Back to the Future…for lunch

September 12th, 2015 at 1:52 pm

While I’m wide open to evidence that I’m wrong, I’ve been skeptical of the claim that the robots are coming for our jobs. To be technical, the economics question is this: is the pace at which labor-saving technology is entering the workforce accelerating? I’ll explain the italics in a moment.

There are various pieces of evidence suggesting that the answer is “no.” Most importantly, if the rate at which machines are replacing workers is increasing, then productivity growth—output/hours worked—should also be increasing. But it has been slowing.

One reason for slower productivity growth is diminished investment in capital goods—like machines—a trend that also doesn’t square with the acceleration hypothesis. By some measures, the lower rate of investment accounts for two-thirds of the deceleration in productivity growth (I discuss all this productivity/investment stuff here). It’s possible that businesses are investing a lot more in robotics and a lot less in everything else, but if you drill down in the investment data, you don’t see that either.

So, what we have is largely anecdote and our own observations. I don’t discount either, but an important nuance here is that when it comes to observations, humans are good at seeing first derivatives (rates of change) and less good at seeing second derivatives (changes in rates of change). We see that iPads and self-scanners are replacing waitpersons and cashiers but it’s hard for us to tell whether “labor-saving technology” (sounds benign when you put it like that) is coming more quickly than it has in the past.

Of course, this time might really be different (some smart people say it is).

Or, as this article I read the other day reminded me (h/t: KN), this time might not be very different at all. It’s about a new quinoa restaurant in San Francisco, called Eatsa, where you order and get your food without ever interacting with a person. [You may insert a quinoa joke here; I quite like it as does my 15 y.o. daughter; OTOH, my 13 y.o. would be happy if it never darkened her plate again.] See the pic below of a happy customer getting her order, placed on an iPad, out of a cubbyhole.

Now, where have I seen that before? Fifty years ago (!), I used to love to go to Manhattan automats, where, as you see in the other pic, a few coins would get you a sandwich, a veggie (not quinoa!), a slice of delicious pie, and so on. For the record, productivity growth was faster and unemployment was lower back then (though at 10, I don’t recall knowing these facts at the time).

All’s I’m saying is that tech change is always with us, and it’s really hard to tell by observation whether the pace with which it’s replacing workers is accelerating. And there are so many more moving parts to this. I’d bet a big difference between the economies in these two pictures is where the machines were manufactured. In other words, technology doesn’t historically kill labor demand. But it does move it around to different industries, occupations, and today, countries.

So before we conclude we’re all robot fodder, let’s see it in the productivity and investment data.

Now, stop hogging all the quinoa!

auto12

Source: NYT (Eatsa)

 

 

Print Friendly

13 comments in reply to "Back to the Future…for lunch"

  1. Richard H. Serlin says:

    Check out this robot brick layer in the current MIT Technology Review:

    http://www.technologyreview.com/news/540916/robots-lay-three-times-as-many-bricks-as-construction-workers/

    In my recent post (in Mark Thoma’s links):

    http://richardhserlin.blogspot.com/2015/08/my-response-to-shiller-on-stock-prices.html

    I talked about how (most) real estate looks like a risky and bad long term investment in part due to the risk of the very real possibility of robot/AI revolution decimating employment, employment security, and wages for the 80%+, and thus their ability to bid on homes and apartments, and in part because these advances may in the not too far future make it possible to build new homes and offices for half, or a quarter, as much, and with much higher quality and features. This robot bricklayer article supports that (as well as a great deal of other evidence I have seen, like in the books The Second Machine Age and Rise of the Robots). I haven’t run the numbers, but it appears that already, at its current sale price, this robot cuts the cost of bricklaying more than in half, and cuts the number of middle class and skilled bricklayer jobs by about three-quarters. And it can easily do some fancy high quality things that are now difficult and rare.


  2. Richard H. Serlin says:

    It’s not just the pace of new machines, and how fast they’re eliminating labor. The pace of labor elimination with these new machines may be going down, but the pace of new jobs created may go down faster (see trend in labor force participation), in part as these new machines may be much more substitutive then complimentary. But also there are other effects, like the machines may be causing lower wages, not just unemployment. But if the market wage starts getting pushed below minimum for more and more people,…

    I would also look at the drop in the labor force participation rate; that’s hard to explain away. And I would note that the statistics you quote are recent and near term. The big danger to employment and wages from robot/AI revolution may really be 5, 10, 20 years from now given the track these developments are on.

    And as far as the vending machines of 50 years ago, this quinoa thing is obviously much higher quality and more of a threat to restaurant jobs.


  3. wendy beck says:

    I’m more concerned about the gig economy than I am about robots taking over everyone’s job. The gig economy, or “on-demand workers” will, if Uber and the like succeed, make it impossible to predict your wages, your hours, etc. I can’t imagine buying a home with that kind of uncertainty.

    I’d be interested in hearing your point of view on this.


  4. Beth says:

    In my opinion, “trickle-down” doesn’t work, the Job Creators won’t, and the investors aren’t interested. If only we had a way to create jobs performing needed work on the public’s behalf…oh wait…

    Americans and many others are working way too hard for far too little return on their energy investment. The old system is stagnant and damaging to our society.just as it was in the Gilded Age. The labor force can’t participate if there aren’t enough jobs nor loans for start-ups, nor grants for living during the time between investment and earnings.

    If we take a cue from the world of stocks and “split” all of the existing jobs, we could have more personal time with which to start our own businesses, more time for whatever we decide to do (or not do), and more economic stimulus due to the higher employment rate as a result of job-sharing and new business creation IF we provide income subsidies where needed for the entire transition. I’d love to patronize a really good Automat with Art Deco design features.

    Mechanization can be great only if its impacts aren’t toxic (creating ever-increasing new expenses/losses without commensurate public benefit), and the profits it creates generate enough revenue which is then redistributed throughout the populace to replace the lost income with services and subsidies. Otherwise we’re just robbing people of the ability to earn income in a consumer society. That’s not Progress. How does a society maintain its cohesion under such a condition? Not peacefully or well, as we know.


  5. Daniel says:

    I look at the self checkout cash registers at stores like CVS (or self check in at the airport) and it seems that those new technologies are transferring work from the worker to the consumer. But it’s hard to see why those types of technologies would produce huge increases in productivity. Is it possible that what’s going on at CVS is reflective of broader changes? Or am I missing something obvious?


  6. Michael Duda says:

    I’m reduced to head-scratching every time I see an article take this tack. The author’s intention seems good: show me “productivity and investment data”, but I can’t follow the reasoning. Investment and the resulting productivity are _lagging indicators_. This entire argument is really driven by people with an (in)ability to see the writing on the wall.

    For example, does Jared not think that self-driving vehicles will eliminate most of the livery and logistics drivers simply because the current annual investment is now only in the tens of millions—tiny in comparison to the trillion dollar automotive manufacturing industry—and you can’t buy even ONE today?

    Or perhaps he thinks that car sharing won’t continue to impact that same industry as shared ownership drives down total sales? Or that aggregate productivity won’t experience a significant bump when traffic is more efficient and people can work during their commute? All within a 10-20 year adoption window? Should we pretend this won’t happen in our lifetime because it hasn’t already? That’s not argumentation, it’s denialism.

    Now extend this thinking to other areas of manufacturing and service: “But you need people for that” is a killer argument…until you don’t need them any more. It’s turtles all the way down, or at least far enough that required human labor could easily become something of a quaint anachronism over the next 100 years—with BIG changes visible in the next decade or three. Laughing at robo-restaurants is very clever (tell that to the 420,000 McDonalds employees and see how many share your mirth).

    The difference between a 7/8ths and 8/8ths “good enough” solution (or one that provides a 10x improvement in productivity for the one worker you need) is what we’re discussing. And if we’re only at 3/8ths now, I don’t think this implies it will never be enough.

    Other commenters seem to get this (bricklaying / 3D printing buildings, etc.). I’m frankly a little baffled that everyone doesn’t see it, since things not expressly forbidden by the laws of physics tend to happen whether we want them to or not. Robocentric production and service is a given for me at this point, and I think it’s high time we stop debating the “if” and start considering how to keep the population happy and fed “when” it happens.


  7. uncle jo says:

    Ray Kurzweil has much to say about the accelerating power of technology https://www.youtube.com/watch?v=IfbOyw3CT6A

    He makes a very convincing argument and I think it’s related to robotic/machine/AI management, production and distribution of most everything…at no cost: The Autonomous, Post Scarcity NonEconomy. A topic for a book, if it hasn’t been written already.


  8. Smith says:

    This is all crazy talk. 3% of the workforce is involved in agriculture. Admittedly this doesn’t account for tractor manufacturers, oil producers for the tractors, grocers, etc. There are some essential clothing items, though mostly produced overseas. And replacement housing stock lags, even though McMansions are sitting empty. As Galbraith wrote over 50 years ago, we are an affluent society. Moreover, our economic activity is not largely not one of necessity. That being the case, robots do not matter, we already have been replaced by you know, threshers, harvesters, and all manner of machinery. It matters much, much less to our well being and way of life if machinery now becomes even more nimble and smarter to assume the few duties we still feel obligated to perform. You are fooling yourself to think otherwise.
    However, distribution of wealth and income is an ongoing issue and capitalists can be counted on to blame the robots to avoid responsibility for a dysfunctional economic system. Don’t play that game. Robots shmobots. Machines already do all the heavy lifting. Seriously.


    • Smith says:

      From 1963 https://www.youtube.com/watch?v=RsDFYUD5KUA

      Taking the opposite view, just how little work we (or robots) really need to do today is unknown, because people are brainwashed into thinking what they do matters. Sometimes stuff needs to get done. Other times we’re working to make our car payments. We need a well running car because we use it … to get to work. Our housing costs are a bit inflated because we need to … be reasonably close to work. Our education costs are sky high because a prestigious degree enables us to … be more successful at work. Everything is inflated because people with money bid up the price of everything, including the hours necessary to hold a job. It’s the gentrification of everything. That’s not including the amount of money we spend on defense. I’m not saying we don’t need to keep high levels or readiness, I’m just remarking on the price we pay.


  9. HCE3 says:

    Autor out of MIT has done some wonderful work here that I think does a good job of diluting down the economics of automation (and why technological unemployment is not really a concern) which you can find in the latest JEP https://www.aeaweb.org/articles.php?doi=10.1257/jep.29.3 The other two papers are also very good.


  10. Manley says:

    I personally think that we are heading in the direction of an AI abundant economy. As you mentioned per example, certain businesses are already taking advantage of the technological advancements that we are continuing to see. From what I have researched, technology will continue to move at an exponential rate. And, knowing most businesses owners, they are probably going to take advantage of these advancements as soon as they can afford to. Most businesses are short sighted, so as soon as they see the direct opportunity to cut costs, I’m sure they will take hold of it. Technology will likely be a decent monetary investment for a few more years, but there will be more and more companies offering consulting in and installment of technological makeovers for a low price.

    The bigger question in my mind is, “what is going to happen to the lower and middle class workers that aren’t necessary any longer?” When technology gets to the point of being able to handle all tedious and “labor” type tasks, there will be massive layoffs. I’m curious to what te plan will be.


Leave a Reply

Your email address will not be published.