Things that went whizzing by that I want to remember: First: Structural change is much more likely to be win-win and much more likely to take place rapidly when it takes place in a time of short supply as opposed to a time of short demand...
Forbes should add still another criterion: Berkeley educates far more students every year than Harvard does (a point made here on several occasions). Not only is Berkeley doing a better job on the percentages of lower income students, it's doing so with much larger nominal admissions.
Structural change at the level of labour allocation isn't going to work if you can't obtain and maintain the robots.
Present circumstances strongly implies that the only way to do that involves shortening the supply chains, which in turn means creating local communities of practice. Which would be tough enough without capital return optimization insisting that you don't do that, the return is higher if the supply chain is long and the communities of practice for any one thing few.
Headline on twitter that the chip shortage is costing the auto industry a couple hundred billion dollars. Aside from desperately not needing this auto industry, that's pretty obviously not enough disruption to cause any market-driven supply-chain shortening. If that's not, well, it's because the market can't. It takes policy and the exercise of the civil power.
The Evergrande thing is rather telling. The US dealt with its housing crisis by bailing out the financial firms and letting millions of people lose their houses. China seems to be dealing with it by letting the financial firms collapse and making sure that people get their new homes. I guess this shows the wretched immorality of Communism in placing people's need for a place to live about the financial sector's need to show rising profits. The horror, the horror.
Keep in mind that the people of China respond to perceived injustice with violent protest on scales that make it a goal of CPC policy to avoid such perceptions occuring.
The axiom of freedom does not survive examination; most people in the Anglosphere are heavily complicit in their subjugation, having accepted the mammonite axiom that their employer does them a kindness to employ them at all. (This is absolutely one of the slavery-derived things in the business practices.)
I've long been saying that the reason we don't see robots raising productivity is that labor is too cheap. There's too big a labor pool to make robots worthwhile. I figured that when the Baby Boomers started retiring we might see more productive robots, but COVID has accelerated this. (It has also accelerated Baby Boomer retirements. I'd love to see if there has been a boom in Social Security retirements.
I think your conjecture on labor vs robots needs to be adjusted for relative productivity. A robot doing a job 1000x faster than a human will likely replace humans, e.g. weaving machines. A robot doing a job that can only work at human speed, e.g. a carer, will not offer any productivity advantage and therefore cost is very important. Such robots will need to be sophisticated and have human-level intelligence. That is expensive. A self-driving car cannot drive faster than a human driver, so the cost must be commensurate with the saving of having no human driver. I suspect a careful analysis of jobs will show that most must operate at human speeds and cannot be automated to increase that speed by orders of magnitude. The tasks that can be automated to be much faster are the best targets for robots.
But here is the rub. Automation is usually done by designed machines, not humanoid robots replacing a human. Industrial robots have some features that mimic human capabilities, but not all. A robot that can replace a human by mimicking a human is more expensive (c.f. Baxter) but not more productive. AMZN may automate with specialty robots the moving of pallets of items around but it still uses humans to do the packing. A robot would not be able to do the job faster as items would break if the forces on them increase by accelerating them too quickly were applied.
I am also reminded in Clark's "A Farewell to Alms" that while weaving machines were used to replace people in England when those machines were exported to India they kept all their workers so that increased productivity per worker was not realized.
Forbes should add still another criterion: Berkeley educates far more students every year than Harvard does (a point made here on several occasions). Not only is Berkeley doing a better job on the percentages of lower income students, it's doing so with much larger nominal admissions.
Value added per gift (and state subsidy) dollar would be a very useful metric that I have occasionally wondered how I would calculate...
University WAR would be very useful. Pity about the measurement issues.
Structural change at the level of labour allocation isn't going to work if you can't obtain and maintain the robots.
Present circumstances strongly implies that the only way to do that involves shortening the supply chains, which in turn means creating local communities of practice. Which would be tough enough without capital return optimization insisting that you don't do that, the return is higher if the supply chain is long and the communities of practice for any one thing few.
Headline on twitter that the chip shortage is costing the auto industry a couple hundred billion dollars. Aside from desperately not needing this auto industry, that's pretty obviously not enough disruption to cause any market-driven supply-chain shortening. If that's not, well, it's because the market can't. It takes policy and the exercise of the civil power.
The Evergrande thing is rather telling. The US dealt with its housing crisis by bailing out the financial firms and letting millions of people lose their houses. China seems to be dealing with it by letting the financial firms collapse and making sure that people get their new homes. I guess this shows the wretched immorality of Communism in placing people's need for a place to live about the financial sector's need to show rising profits. The horror, the horror.
Keep in mind that the people of China respond to perceived injustice with violent protest on scales that make it a goal of CPC policy to avoid such perceptions occuring.
The axiom of freedom does not survive examination; most people in the Anglosphere are heavily complicit in their subjugation, having accepted the mammonite axiom that their employer does them a kindness to employ them at all. (This is absolutely one of the slavery-derived things in the business practices.)
How true.
I've long been saying that the reason we don't see robots raising productivity is that labor is too cheap. There's too big a labor pool to make robots worthwhile. I figured that when the Baby Boomers started retiring we might see more productive robots, but COVID has accelerated this. (It has also accelerated Baby Boomer retirements. I'd love to see if there has been a boom in Social Security retirements.
I think your conjecture on labor vs robots needs to be adjusted for relative productivity. A robot doing a job 1000x faster than a human will likely replace humans, e.g. weaving machines. A robot doing a job that can only work at human speed, e.g. a carer, will not offer any productivity advantage and therefore cost is very important. Such robots will need to be sophisticated and have human-level intelligence. That is expensive. A self-driving car cannot drive faster than a human driver, so the cost must be commensurate with the saving of having no human driver. I suspect a careful analysis of jobs will show that most must operate at human speeds and cannot be automated to increase that speed by orders of magnitude. The tasks that can be automated to be much faster are the best targets for robots.
But here is the rub. Automation is usually done by designed machines, not humanoid robots replacing a human. Industrial robots have some features that mimic human capabilities, but not all. A robot that can replace a human by mimicking a human is more expensive (c.f. Baxter) but not more productive. AMZN may automate with specialty robots the moving of pallets of items around but it still uses humans to do the packing. A robot would not be able to do the job faster as items would break if the forces on them increase by accelerating them too quickly were applied.
I am also reminded in Clark's "A Farewell to Alms" that while weaving machines were used to replace people in England when those machines were exported to India they kept all their workers so that increased productivity per worker was not realized.