Thursday, 28 March 2013

Will humans be needed to screw in light bulbs?


How many people does it take to screw in a light bulb? None, the light bulb screws itself in now. Once humans are out of the equation, they don’t receive money for screwing in the light bulb, thus they have no money to pay for light. But should a human benefit from the light bulb? Without humans, was it even necessary for the light bulb to screw itself in? A human comes along with an idea for a better self-screwing light bulb, but no one has the money to pay for it. What then?

Productivity is up, wages are down; capital is up, jobs are down. It is pure idealism to think that technology has played no role in that. Whether it is a temporary situation or more structural, only time will tell.

Some people suggest the crisis is not from rising automation, but from a lopsided distribution of wealth. While I’m happy to concede this might be the case, I’d be less certain about implementing any policies based on this finding. Plenty of folks seem to cry loudly for redistribution (the OWS for instance) without realising that doing so inhibits entrepreneurial spirit and investment. If the government is simply going to forcibly take one’s profits and give them to others, what motivation would one have to initially invest? Sure, we’re all struggling after the GFC, but a top-down approach is sure to mess things up. I guess to cheapen a summation I’d say a sloppy understanding of socialism is not the answer here.

While I can agree that the nature of jobs will continue to shift away from manufacturing especially, the argument discounts the power of supply and demand to assume large swaths of the population would be unemployed. As unemployment increases, the cost of labour decreases. At some point, there should be a crossing of the cost/benefit curves of machines versus men. Then, employment will once again increases. It could be argued this is a race to the bottom, and that income divide would markedly increase. But that is an entirely different conversation altogether.

That said, a far more likely scenario is that the enterprising and entrepreneurial among us will continue to employ the population, because it is a usable resource. I can’t name a single natural resource which isn’t used by someone to produce something. I’m loathe to discount human ingenuity to find us all something to do. Somebody somewhere will always be willing to trade a little bit of their money, for a little bit of your time.

Then again, as I say in my article, I’m not sure a bottom-up approach will work either. After all, if profits are priority one, and jobs are increasingly completed by machines, then we can’t trust the market to fix unemployment. When I say there “isn’t enough work to go around”, I’m implicitly assuming the amount of total demand for stuff is fixed. But if machines can produce X amount of stuff, why won’t they be the preferred way to produce 5X the amount of the stuff? Why will we use inefficient human labour to make up the surplus? Once a job is taken over by machines it’s probably gone forever, irrespective of the amount of stuff required. So capitalism doesn’t seem to be the answer either.

But then there’s “just-in-time” inventory and centralised planning. With the computing power we now have at our disposal, centralised planning could work. At least for core products and services (staple foods, essential garments, energy, water, etc). I’d prefer a free marketplace remains for ideas and innovation, but the core items could be managed centrally. The more “commodity” the industry, the more that centralisation seems to make sense. The more “creative” or “intellectual” the industry, the more that a laissez-faire approach will encourage innovation and steady improvement.

Basically, the question seems to be: are we done with defining the industry’s inputs and outputs? For something like agriculture, maybe so. For web-design, clearly not. But even agriculture has plenty of technological innovation going on all the time, so it’s only the manual labour part which is a “commodity”. So, better to ask: “can we automate the given industry?”, and if so, “how much?”. Where do we still need/want human beings to be involved.

A good example is online banking. When banks figured out they could offload all the transaction work onto customers (debit, credit, transfers, etc) they were extremely keen. This has saved banks millions of dollars in redundant human workers. Another example is self-checkouts at supermarkets. While they’re not everywhere, they are preferred by customers in shops using them.

I saw another example last year. HOP cards in Auckland have removed almost entirely the need for ticket conductors on Auckland trains. Once the plan to implement them all over Auckland public transport wraps up, why would one need conductors at all? I can foresee the transport machines themselves becoming automated in the future. If it saves business owners money to get rid of drivers, then why wouldn’t they? If there were technologies in place to produce and distribute everything a person might need, why not go for that?
There are already technologies in the works to make this a reality. Google’s self-driving cars, for instance, could distribute food from where it’s grown to where it’s consumed. And then there’s the entire field of robotics. If we can design self-repairing robots, and robots which do all the work for our basic necessities, at some point it’s going to become easier to simply give that stuff to the people who need it rather than trying to sell it. The same goes for clothing, and shelter, and water. Automated systems can be designed, and they can be built, and they can be maintained indefinitely by robots. If people don’t have to pay for necessities in the way they do now, not having to work shouldn’t bother them. But this changes society immensely.
Yet while human labour may not be necessary to create more efficiency and thus capital, most of the economy is composed of unnecessary activities that exist purely for the sake of economic opportunity. In other words, while technology increases the efficiency of the transportation industry such that driverless cars may replace taxi, bus, and truck drivers, that logic applies internally with the goal of improving the efficiency of the transportation industry. The economy as a whole can create new jobs that don’t require efficiency, like making oreos. Nobody “needs” oreos, but they exist because someone can make money producing them and essentially creating a market for their existence
Another way of looking at it is this: the shift to robotic labour will also be one of mind. If we want something in the future, or want to provide a service or product to someone, figuring out a way to do it will be a matter of either (a) figuring out how to “program” a machine to make it for you or (b) simply asking a machine to do it for you. The idea of getting a person to do something for you will seem strange.
It would be a little bit like asking for a way to cross-reference phone numbers of all the people in a city with their street address and show it on a map. If I asked someone from 50 years ago to do this, they would come up with some sort of an index card system to do this whereas to us this is clearly a computational problem. Get the right databases and cross-reference them on a computer. It would never even occur to us to figure out another way to do it. In this way, it would never occur to someone in the future to figure out how people can make things in a factory, or accomplish simple service tasks.
Where I worry, is that western society is founded on principles of work ethic. We compel people to work with things like healthcare and debt to a point where the decision is really work or die. The truly sick part is that we maintain this pressure cooker, in the face of an economy that is rapidly outgrowing a need for workers. Something needs to be done.
Kurt Vonnegut, the modern American prophet, wrote a book about this called “Player Piano”– brilliantly titled. If you haven’t already, I highly recommend picking it up if you’re interested in a future where a significant portion of the population is not needed to work to keep things going.

What about the most interesting question, one I didn’t ask in my previous article: How do business people who want to make money think about making money if nobody else has money to spend, due to not being able to work because there are no jobs? My point is exactly that, as it starts to make less and less economic sense to keep us employed, what happens to us? Do we really need to maintain such draconian social constructions to keep us working in a society that is doing more and more with less and less labour?

No comments: