3 ms·

If we adopt a 4 day / 36 hour workweek as normal, we can expect substantial social benefits: - Reduced commuting, etc. - Improved health / quality of life. -

by mlyle 3y ago

If we adopt a 4 day / 36 hour workweek as normal, we can expect substantial social benefits: - Reduced commuting, etc. - Improved health / quality of life. - Reduced supply of labor (due to overtime kicking in) should push up equilibrium price. > It feels strange to be offered the chance to work less than half as much as another human and receive substantially greater rewards for my efforts This is the nature of any job where there's substantial leverage effects: if your work can make a hundred people fractionally more productive, you're going to reap outsized rewards. LLMs are just going to amplify this, by devaluing a lot of work but allowing other people to be fantastically more productive thanks to their leverage.

ttoinou 3y ago
Lets make it 1 day workweeks then. The benefits will be even greater
- mlyle 3y ago
  Alternatively, let's go back to sweatshop, all-day-long backbreaking labor. (It turns out this style of argument is not very useful). We have a high enough standard of living to shoulder the downsides of 10% fewer working hours (which should lower productivity by less than 10%; perhaps much less) in exchange for the upsides. Automation poses substantial risk of high unemployment; shortening the work week a little is a good hedge (and offers other benefits).
  - gottorf 3y ago
    No need to make any policy changes; existing policy appears enough to make people work less. The labor force participation rate has been steadily decreasing from the 1990-2000 peak[0]. As long as the ones who are working 10% fewer hours are also taking home 10% less, I don't see the problem with it. But the current system of the welfare state that depends on redistributive taxation has a strong incentive against that change; if everyone working works 10% less, that's 10% less available to be transferred to those who aren't. [0]: https://fred.stlouisfed.org/series/CIVPART https://fred.stlouisfed.org/series/CIVPART
    - mlyle 3y ago
      > No need to make any policy changes; existing policy appears enough to make people work less I interpret this in the opposite way you do. Labor force participation includes effects where people think it's not economically worth their time to work. Reducing supply of labor could be expected to increase labor force participation. Not to mention this version of the statistic is a completely worthless one: it's not corrected for demographic (aging) effects. > As long as the ones who are working 10% fewer hours are also taking home 10% less This is unlikely: when you reduce the supply of something, the price tends to go up. > if everyone working works 10% less, that's 10% less available The marginal benefit of the factor (e.g. per-hour productivity) tends to go up, too. (Declining marginal factor product, blah blah blah). Indeed, this is why the price of the factor increases.
      - gottorf 3y ago
        > effects where people think it's not economically worth their time to work This is only possible with generous social benefits given to those who aren't working. In the top 8 most generous states, transfer payments given to those not working exceed the median income[0]. > it's not corrected for demographic (aging) effects You're right. The prime working-age group (25-54) reversed the same downward trend after the 90-00 peak sometime in 2015, and recovered relatively quickly post-Covid[1]. But the picture isn't as rosy for young (20-24) entrants to the employment marketplace[2], which is worrisome. > The marginal benefit of the factor (e.g. per-hour productivity) tends to go up, too I understand, but aggregate production will go down, would it not? [0]: https://www.cato.org/sites/cato.org/files/pubs/pdf/the_work_versus_welfare_trade-off_2013_wp.pdf https://www.cato.org/sites/cato.org/files/pubs/pdf/the_work_... [1]: https://fred.stlouisfed.org/series/LNS11300060 https://fred.stlouisfed.org/series/LNS11300060 [2]: https://fred.stlouisfed.org/series/LNS11300036 https://fred.stlouisfed.org/series/LNS11300036
        mlyle 3y ago
        > This is only possible with generous social benefits given to those who aren't working. No; for example, as discussed, retirees are in the set. If supply of labor falls, a greater proportion of people able to be retired will work due to economic incentives. > I understand, but aggregate production will go down, would it not? Yes, but it contradicts your "10% less". Fighting the 10% less is: - For each worker, their least productive hours will be cut rather than their most productive - For some of hours lost, additional workers will be employed in higher value jobs. - Some types of output will not be affected at all; others it will be worth paying overtime. - For some of hours lost, automation will be employed. - Lowering commute use of roads and transport will increase productivity. All in all, I'd guess lowering the work week by 10%, and the nominal days-per-week to 4, might lower aggregate output by a couple percent. I would also expect wages to increase, arresting trends where wages have stagnated as productivity has increased.

by mlyle 3y ago

ttoinou 3y ago

Lets make it 1 day workweeks then. The benefits will be even greater

mlyle 3y ago
Alternatively, let's go back to sweatshop, all-day-long backbreaking labor. (It turns out this style of argument is not very useful). We have a high enough standard of living to shoulder the downsides of 10% fewer working hours (which should lower productivity by less than 10%; perhaps much less) in exchange for the upsides. Automation poses substantial risk of high unemployment; shortening the work week a little is a good hedge (and offers other benefits).
- gottorf 3y ago
  No need to make any policy changes; existing policy appears enough to make people work less. The labor force participation rate has been steadily decreasing from the 1990-2000 peak[0]. As long as the ones who are working 10% fewer hours are also taking home 10% less, I don't see the problem with it. But the current system of the welfare state that depends on redistributive taxation has a strong incentive against that change; if everyone working works 10% less, that's 10% less available to be transferred to those who aren't. [0]: https://fred.stlouisfed.org/series/CIVPART https://fred.stlouisfed.org/series/CIVPART
  - mlyle 3y ago
    > No need to make any policy changes; existing policy appears enough to make people work less I interpret this in the opposite way you do. Labor force participation includes effects where people think it's not economically worth their time to work. Reducing supply of labor could be expected to increase labor force participation. Not to mention this version of the statistic is a completely worthless one: it's not corrected for demographic (aging) effects. > As long as the ones who are working 10% fewer hours are also taking home 10% less This is unlikely: when you reduce the supply of something, the price tends to go up. > if everyone working works 10% less, that's 10% less available The marginal benefit of the factor (e.g. per-hour productivity) tends to go up, too. (Declining marginal factor product, blah blah blah). Indeed, this is why the price of the factor increases.
    - gottorf 3y ago
      > effects where people think it's not economically worth their time to work This is only possible with generous social benefits given to those who aren't working. In the top 8 most generous states, transfer payments given to those not working exceed the median income[0]. > it's not corrected for demographic (aging) effects You're right. The prime working-age group (25-54) reversed the same downward trend after the 90-00 peak sometime in 2015, and recovered relatively quickly post-Covid[1]. But the picture isn't as rosy for young (20-24) entrants to the employment marketplace[2], which is worrisome. > The marginal benefit of the factor (e.g. per-hour productivity) tends to go up, too I understand, but aggregate production will go down, would it not? [0]: https://www.cato.org/sites/cato.org/files/pubs/pdf/the_work_versus_welfare_trade-off_2013_wp.pdf https://www.cato.org/sites/cato.org/files/pubs/pdf/the_work_... [1]: https://fred.stlouisfed.org/series/LNS11300060 https://fred.stlouisfed.org/series/LNS11300060 [2]: https://fred.stlouisfed.org/series/LNS11300036 https://fred.stlouisfed.org/series/LNS11300036
      - mlyle 3y ago
        > This is only possible with generous social benefits given to those who aren't working. No; for example, as discussed, retirees are in the set. If supply of labor falls, a greater proportion of people able to be retired will work due to economic incentives. > I understand, but aggregate production will go down, would it not? Yes, but it contradicts your "10% less". Fighting the 10% less is: - For each worker, their least productive hours will be cut rather than their most productive - For some of hours lost, additional workers will be employed in higher value jobs. - Some types of output will not be affected at all; others it will be worth paying overtime. - For some of hours lost, automation will be employed. - Lowering commute use of roads and transport will increase productivity. All in all, I'd guess lowering the work week by 10%, and the nominal days-per-week to 4, might lower aggregate output by a couple percent. I would also expect wages to increase, arresting trends where wages have stagnated as productivity has increased.