Tag Archives: Jobs

Automation: So Far, Business As Usual

Since at least 2013, many have claimed that we are entering a big automation revolution, and so should soon expect to see large trend-deviating increases in job automation levels, in related job losses, and in patterns of which jobs are more automated.

For example, in the October 15 Democratic debate between 12 U.S. presidential candidates, 6 of them addressed automation concerns introduced via this moderator’s statement:

According to a recent study, about a quarter of American jobs could be lost to automation in just the next ten years.

Most revolutions do not appear suddenly or fully-formed, but instead grow from precursor trends. Thus we might hope to test this claim of an automation revolution via a broad study of recent automation.

My coauthor Keller Scholl and I have just released such a study. We use data on 1505 expert reports regarding the degree of automation of 832 U.S. job types over the period 1999-2019, and similar reports on 153 other job features, to try to address these questions:

  1. Is automation predicted by two features suggested by basic theory: pay and employment?
  2. Do expert judgements on which particular jobs are vulnerable to future automation predict which jobs were how automated in the recent past?
  3. How well can we predict each job’s recent degree of automation from all available features?
  4. Have the predictors of job automation changed noticeably over the last two decades?
  5. On average, how much have levels of job automation changed in the last two decades?
  6. Do changes in job automation over the last two decades predict changes in pay or employment for those jobs?
  7. Do other features, when interacted with automation, predict changes in pay or employment?

Bottom line: we see no signs of an automation revolution. From our paper‘s conclusion:

We find that both wages and employment predict automation in the direction predicted by simple theory. We also find that expert judgements on which jobs are more vulnerable to future automation predict which jobs have been how automated recently. Controlling for such factors, education does not seem to predict automation.

However, aside perhaps from education, these factors no longer help predict automation when we add (interpolated extensions of) the top 25 O*NET variables, which together predict over half the variance in reported automation. The strongest O*NET predictor is Pace Determined By Speed Of Equipment and most predictors seem understandable in terms of traditional mechanical styles of job automation.

We see no significant change over our time period in the average reported automation levels, or in which factors best predict those levels. However, we can’t exclude the possibility of drifting standards in expert reports; if so, automation may have increased greatly during this period. The main change that we can see is that job factors have become significantly more suitable for automation, by enough to raise automation by roughly one third of a standard deviation.

Changes in pay and employment tend to predict each other, suggesting that labor market changes tend more to be demand instead of supply changes. These changes seem weaker when automation increases. Changes in job automation do not predict changes in pay or employment; the only significant term out of six suggests that employment increases with more automation. Falling labor demand correlates with rising job education levels.

None of these results seem to offer much support for claims that we are in the midst of a trend-deviating revolution in levels of job automation, related job losses, or in the factors that predict job automation. If such a revolution has begun, it has not yet noticeably influenced this sort of data, though continued tracking of such data may later reveal such a revolution. Our results also offer little support for claims that a trend-deviating increase in automation would be accompanied by large net declines in pay or employment. Instead, we estimate that more automation mainly predicts weaker demand, relative to supply, fluctuations in labor markets.

GD Star Rating
loading...
Tagged as: , , ,

The value of time as a student

When I was at college, many of my associates had part time jobs, or worked during school breaks. They were often unpleasant, uninspiring, and poorly paid jobs, such as food preparation. Some were better, such as bureaucracy. But they were generally much worse than any of us would expect to be after graduating. I think this is normal.

It was occasionally suggested that I too should become employed. This seemed false to me, for the following reasons. There are other activities I want to spend a lot of time on in my life, such as thinking about things. I expect the nth hour of thinking about things to be similarly valuable regardless of when it happens. I think for a hundred extra hours this year, or a hundred extra hours in five years, I still expect to have about the same amount of understanding at the end, and for hours in ten years to be about as valuable either way.

Depending on what one is thinking about, moving hours of thinking earlier might make them more valuable. Understanding things early on probably adds value to other activities, and youth is purportedly helpful for thinking. Also a better understanding early on probably makes later observations (which automatically happen with passing time) more useful.

This goes for many things. Learning an instrument, reading about a topic, writing. Some things are even more valuable early on in life, such as making friends, gaining respect and figuring out efficient lifestyle logistics.

Across many periods of time, work is roughly like this. It is the total amount of work you do that matters. But between before and after graduating, this is not so!

If activity A is a lot more valuable in the future, and activity B is about as valuable now or in the future, all things equal I should trade them and do B now.

Yes, work before graduating might get you a better wage after graduating, but so will the same amount of work after graduating, and it will be paid more at the time. Yes, you will be a year behind say, but you will have done something else for a year that you no longer need to do in the future.

On the other hand, working seems a great option if you have pressing needs for money now, or a strong aversion to indebtedness. My guess is that the latter played a large part in others’ choices. In Australia, most youth whose families aren’t wealthy can get enough money to live on from the government, and anyone can defer paying tuition indefinitely.

It seems that college students generally treat their time as low value. Not only do they work for low wages, but they go to efforts to get free food, and are happy to spend an hour of three people’s time to acquire discarded furniture they wouldn’t spend a hundred dollars on. This seems to mean they don’t think these activities they could do at any time in their life are valuable. If you are willing to trade an hour you could be reading for $10 worth of value, you don’t value reading much. When these people are paid a lot more, will they give up activities like reading all together? If not, it seems they must think reading is also more valuable in the future than now, and the relative values are jumping roughly in line with the value of working at these times. Or do they just make an error? Or am I just making some error?

GD Star Rating
loading...
Tagged as: , , ,

Jobs Explain Lots

Different people do different things; why? When we look features of individuals to explain their differing individual behavior, there are a few favorites: age, gender, race, income, education, IQ, and personality-type. Some people look at location, such as zipcode or nation. But it seems to me that one’s job (i.e., occupation) is a neglected strong predictor of many interesting things. For example:

  • A few days ago I blogged on a recent study of how jobs predict the chances of divorce. Job risk-ratios range over about a factor of two, after controlling for age, gender, race, and income.
  • I start my health econ class with this ’99 study of how jobs predict death rates. Job risk-ratios range over about a factor of two, after controlling for age, gender, race, income, and education. (Key chart below the fold.)
  • A February analysis found occupation strongly predicts the direction of political contributions, and an ’07 study said academic discipline strongly predicts professor political affiliation. This page of aneqdotes suggests that jobs often predict political affiliations well.

More generally, I’d love to see a factor analysis seeking the few strongest job factors that can simultaneously predict variations in divorce, mortality, political affiliation, and whatever else interesting one can throw into the mix. Seems like a great project for a data-oriented grad student. Continue reading "Jobs Explain Lots" »

GD Star Rating
loading...
Tagged as: , ,