, citing Boyce, Brown, and Moore (2010), “Money and Happiness: Rank of Income, Not Income, Affects Life Satisfaction.” If relative rank matters, then more output does not translate into more leisure unless the distribution changes; people may work to keep or improve their position.

Labor supply and immigration

One commenter noted that comparisons between the 1970s and today often omit the large influx of women and immigrant labor into the workforce. The commenter did not take a position on who should work, but argued that going from women being about 30 percent of the workforce in 1970 to more than 50 percent in 2026, alongside expanded immigrant labor, must affect workers’ leverage. That is a demographic counterweight to a purely technological story.

The investor class and the demand for savings

One commenter put the distributional point bluntly: productivity gains have mostly gone to the investor class, who are “not only not working but making more for not working than they ever have.” Another commenter emphasized that money comes from other people, who do not give it away unless you satisfy them, and that people who want work from you will take all you can perform up to the limit of labor laws.

A different commenter focused on collective action and savings demand: motivation to work less than 40 hours is low when everyone around you works 40 hours, and the United States has policies that increase the demand for savings. That connects to the essay’s point that a second ticket must be bought, but it also suggests that preferences and institutions, not only technology, set hours.

Some commenters proposed structural priorities

One long comment argued that the problem is insufficient collective drive to provide for everyone before the wealthy are allowed to extract. If two people need something to survive, the group needs to produce two of it; if one demands a thousand, the group needs to produce 1,001 to satisfy demand. The commenter’s proposal is to prioritize food, water, shelter, and medical care before satisfying the desires of the rich, so that rich people can still buy private jets but must pay more to convince people to give up leisure. The commenter called the political focus on “jobs jobs jobs” instead of needs and rest “the greatest tragedy of modern capitalism.”

Another commenter argued for shifting taxes off individual income toward consumption and wealth taxes, with companies paying wealth and turnover taxes appropriately scaled. The claim is that this could stimulate diversity of small and medium companies, reduce monoculture markets, and stop incentivizing the established over the emerging. That is a tax-structure counterpoint to the essay’s ownership-focused proposal.

AI may already be quietly reducing hours

One commenter observed that nobody wants to admit they are having a 15-hour work week with AI, but office attendance suggests many knowledge workers are starting to reduce hours, especially remotely. Some fill the time with more work, but many are not increasing work volume; they are quietly reducing labor. That is anecdotal, but it is a live hypothesis the essay’s model does not measure.

The “productivity reset” metaphor

One commenter summarized the dynamic with a metaphor: there is a machine called Productivity with a knob and a button labeled “reset.” Whenever productivity increases, someone presses reset, the new level becomes the norm, and the extra time is not for rest but for finding ways to produce more. That captures the essay’s institutional point without requiring the model: gains are absorbed by changed expectations.

The bottom line

The essay’s strongest factual claim is that the United States has already freed the hours: a 40-hour week of 1973 output takes about 16 hours today, while hours worked per person are higher. Its strongest analytical claim is that the wage cannot simultaneously be a neutral price for time and the only ticket most people have to survival. Its policy claim is that the second ticket must be ownership, built through a common share, personal saving, and a tax shift, and that the price of buying that ticket rises as machines improve.

The Hacker News discussion adds three durable qualifications. Competition can force long hours even when productivity rises, because firms and workers are running against each other rather than against the past. Measurement of productivity and inflation is contested when product quality and composition change. And the essay’s model is deliberately simple, with one kind of worker and no politics, so its predictions are conditional on assumptions that commenters did not all accept. The debate is less about whether the productivity–pay gap exists than about which mechanism deserves the most weight, and whether ownership policy can be adopted before the price of a second ticket becomes prohibitive.

Sources