Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The Automation Myth

vox.com

Post reply on HN
Print view View original

31–40 of 54 posts

Re: The Automation Myth

#31

Post by yasth »

aetherson wrote: post #27
Terribledactyl wrote: post #11

This article was very confusing and difficult to read. There seems to be a conflation of workers being replaced 'robots' (Human teller -> ATM) and the process by which humans are simply more productive through augmentation (Manager with paper ledger -> Manager with inventory system). There are some nasty potential error sources when comparing productivity per country over time because they use different currencies wi…

It's all the same thing from a high enough level. If your manager can do the work of two managers because he has an inventory system, you don't suddenly manage twice as much inventory (because everything else about your business doesn't change at the same time), you fire every other manager. But the point is that society as a whole is richer because you can get a good from a manufacturer to a customer with less overa…

> you fire every other manager.

Except managers are "sticky", and generally aren't fired merely because they are less burdened. All jobs not just managers are in truth sticky, but in general the more status a job has, and the more high status interactions a job has the greater the difficulty in removing the position.

A lot of companies became much more productive on paper during the recession, not because they did massive changes, but because they were forced to finally cut some staff. Like a cheesy 80s movie they had the productivity inside themselves all along, but it took the need to trim budgets to expose that productivity.

Of course a lot of companies never actually really pared down things, or not enough to really boost their productivity. It just isn't simple to get that productivity to reveal itself, especially outside of an existential threat.

Re: The Automation Myth

#32

Post by Retric »

pdonis wrote: post #26
chipsy wrote: post #21

An underlying premise of productivity tracking is that we are able to consume more indefinitely. But in all likelihood there is a logistic curve to human material needs as well. If we did have $30,000 more on average, as the article states, how would it be used? The case is obvious for the poor, but as you go to higher income brackets more and more of the possibilities fall into "luxury" and "status symbol". Take, fo…

> in all likelihood there is a logistic curve to human material needs as well. Material needs, probably yes. But not all needs are material needs. > as you go to higher income brackets more and more of the possibilities fall into "luxury" and "status symbol". Those are loaded words; one person's "luxury" is another person's preferred entertainment. > It's absolutely bizarre that in an era of low fiction information,…

Status based economies are fragile.

There have been a few waves such as the Beatniks where people say eff status I want to relax. Electronics pushed that off for a while, but that treadmill seems to have slowed down. So, I expect the next wave of young people to accept they can live well on a 2 day a week job. https://en.wikipedia.org/wiki/Beatnik

Such is fine for 20-30 year olds without health issues or kids, but does not keep widget factories staffed or pay for pension checks.

Re: The Automation Myth

#33

Post by exelius »

This is an awful article. Let me count the ways:

1. You don't measure productivity by the number of hours worked. There are just different cultural norms at play and it doesn't make sense.

2. Comparing an economy that started out with low productivity and grew quickly to an economy that started out with high productivity and grew more slowly isn't really telling me anything other than the low-productivity economy caught up. Which happens, because it's harder to push the frontier than follow in someone else's footsteps.

3. The effect of automation is very hard to quantify because it tends to displace low-skilled jobs that don't pay much in the first place. But the social impact of this is huge -- you take the people at the bottom of the economy and take away the only jobs that were accessible to them. Social unrest is inevitable. This will likely impact the economy in negative ways in years to come, and we need to understand this.

Anyway, all around a terrible article. The author has no grasp of basic economics and seems to have cherry picked statistics that he thinks back up his claims. Has anyone created a "top 10 logical fallacies of data analysis"? Because this article would tick all 10 of them.

Re: The Automation Myth

#34

Post by stdbrouw »

quanticle wrote: post #28

The power of Moore's Law — which states that the power of computer chips doubles roughly every two years — is such that the next five years' worth of digital progress will involve bigger leaps in raw processor power than the previous five years. It's at least possible that we really will have a massive leap forward in productivity someday soon that starts substantially reducing the amount of human labor needed to dri…

> But out in the wider public, there seems to be a perception that chips will get "twice as good" every year into perpetuity.

I'd think people in the wider public have no clue as to exactly how much faster computers get every year. Considering that software usually gets slower at about the same rate as processors get faster, I'd imagine many people would think the increases in speed are much smaller than they actually are.

Re: The Automation Myth

#35

Post by mkozlows »

norea-armozel wrote: post #2

This is something that I've always found odd about how the actual value in the economy is tracked. Specifically, how the stock markets continue to rise despite the decline in productivity. If I was told as a shareholder or private owner of a company, that an hour of labor and inputs are making me mess than I was making 20 years ago I would be upset but that doesn't seem to be reflected in the volumes and prices of th…

To make this clear: Productivity is not declining. The growth of productivity is slower than it was in the past, but it's still growing. The first graph here is annualized growth rates, and you'll note they're all positive: http://www.bls.gov/lpc/prodybar.htm

Re: The Automation Myth

#36

Post by Taek »

This article encouraged my to look at why my life was so expensive.

1/3 of my expenses are related to living in Boston. Rent is 2/3 of my total expenses, and 1/2 of my rent is only because of the city I live in, if I chose a cheaper city or lived 15 minutes further away, I could have the same house for 1/2 the rent. Of the remaining 1/3 expenses, many of those are cost-of-living related as well.

Location is something that's never really going to get cheaper, because location is fundamentally scarce. You can have as many cities as you want, but you can only have 1 SF, 1 Boston, etc. I don't want to live in a city, I want to live in Boston. And that's expensive compared to other cities around the world.

I think this is a testament to how cheap technology has made our lives. 100 years ago, things like clothes and food were a much more significant part of lifetime expenses. And as technology continues to improve, location is going to continue to eat up a larger portion of total expenses.

Re: The Automation Myth

#37

Post by itsybitsycoder »

That first graph stinks. It only counts job work hours worked per job worker, not work hours per person. In 1950 there were an awful lot of households with one adult working a paid job and one adult handling most of the household work. According to this graph, that one worker did 1900 hours per year plus a bit of household work, and the other adult did ??? hours of work. Now both adults are doing 1700 with the household work on top. Granted, the amount of household work has gone down, but has it really gone down by over 1500 hours per year? If not, we're working more, not less.

Re: The Automation Myth

#38

Post by strange_quark »

exelius wrote: post #33

This is an awful article. Let me count the ways: 1. You don't measure productivity by the number of hours worked. There are just different cultural norms at play and it doesn't make sense. 2. Comparing an economy that started out with low productivity and grew quickly to an economy that started out with high productivity and grew more slowly isn't really telling me anything other than the low-productivity economy cau…

> You don't measure productivity by the number of hours worked. There are just different cultural norms at play and it doesn't make sense.

The article doesn't. It measures productivity in terms of actual monetary value per worker and then compares it to the number of hours worked.

> The effect of automation is very hard to quantify because it tends to displace low-skilled jobs that don't pay much in the first place. But the social impact of this is huge -- you take the people at the bottom of the economy and take away the only jobs that were accessible to them. Social unrest is inevitable. This will likely impact the economy in negative ways in years to come, and we need to understand this.

Again, addressed by the article. Would you have considered milkmen or household servants high-skilled jobs? Both these jobs have been all but replaced for the majority of people, yet there wasn't that much social unrest when the refrigerator or washing machine was invented.

Re: The Automation Myth

#39

Post by AnimalMuppet »

VLM wrote: post #29
aetherson wrote: post #27

Earlier quoted context omitted.

It's all the same thing from a high enough level. If your manager can do the work of two managers because he has an inventory system, you don't suddenly manage twice as much inventory (because everything else about your business doesn't change at the same time), you fire every other manager. But the point is that society as a whole is richer because you can get a good from a manufacturer to a customer with less overa…

"and there are more jobs created than destroyed" Why? Well, I know its tradition to say it, but there appears to be no causal relationship, just occasional coincidence. The mathematical model would be interesting to see.

Just occasional coincidence? Looks to me more like there's very common (almost uniform) coincidence, with occasional exceptions. But we don't actually call that "coincidence", we call it "normal behavior".

Re: The Automation Myth

#40

Post by gitah »

Taek wrote: post #36

This article encouraged my to look at why my life was so expensive. 1/3 of my expenses are related to living in Boston. Rent is 2/3 of my total expenses, and 1/2 of my rent is only because of the city I live in, if I chose a cheaper city or lived 15 minutes further away, I could have the same house for 1/2 the rent. Of the remaining 1/3 expenses, many of those are cost-of-living related as well. Location is something…

Autonomous vehicles or high-speed subways that are cheap to build will destroy housing prices. Technology can reduce the cost of transportation increasing the usable land around a city.

Other advances include new construction techniques making it cheaper to build taller buildings. ie prefabs or 3D printing

Post reply on HN