Assorted links

Comments

2. These numbers seem awfully fishy.

I concur. Based on the geographical regions impacted, my first thought is simply net migration of the healthier, wealthier individuals out of those areas, but migration isn't even mentioned.

Yes, but migration occurs for good reasons: poor health care, no jobs, no public services.

Notice the life expectancy is declining in the South, where the social net has been guttered by Republicans for mostly racist reasons. This is entirely expected based on the GOP's actions. People move to liberal areas with high taxes and activist governments, because life is better there.

He meant migration INTO the states.

Sorry, my error. Here's the problem with drawing too many conclusions from the study as presented by the LA Times. The most improved county is New York County, with a population of more than 19m. Three of the "worst" counties are in South Dakota. Their entire state has a population of 814K. Mellette County, one of the big decliners, has a population of a little more than 2,000. It doesn't take much to move that neddle. Sunflower County has the largest population of any of these counties at 68K.. Colier County in Florida, the smallest on the improved list, is bigger than all the decliners combined.

Ah, racism. The universal explanation.

Could the decline of life expectancy have something in common with massive epidemics of morbid obesity in the USA? Any European visiting the States is, frankly, shocked by the sight.

Of course, you may construct an elaborate connection between reductions of social expenses and people getting fatter. But I am not aware of any program (like "anywhere in the world") that would cut down obesity rates once they start to grow. High-energy food is cheap and people are built by Mother Nature to love it.

could it also have something to do with greater female participation in the workforce.

Could the decline of life expectancy have something in common with massive epidemics of morbid obesity in the USA?

Yes. But look at the data. Women's life expectancy has declined in about 1/4 of US counties. Men's has declined in a tiny handful.

Just from that, you'd expect we're talking about maybe 1/8 of the population who die younger, and 7/8 who on average live longer.

But as Ted Craig points out, it's mostly in places where there aren't a lot of people (plus some cities). So it's less than 1/8 the population.

Communities with large immigrant populations — Southern California, for example — fared considerably better than average despite relatively high poverty rates. The worst-performing counties were clustered primarily in Appalachia, the Deep South and the lower Midwest. In those places, women died as much as a year younger in 2007 than women did a decade earlier. Life expectancy for women slipped 2 1/2 years in Madison County, Miss., which recorded the biggest regression.

They blame smoking, obesity, and high blood pressure. Women who started smoking decades ago are have been dying of it more in the last 10 years. They didn't say so, but it's a Red State thing.

They say that healthy latino immigrants bring up the average some places, and argue that it isn't just that poor latinos are healthier when they get here because healthy people are more likely to immigrate -- because immigrants from other places don't show the effect. (This reasoning is not conclusive but it is suggestive.)

That 1927 dictionary was pretty optimistic: "rayon, n: A glossy fiber, resembling silk." At the very least, it's generous with its definition of "resembles."

re: Bank tellers, surely we can all agree that there is a trade off between labor and capital that firms make, and that trade off can have short run effects on employment?

LE has some interesting quirks. Hispanic populations actually have some of the best LE despite lower incomes, blacks have the lowest (besides the male mortality due to violence, Vitamin D deficiency is suspected) and notice the spread across U.S. counties is quite large.

My favorite LE quirk though was the horror displayed at Megan McArdle's suggestion that all the people who used U.S. LE as a basis for arguing for PPACA should measure the results. Such a fear of empiricism has been rarely seen outside medieval clergy..

The "Bank Tellers" link is to an Economist article that faults Obama for saying ATMs reduce the number of bank tellers. U.S. population is growing at something like 1.2% annually. They cite numbers that show 0.5% growth in the number of bank tellers from 1985 to 2002, a period in which I seem to recall bank branch offices springing up in every strip mall; and they project 0.6% growth from 2008 to 2018. So their point was what exactly?

Thanks, gVOR08

Your point is obvious, but missed by both the Economist and Marginal Revolution. Hmm, I wonder why ...

If bank teller jobs grew at the rate of population, we would have had 64,156 more bank tellers.

Why would you assume that the steady-state growth rate for bank tellers is equal to the population growth rate?

Did you entirely miss the point about the creation of new jobs with the introduction of ATMs, which is the largest part of why people are mocking Obama?

Who dares mock Obama? Must be racist hicks at work.

I'm certain that if President Palin were to blame 17% underemployment on ATM machines, we'd get entirely reasonable, fact-filled articles in places like The Economist disputing the assertion ever-so-politely, and concluding that it's natural to blame "whoever is in office" for the employment disaster, "no matter who is really to blame."

Uh huh.

"6. Meandering but interesting blog post"

There are so many disingenuous claims in that article, I wouldn't knwo where to begin. They start at the very offset, with the contrived model of "Culture-Politics-War-Business." Business, like Culture and Politics, are skeptical of technology where it appears to threaten extant organizations - and this can be just as or more damaging to tech than the other three groups.

Corporations aren't "ending." If anything, power will change its model of control - but power will remain accumulated; the powerful will continue their expansion of the same.

OK, ATMs reduce bank teller jobs. And increase other worker's productivity as well. We are not rushing to the bank some time during the day to get or deposit money. Increasing other workers productivity reduces jobs there as well as less workers are needed at other places of business. We cannot ignore the second and third order effects.

What do I say to that? Keep the ATMs. I should not have to pay more for banking and reduce my personal productivity to keep more bank tellers employed. This allows us to create more and better jobs in other areas. There may be a painful lag for those former bank tellers, but it is better for society overall in the long run.

This is an excellent point... It is not enough to just measure the number of employed bank tellers.

Of course ATMs reduce the need for bank tellers. But they do not obviate the tellers.

Imagine our world with no ATMs. We would have many, many more tellers.

On the other hand, by having fewer tellers, banks save money, which they can then use on something more profitable than tellers. One of those things might be lending money to people to be used as capital.

I would argue that ATMs are an economic efficiency that leads to a greater number of non-bank teller jobs and a lower number of bank teller jobs. I would further hypothesize that the number of jobs created is more than the number of jobs lost.

I would argue that ATMs are an economic efficiency that leads to a greater number of non-bank teller jobs and a lower number of bank teller jobs. I would further hypothesize that the number of jobs created is more than the number of jobs lost.

I was with you up to that last sentence.

ATMs do the same sort of transactions that tellers do, but cheaper and more efficiently. Because they're machines.

Tellers do the jobs that were not considered automating. Like, if I want a roll of quarters I go to the teller, because they didn't bother to let me do that at an ATM.

Doesn't that generalize? Anything that doesn't take sophisticated judgement or eyesight, can be better done by machines -- provided it's routine enough to be worth the bother of automating.

Things that take sophisticated judgement? Then the employer has to trust his employee, who could use his judgement to cheat him. And employees who're supposed to do that sort of work are not so unlikely to look for ways to go into business for themselves. This sort of work tends to pay well, but employers prefer not to create too many jobs like that.

Eyesight? Automation keeps incringing on that. Driving is the obvious thing that will never be automated, as note the previous discussion here about why it hasn't already taken over.

So what we're left with is lots of little niches, jobs where people can do routine work that is worth training them for but not worth automating because the returns are too low.

I'm not sure about this reasoning, but it seems to me more plausible than the simple assumption that jobs will automaticly show up for people who want them.

And yet, there is the example of the industrial revolution. A whole lot of people got pushed off the farms in England, because it was more profitable to do agriculture that didn't need them. And they got pushed into the cities where immediately factory jobs turned up for all of them, factory jobs that paid far better than farming ever had, and they were much better off.

So maybe it will happen like that again.

"Driving is the obvious thing that will never be automated" reminds me of Arthur C. Clarke's citation: "When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong."

You already have fully automatic metro trains, for example in Dubai.

Granted, it is easier to drive a train automatically than a car, but Rome wasn't built in a day.

I would gladly bet a bottle of good wine that you'll have fully automatically driven cars capable of driving through heavy traffic in 2020.

Full disclosure: I am mathematician by training and software developer by living, so I do have some insight into the corresponding algorithmic problems, and thus I am probably morally disqualified from taking bets on this topic.

I would gladly bet a bottle of good wine that you’ll have fully automatically driven cars capable of driving through heavy traffic in 2020.

Suppose you're right. Doesn't that eliminate a whole class of jobs? Things that currently cannot be automated, that will be automated.

If we have a system set up for fully automatic cars, what will be the marginal cost for automating away one more truck driver? Higher than a driver's salary? Is that likely?

It looks to me like we will have two remaining classes of jobs.

There will be jobs that are specialized or transient to the point that it isn't worth the cost of automating them, but it is still worth the cost of training people to do them.

And there will be jobs that involve decisions which humans can make, for which no algorithm can be found and which genetic algorithms etc do not work. Japanese flower arranging, perhaps. Playwrighting. Prostitution. Things where you can admire the humanity of the creator, knowing that it simply could not have been created by machine intelligence because machines lack that je ne sais quoi.

"And there will be jobs that involve decisions which humans can make, for which no algorithm can be found and which genetic algorithms etc do not work."

So true. Though people who observed Deep Blue playing with Kasparov often said that DB's actions had (non-human) beauty in them.

I would say that there is a rough criterion: if a job involves a lot of "toil" (by which I mean tiring, but not really challenging work), and if there are significant risks connected with the worker's fatigue or inattention or intoxication or sudden worsening of health (like a stroke), it is highly probable that this job will eventually be "outsourced" to autonomous machines.

In re #1

There goes the Japanese plan of getting out of debt.

Yes, of course ATMs reduce the number of bank teller jobs. This is the point of it all.

To find those tasks which we can mechanise and mechanise them, thus freeing up that labour to do other things.

We do indeed have unlimited wants and desires and a limited amount of labour to satisfy them with.

We do indeed have unlimited wants and desires and a limited amount of labour to satisfy them with.

Well, yes. But what if something other than labor is in shorter supply? Like, say, oil?

I try to imagine what I'd do if I owned the world. I might want to hire a bunch of people to build me an authentic medieval castle, using authentic medieval stonemason's tools. Or I might not.

Would I want a whole lot of lumber? Probably not, I could get more than I want for myself from a single big sequoia tree. I might rather have a whole lot of primeval forests. If I don't need to build cottages for 300 million americans, do I really need all that wood?

And the oceans? Wouldn't it be a good thing to restore the oceans? A single fishing boat could provide me with all the fish I want. Why have more?

The earth is rather overpopulated, isn't it? Maybe I could find jobs for 7 billion people, but when there's an oversupply of them wouldn't it be better to adjust the population down to meet my demand?

Even when it isn't just me, don't the laws of supply and demand work? When there are more people than the economy needs, won't supply and demand tend to drop the price of labor until supply falls to meet demand?

But what if something other than labor is in shorter supply? Like, say, oil?

Then if a technological change comes along that means we can get more of what we want, with less oil (and that doesn't otherwise impose worse costs than just using the oil), we should grab that change eagerly.

As for supply and demand, you forget that, to reverse Julian Simon's famous quote, every pair of hands and a brain comes with a mouth. You add more people to the world, you add more mouths that need feeding, and bodies that need sheltering, and minds that would like entertaining, and will do things like fall in love and want to get married and have a big celebration and so forth. Increasing the supply of labour increases the demand for labour. .

As for supply and demand, you forget that, to reverse Julian Simon’s famous quote, every pair of hands and a brain comes with a mouth. You add more people to the world, you add more mouths that need feeding, and bodies that need sheltering, and minds that would like entertaining, and will do things like fall in love and want to get married and have a big celebration and so forth. Increasing the supply of labour increases the demand for labour. .

That would be true only if the extra bodies could pay for what they need.

But if the people who can pay them only need X number of bodies for labor, why should they pay extra if they happen to have X+1,000,000 bodies available? By supply-and-demand they should pay less. Because when there is surplus labor, the price should go down, right?

So the extra million bodies don't provide more demand, they provide less demand. X + 1,000,000 workers demand less than X workers do.

And almost every machine comes with a plug / gas tank. You need to power it in order to work.

Given current hysteria about nuclear energy, I am not so sure that the contemporary easy availability of cheap electricity will last. Civilizations can't be run on wind power.

I'm not sure where you're going with this.

Are you suggesting that if we can afford to feed a whole lot of people, that we can use them for muscle power? We might be able to power things with human muscle that we can no longer afford to run on fossil fuels?

I'd certainly rather pull a rickshaw than starve, myself.

Actually, while writing this comment, I was thinking about the visible effect that the rising costs of gasoline had on use of bicycles.

In Czechia, the cost of 1 litre of gasoline grew from approx. 1,40 USD to 2,10 USD in last 3 years.

Quite a lot of rural folk reacted by switching from cars back to the human-powered bicycle for shorter trips.

Yuan, n: The monetary unit of the Chinese Republic, since January, 1914; also, a silver coin containing 23.98 grams of pure silver. It is equivalent to .644 of a haikwan tael, or .464 cents, U.S. Currency. Called also Yuan dollar.
!!!!!!!!!!!!!!!

from .464c to 15.45c today = 33 fold increase
Silver 23.98 grams of pure silver = $27.57 today = 594 fold increase

Comments for this post are closed