What this covers

Daron Acemoglu and Russ Roberts examine whether technological progress automatically distributes its gains across society. Acemoglu argues that productivity gains do not inherently raise broad living standards. Instead, whether innovation translates into higher wages depends on three specific conditions: genuine competition in product markets, absence of coercive power in labor relations, and technology that creates new tasks rather than purely automating existing work. Roberts presses back with cases where market forces—price drops, alternative employment, labor mobility—have historically absorbed displacement. The conversation moves between economic theory, historical episodes, and contemporary concerns about automation and market concentration.

Acemoglu draws on concrete history: medieval windmills that increased productivity tenfold but enriched only lords who monopolized them; the cotton gin's role in spreading slavery deeper into the American South; and the first ninety years of British industrialization, during which children worked eighteen-hour days in mines and life expectancy fell below thirty despite industrial progress. He contends that frictional labor market conditions mean workers and employers bargain over quasi-rent—surplus value neither party can capture alone—and an employer with overwhelming power can implement mediocre automation purely to strip rents away, even when it damages overall efficiency. He also implicates macroeconomic forces: tax codes that favor capital over labor, and societal "Vision as Driver of Technology" toward machine intelligence, pushing innovation down excessive automation paths. Roberts counters that labor mobility and broad price decreases protect workers more than the bargaining-power account acknowledges, and that competition eventually erodes monopoly rents. The debate circles through antitrust, trade unions, digital platforms, and whether artificial intelligence will concentrate power or diffuse it.

Sharpest takeaway

Acemoglu argues that innovation does not automatically produce shared prosperity; whether technology raises wages broadly depends on competition, worker bargaining power, and the balance between automation and the creation of new labor tasks, all of which require deliberate institutional shepherding rather than faith in market automaticity.

  • Historical episodes (medieval windmills, the cotton gin, the early Industrial Revolution) show productivity gains that did not reach workers due to monopoly, coercion, or pure automation.
  • Prosperity from technology requires three preconditions: competition, non-coercive labor markets, and technology that raises labor's marginal productivity (new tasks) rather than only automating.
  • Macroeconomic forces (tax bias toward capital, rent-shifting automation, and societal 'vision') can push technology toward excessive automation that does not benefit workers.

The argument · threads6 threads · 30 claims
0.80

Labor markets are imperfectly competitive; quasi-rents from search frictions create bargaining problems where organized worker power is needed to check employer dominance.

5 pointscentrality 5/5
  • Rather than treating labor markets as fully competitive, one should model them as frictional: search frictions mean every employment relationship contains a quasi-rent (because both parties are locked in and lack perfectly indifferent outside options in the short run), and how this quasi-rent is divided is a bargaining problem; if one side has all the bargaining power, the division will be unfair and not even necessarily efficient.

    “if you have a frictional labor market where there are, for instance, search frictions--which mean that every employment relation has a quasi-rent in it.”

  • Large, well-organized companies facing completely disorganized individual workers create an imbalance in splitting employment quasi-rents: an individual confronting a bureaucratic organization can become powerless through divide-and-rule, which is why—per Galbraith's countervailing powers—worker coordination is needed to balance large corporations.

    “if you have large companies that are very well organized and completely disorganized workers, it's going to be in the splitting of this quasi-rent that there's going to be an imbalance.”

  • Worker voice was decisive: trade union activity and democratic movements were harshly suppressed in early industrial Britain (the Chartists collected 3 million signatures for minimal political reform and were ignored, their leaders jailed), and only after the Master and Servant Acts were addressed and trade unions legalized did factory conditions improve substantially.

    “Chartists collected 3 million signatures... Three million signatures for asking for minimal changes in the political system so that people could vote and could participate in the political process.”

  • The free snacks, coffee, and entertainment that companies like Google and Facebook lavished on employees are evidence of quasi-rent splitting, not competitive labor markets, because in a truly competitive market a firm would only pay a worker's marginal product; throwing money at workers signals that valuable, specifically-skilled employees generate quasi-rents the firm fears losing.

    “In a competitive labor market, I would never throw money at you like that because I'm just paying for your marginal product anyway. So, the fact that I am, like, throwing money left, right, and center is already sort of a sign that there are some quasi-rents being split there.”

  • While long-run effects of trade unions are very difficult to know, in the short to medium run there are both theoretical reasons and empirical evidence that effective unions can have significant effects without creating major inefficiency, particularly in the modern factory system by leveling the playing field; Scandinavian data show unions acting in concord with employers compressing the wage structure with massive measurable effects.

    “if you look at Scandinavian countries, there you are actually seeing how unions, when they are acting in a way that's in concord with employers... also compressing the wage structure, they are having massive effects that you can see in the data.”

0.75

Technology's direction is chosen, not inevitable; profit-driven innovation may bias toward automation over complementary tasks, and concentrated firms distort this balance further.

4 pointscentrality 5/5
  • New tasks are not exogenous but endogenously generated by profit-motivated entrepreneurs and innovators within directed technological change models; however, there is no necessity that the efficient balance between new tasks and automation will be created, nor that it will occur at the same rate regardless of market structure—so a tech sector monopolized by a few firms may distort the balance toward automation.

    “there is no necessity that the efficient balance of new tasks and automation will be created. Nor is that always going to be taking place at the same rate irrespective of the market structure.”

  • Government should not slow down automation—rapid automation is desirable—but should use programs and competitions (as DOD, NIH, and NSF have historically done) to encourage human-complementary uses of AI and digital technology so that worker productivity remains central, without sitting in the driver's seat of innovation.

    “Definitely not automation taxes, definitely not slow down automation. We do not want to slow down automation. We want automation to be rapid, but at the same time find other things that we can do for workers”

  • In an imperfect labor market with quasi-rents, mediocre automation technology can have a rent-shifting role: an employer who was sharing rents with employees (creating welfare valued by a social planner) may be tempted to adopt so-so automation that strips those rents away, making private and social incentives to automate diverge.

    “now you give me an automation technology. That's so-so... but it strips those rents away. I might be tempted to use it and the social planner doesn't want me to.”

  • A society-wide 'vision'—such as an obsession with implementing machine intelligence and getting rid of humans—can, even if temporary, carry the economy down an excessive automation path; vision is therefore a central macroeconomic force shaping how technology develops.

    “The whole industry--the whole society--may get obsessed with a vision... that says, 'Oh yes, we have to implement machine intelligence, artificial intelligence. Let's get rid of humans.' And, that can carry the economy for quite a while down an excessive automation path.”

0.66

History shows technological progress does not automatically raise worker wages unless accompanied by new task creation and institutional conditions that prevent rent capture.

5 pointscentrality 4/5
  • Medieval windmills increased productivity more than tenfold in some tasks, but because lords and abbots monopolized them, forcibly suppressed competition, and coerced people into using their mills, few if any benefits spread beyond a small narrow elite.

    “Windmills, for example, increased productivity more than 10-folds in some tasks, but because they were monopolized by lords and abbots who actually forcefully did not even allow any competition and coerced people to use their mills, few, if any of the benefits spread beyond that small, narrow elite.”

  • The cotton gin transformed the United States from an agricultural backwater into the largest cotton exporter and revolutionized the South's economy, but enslaved black workers fared much worse because they were moved into harsher Deep South plantations as cotton cultivation spread.

    “the workers--black, enslaved people--actually fared much worse under the cotton gin because they were moved into much, much harsher plantations in the Deep South, as cotton plantations spread.”

  • During roughly the first 90 years of the Industrial Revolution (about 1750 to 1840), working people did not much benefit despite industrial progress; evidence includes an 1834 Parliamentary Report of children as young as five or six working 18 hours in deep coal mines, life expectancy at birth falling below 30 in British cities, rampant infectious disease, no sewage systems, and significantly expanded working hours.

    “In 1840 [1834--Econlib Ed.], there was a Parliamentary Report about the condition of England that reported children as young as five or six working 18 hours in deep coal mines and pushing carts with their heads.”

  • Roberts' account that automation cuts prices and reallocates displaced workers is only half the story; if automation were all that happened, labor's share of income would decline steadily, but for most of the 20th century it did not, because scientific knowledge is simultaneously applied to create many new tasks—including for unskilled labor—and it is these new tasks that reinstate employment, raise wages, and keep labor's share constant.

    “if that's all that's going on--automation and all that's going on--labor's share would decline steadily both at the industry level and at the economy level. But, that's not what we see in the data for much of the 20th century.”

  • Because skilled labor was in very short supply in the United States, American innovators' creative energy was devoted to raising the productivity of unskilled labor and integrating it centrally into production; this drove the interchangeable-parts 'American system of manufacturing,' making unskilled labor's productivity—and hence wages—high, a phenomenon first hypothesized by economic historian H.J. Habakkuk.

    “Because, skilled labor was in very short supply in the United States, the whole creative energy of American innovators and entrepreneurs was devoted to increase the productivity of unskilled labor.”

0.44

Digital platforms' advertising business model creates harmful externalities and chokes alternative models, warranting tax intervention to enable ad-free competition.

6 pointscentrality 2/5
  • The digital ad business model misuses online technology by creating an ecosystem of outrage, emotional engagement, and extreme reactions with negative externalities especially on teenagers, and it chokes out alternative business models; a digital ad tax could open space for ad-free models (like Wikipedia and Netflix, which lack misinformation, unlike Facebook, Twitter, and Reddit, whose business model produces it).

    “Wikipedia doesn't have misinformation. Netflix doesn't have misinformation, but Facebook, Twitter, Reddit do. Why is that? It's because of their business model.”

  • Contrary to the view that ChatGPT makes Google vulnerable, current foundation models like GPT-4 are so resource-intensive and developing in such a way that they may actually strengthen the hold of Google and Microsoft, since OpenAI is now effectively part of Microsoft (which is embedding GPT-4 across its products) and Google is advanced in foundation models too.

    “the foundation models which--like ChatGPT-4--are so resource-intensive, and they're going in a particular way that they might actually strengthen the hold of Google and Microsoft.”

  • Thanks to technological breakthroughs starting around the mid-18th century, applying scientific knowledge to industrial processes, we enjoy much better health, comfort, and prosperity, making us huge beneficiaries of technological progress.

    “today we are extremely fortunate, from a historical sweep, because of the technological breakthroughs that started sometime around the middle of the 18th century”

  • Immigrants came to the United States because living conditions and wages were genuinely higher there, while workers flocked to British cities like London mainly because of push factors: rising agricultural productivity reducing labor needs and the Enclosure Movement, which cut off traditional means of survival (collecting wood, hunting) in a harsh and rapid way.

    “the reason why they went to London is mostly push factors. So, labor was pushed out of the countryside in the United Kingdom for two reasons.”

  • While humanity has hugely benefited from technological progress since the mid-18th century, there is nothing automatic about technology bringing widespread prosperity; believing it lifts all boats automatically causes us to abrogate the responsibility to shepherd it through institutional adjustments so society benefits broadly.

    “there is nothing automatic about technological progress bringing widespread prosperity”

  • We should return to and rethink antitrust because we have never seen private organizations as large as Google, Microsoft, Facebook, and Amazon, which hold enormous social, political, and economic power; whether this is justified by economies of scale or should be targeted like Standard Oil is an open and important question.

    “We have never seen as large organizations in the world--private organizations--as Google, Microsoft, Facebook, Amazon. That means they have enormous amount of social, political, as well as economic power.”

0.43

Managerial ideology prioritizing shareholder value over worker welfare has reduced wage growth and labor's income share independent of technological change.

3 pointscentrality 3/5
  • The Friedman Doctrine that the social responsibility of business is profits motivated a movement among managers to stop being generous with employees, and research shows CEOs with MBAs—trained that shareholder value is key—are keen on not giving workers raises compared to comparable companies, producing a negative effect on wage growth and labor share.

    “I have some work, for example, with Alex He and Daniel le Maire showing how CEOs with MBAs... are very keen on not giving raises to their workers, whereas comparable companies do so. So, they have a negative effect on wage growth and labor share.”

  • For technology to translate productivity into broad prosperity, three critical ingredients must hold: competition (which is not automatic), absence of coercion in labor markets (when employers have too much power, higher marginal productivity need not raise wages), and technology that raises labor's marginal productivity rather than pure automation (which raises capital's productivity but may break the chain from higher output to higher wages).

    “One is: competition... Second: coercion in labor markets... And, Third: automation. That, by itself automation may increase capital's productivity--may increase profits--but it may not increase the marginal productivity of labor”

  • US tax policy biases firms toward automation: the marginal tax rate on labor is about 25% while on equipment capital it is about 5%, so profit-maximizing companies are encouraged to use and innovate on capital rather than hire and innovate on labor; removing these distortions is one corrective.

    “the marginal tax rate on labor is about 25%, on capital is about 5%--equipment capital. So, we're creating a tax system... that subsidizes people to do more automation and less of hiring workers.”

0.38

Economic theory's reliance on the Cobb-Douglas function obscures how automation actually works and prevents recognition that productivity gains need not translate into wages.

2 pointscentrality 3/5
  • Economists' view of technology is colored by the convenient Cobb-Douglas production function, which makes marginal product and average product proportional, so any increase in average productivity automatically raises labor's marginal productivity by the same amount and translates into higher wages; the real world is not like that, especially with many automation technologies.

    “the Cobb-Douglas production function is special... what it implies is that marginal product and average product are proportion.”

  • The 'factory of the future will have two employees, a man and a dog'—the dog there to keep the man from touching the equipment and the man to feed the dog—illustrates how, with extensive automation, the remaining workers contribute little to average productivity, so the competitive process will not pay them high wages; this is the essence of automation.

    “the factory of the future will have two employees, a man and a dog. The dog is there to make sure that the man doesn't touch the equipment and the man is there to feed the dog.”