3 ms·

> it seems logical to conclude that the sum total of available commercially viable jobs will decrease with time. It might seem logical, but it isn't true. The

by brc 14y ago

> it seems logical to conclude that the sum total of available commercially viable jobs will decrease with time. It might seem logical, but it isn't true. The Luddites were people who were frightened of the introduction of machinery, and used to go around attacking factories, thinking there would be no more jobs. In reality, whenever automation provides benefits through lower prices, what this does is reduce the amount of hours labor a person needs to spend in order to fulfill that need. If you have a cow, it will take a person a couple of hours to obtain a gallon of milk, and then you have to house, feed and look after the cow. Instead, with automated and factory-produced milk, the average person spends a fraction of that time - maybe 10 minutes - to earn the money to purchase that milk at a convenient location. And so it goes with everything - whenever a product is automated and the jobs (hours) required to produce that thing is reduced, then the cost of the product falls. The consumer then has a surplus of cash, which is really just a surplus of their time spent, to either keep for themselves, or spend on something else. As human needs and wants are essentially unlimited (most of us want to go to the moon if it was possible) then, for each good that falls in price, demand will shift to other goods. As automation lowers the cost of each additional good, a person is able to consume more of that good or more additional goods. Now, you might say that 'consumption' is bad - but consuming a good might also mean consuming a service, like a movie, a local band, or an art gallery, which is still consumption, but not in the 'using up resources' manner (although nearly every consumption save for walking to a tai-chi class on the local beach uses up some kind of resources) This curve of increasing standard of living is intrinsically linked to increases in productivity, which roughly means the amount of goods + services produced for the given input of working hours. Thus increasing amounts of automation will never, ever reduce the supply of jobs, because each automated thing that frees up human time for something else, in itself creates activity which generally means creating more jobs. You don't get an iPhone 5 to buy unless farmers can milk a herd of 150 cows singlehandedly. And so we'll never have tourist trips to the moon unless there is increasing automation amongst industries that are currently labour-intensive. This has been going on for a millenia, and there is no tipping-point or transition point, as it is a very smooth curve. Yes, it can be disruptive for small pockets of workers who become no longer required - there are no more pools of typists - but at the macro level the transition is relatively smooth.

doktrin 14y ago
>> As human needs and wants are essentially unlimited (most of us want to go to the moon if it was possible) then, for each good that falls in price, demand will shift to other goods. As automation lowers the cost of each additional good, a person is able to consume more of that good or more additional goods. I'm well aware of the Luddite Fallacy :) While it is true that the economy has thus far been able to re-absorb a workforce displaced by automation, we should not take as a given this will be the case forever. Certainly not based on the premise of "infinite human wants". Frankly, I am not convinced our desires are in fact infinite. Our demands do adjust with opportunity, but only to a point. For instance, many people are satisfied after achieving a certain relative level of security and comfort. However, more important than demand is how many workers are required to meet the supply. If innovations in automation continue at even a linear pace (and, frankly, we're probably talking quadratic), then we will eventually reach a point where staggeringly few actual workers are required for the production of a given good. On a slightly less speculative note, a documented phenomenon of workplace automation in developed nations in recent years is the increase in income & labor polarization (or, inequality) [1]. Essentially, while total employment is somewhat maintained, skill (and, to a large extent) income become polarized at the ends. From the MIT study [1] : In net, employment changes in the U.S. during this period were strongly U-shaped in skill level, with relative employment declines in the middle of the distribution and relative gains at the tails ... Technological progress in our model takes the form of an ongoing decline in the cost of computerizing routine tasks, which can be performed both by computer capital and low skill (‘non-college’) workers in the production of goods. The adoption of computers substitutes for low skill workers performing routine tasks–such as bookkeeping, clerical work, and repetitive production and monitoring activities–which are readily computerized because they follow precise, well-deﬁned procedures. Importantly, occupations intensive in these tasks are most commonplace in the middle of the occupational skill and wage distribution The same paper does mention that some low skilled workers are able to shift (and even see some relative wages rise) towards service economy jobs. However, that certainly doesn't change the macro picture. Lastly, in the near-mid term the greatest challenge is most likely not the absolute out-and-out displacement of work, but rather that displacement vastly outpaces the ability of the economy to re-generate employment. >> This has been going on for a millenia, and there is no tipping-point or transition point, as it is a very smooth curve. A very smooth J-curve perhaps. Implying that historical trends over the past millenia are a guaranteed predictor of future events is a highly dubious claim. We are living in a relatively unprecedented age of advancement and growth by just about every possible measure, and it's hardly a given that what has held true in the past (e.g. "Luddite's fallacy") will always hold true. [1] http://economics.mit.edu/files/1474 http://economics.mit.edu/files/1474 [2] http://www.economist.com/blogs/babbage/2011/11/artificial-intelligence http://www.economist.com/blogs/babbage/2011/11/artificial-in...

by brc 14y ago

doktrin 14y ago

>> As human needs and wants are essentially unlimited (most of us want to go to the moon if it was possible) then, for each good that falls in price, demand will shift to other goods. As automation lowers the cost of each additional good, a person is able to consume more of that good or more additional goods. I'm well aware of the Luddite Fallacy :) While it is true that the economy has thus far been able to re-absorb a workforce displaced by automation, we should not take as a given this will be the case forever. Certainly not based on the premise of "infinite human wants". Frankly, I am not convinced our desires are in fact infinite. Our demands do adjust with opportunity, but only to a point. For instance, many people are satisfied after achieving a certain relative level of security and comfort. However, more important than demand is how many workers are required to meet the supply. If innovations in automation continue at even a linear pace (and, frankly, we're probably talking quadratic), then we will eventually reach a point where staggeringly few actual workers are required for the production of a given good. On a slightly less speculative note, a documented phenomenon of workplace automation in developed nations in recent years is the increase in income & labor polarization (or, inequality) [1]. Essentially, while total employment is somewhat maintained, skill (and, to a large extent) income become polarized at the ends. From the MIT study [1] : In net, employment changes in the U.S. during this period were strongly U-shaped in skill level, with relative employment declines in the middle of the distribution and relative gains at the tails ... Technological progress in our model takes the form of an ongoing decline in the cost of computerizing routine tasks, which can be performed both by computer capital and low skill (‘non-college’) workers in the production of goods. The adoption of computers substitutes for low skill workers performing routine tasks–such as bookkeeping, clerical work, and repetitive production and monitoring activities–which are readily computerized because they follow precise, well-deﬁned procedures. Importantly, occupations intensive in these tasks are most commonplace in the middle of the occupational skill and wage distribution The same paper does mention that some low skilled workers are able to shift (and even see some relative wages rise) towards service economy jobs. However, that certainly doesn't change the macro picture. Lastly, in the near-mid term the greatest challenge is most likely not the absolute out-and-out displacement of work, but rather that displacement vastly outpaces the ability of the economy to re-generate employment. >> This has been going on for a millenia, and there is no tipping-point or transition point, as it is a very smooth curve. A very smooth J-curve perhaps. Implying that historical trends over the past millenia are a guaranteed predictor of future events is a highly dubious claim. We are living in a relatively unprecedented age of advancement and growth by just about every possible measure, and it's hardly a given that what has held true in the past (e.g. "Luddite's fallacy") will always hold true. [1] http://economics.mit.edu/files/1474 http://economics.mit.edu/files/1474 [2] http://www.economist.com/blogs/babbage/2011/11/artificial-intelligence http://www.economist.com/blogs/babbage/2011/11/artificial-in...