It is time to stop quivering in our boots in pointless fear of the future and just roll up our sleeves and build it.
- Ray Pierrehumbert

Wednesday, June 15, 2011

Automation: Obama vs The Economist

The Economist chooses exactly this moment to address the Automation Crisis and the current employment situation! They start by quoting O:
There are some structural issues with our economy where a lot of businesses have learned to become much more efficient with a lot fewer workers. You see it when you go to a bank and you use an ATM, you don’t go to a bank teller, or you go to the airport and you’re using a kiosk instead of checking in at the gate.
And they refer to this blog entry by Karl Smith:

Lets take some obvious examples. Suppose to create welded metal I need both a welder and welding torch. The welding torch goes down in price. That means that its actually cheaper to create each piece of welded metal. This will allow me as a factory owner to either lower my price, sell more welded metal while maintaining my profit margin.

However, to do this I will need more welders. So a fall in the price of welding torches, increases the demand for welders.

On the other hand suppose that I am an airline considering whether to have more booking agents or whether to invest in more sophisticated booking software. Specialized software can run well into the multi-millions but if it gets just cheap enough it might actually be a better deal than new agents.

In the end, The Economist does me the favor of concluding:
I wouldn't blame ATMs on our jobless recovery, but surely the general skill-bias of technological change is an important part of the issue. I suspect Tyler Cowen may be right that the recession created an occasion for firms to shed "zero-marginal-productivity workers". In that case, the ranks of the unemployed are filled with wannabe workers whose labour is at present worth less to employers than the cost of employing them. This puts Mr Obama in a politically perilous position. We can expect rising aggregate demand to make it pay for some firms to once again employ some significant number of relatively low-productivity workers, but we probably can't reasonably expect the unemployment rate to return to its pre-recession level, at least not in the absence of politically unlikely employment subsidies or government make-work schemes. Given the current creeping pace of growth, the unemployment rate may not improve very much before next fall, which would bode ill for incumbents. Mr Obama can blame it on the machines and deny Republican charges that his administration made the recession worse. But jobless voters and the voters that love them tend to blame the guy in office, no matter who's really to blame.
Yeah. Something like that. This is on the time scale where conventional economic thinking makes sense to me, and that's pretty much what I think. Losing the last election eliminated the possibility of "politically unlikely employment subsidies" which clinches a very long period of relative unemployment.

Where I part company from the mainstream is only in thinking it's likely permanent, and we ought to get used to it. That is, I expect them to get the short run right and the long run wrong, as usual.


Rich Puchalsky said...

I know that I'm supposed to err on the side of civility, but that discussion is awfully stupid. The zero productivity worker links leads to a blog post by Tyler Cowen. Cowen wonder why output has recovered even while employment hasn't, and of course hypothesizes, a ha, they must have been zero productivity workers. Then he looks at the characteristics of the low-term unemployed and sees that a big chunk of them are older, highly educated workers. Were they zero productivity too? What a puzzle!

Libertarians being who they are, it will never ever occur to Cowen or any of the other people blithering on about productivity that there is a simpler, alternative explanation. Output recovers while employment doesn't because employers use widespread unemployment to squeeze concessions out of workers. They work harder, for longer hours, for the same pay, or it's made clear that they will lose their jobs. Why do older educated workers stay unemployed? It's not because their skills have suddenly and magically degraded. It's because they are harder to bamboozle. An employer wants a young person who is thrilled to get their first job, not an older one with higher productivity who has the experience to know how the employer is squeezing the workforce.

Of course, to libertarians, that's all good. They are on wingnut welfare, after all, and they aren't going to lose their jobs. But analytically, it's crap.

And your thesis that it's all going to happen eventually so why not get used to it now plays in, very unpleasantly, with this kind of pro-elite propagandizing. You can say that it's going to happen eventually, sure. And we'll have to get used to it then. Why should we start now, given that nothing is actually being done to help society change other than to squeeze workers more?

Steve Bloom said...

IIRC Krugman makes essentially the same point as Rich.

Michael Tobis said...

I certainly am not proposing squeezing anybody any harder!

Squeezing concessions out of workers is indeed a key to the story.

But I still contend that resource constraints have finally entered the picture and we can't pretend them away.

Cowen's points seem peripheral to where I am going, but I find them reasonable. This doesn't mean that I approve of a system which treats human beings only as "resources", but there is no reason to shut down thinking about the core aspect of humans as employees. That there is more to the relationship than the economics of the employer-employee trade doesn't make the core transaction go away.

The linked article from Kling, which takes us even further afield from the core concerns around here, struck me as more interesting. He points out that *most* people have zero value in *most* jobs, and that as society becomes more complex the matching task becomes more difficult on both sides.

Sorry this is off topic. But the fact that Richard doesn't like somebody that the article I quoted links to in passing seems to me off topic as well.

Steve, URL for Krugman's article please?

Rich Puchalsky said...

"But I still contend that resource constraints have finally entered the picture and we can't pretend them away."

But you have no evidence for this. Or rather, as I commented previously, the oil-price-shock theory that you hold to is a minority view, held by one notable economist. We really are in "why are people worried about CO2? Solar variability has finally entered the picture and we can't pretend it away" territory.

As for the "it will happen eventually" argument -- I remember reading once an interview with a transhumanist of the sort who believes that technology will some day extend lifespan indefinitiely. That is ... problematic, but in any case what I remember is the interviewer, or someone, saying that we needed to learn how to die. The transhumanist said that we have no need to learn how to die: people have done it since the beginning of the species with a 100% success rate.

Funny, and intended to be funny, but it's a good point too. Why should we "learn how to die" -- get ready in advance to accept a new low-employment regime? If it's going to happen, it will happen whether we've prepared for it or not. I could see preparing for it, perhaps, if that consisted of actions that would actually help. But we're not taking actions like that. In the absence of the political will to make changes in society, all that acceptance of a new low-employment regime gets us is acceptance that workers should be squeezed more.

EliRabett said...

While we are at it Karl Smith is also an idiot. Most welding in factories today is done by robotic welders. Hand welding is done on job sites, which with housing down in the ditch, is a lot less busy.

Shining Raven said...

I don't know why I still comment, Rich is saying things much more eloquently that I wish to say.

Nonetheless I want to emphasize one point: Your arguments are really reminiscent of the usual neo-liberal stuff that is indeed used to squeeze workers.

That does not as such make an argument wrong, necessarily, but it means that this argument has already been examined in quite a bit of detail.

On a more abstract level, your argument as well as that of the neo-liberals boils down to the question whether high unemployment in a particular situation is due to a shortfall in aggregate demand (like presumably Rich and I would be arguing in the current situation), or due to some kind of "structural problems". In the case of the neo-liberals, this would be e.g. an ossified labor market with to little flexibility to hire and fire or too much regulation in general. In your argument, this would be the limits of our resources that we are currently feeling, as you claim.

In both cases it would be futile to try and raise aggregate demand by a government program, but there is no evidence that this is true. On the contrary, the analysis shows that the stimulus has been beneficial for the development of the employment numbers - which would be much worse in the absence of the stimulus.

So on the one hand there is a proven theory (in the sense that it has a lot of support from past experience) that the current problems are due to a standard shortfall of aggregate demand, and the analysis supports this (see eg. Krugman here ).

On the other hand there is a theory that claims that things are due to "structural problems", which really does not have much to support it in terms of evidence, and which is awfully convenient to squeeze workers. Wherever these ideas have been implemented in wide-ranging reforms, they have not delivered what was promised.

I know which theory I favor.

Michael Tobis said...

The upshot of my argument is going to be a universal minimum income and universal health care. That is, dignity decoupled from employment.

Unions tying everything to employment made sense when there was enough work to do that workers were valuable.

Shining Raven said...

I can get behind that, but I still think your premise is wrong.

I think Arthur had it right in the Automation thread. People will always find work to do that other people pay them for. I do not see that we are getting into a society where there is an overall shortage of work.

For your argument to work you have to show that really something has changed structurally, and that it is not just another recession of the type that we understand. I do not think that you can do this, but I think this is the point that still needs to be addressed.

Rich Puchalsky said...

People have been advocating a universal minimum income and universal health care for a long time. And in reality, they exist in a lot of countries right now.

So, as Shining Raven said, I don't see what has changed this time. If you were for those things before, you still are. If you weren't, you still aren't. The argument that this time we're running into resource constraints so we have to do something differently now has the problem that it appears to be factually false.

Moreover, I don't see what resource constraints have to do with the question, really. I can easily imagine a Europe-after-oil-runs-out in which people still have universal health care and income, and an America-after-oil-runs-out that is still dominated by elites who squeeze the have-nots for all they're worth. Changing the size of the pie doesn't necessarily change a society's preferred means of dividing the pie.

In general, I think that what you're doing is a variant of what Republicans do when they come up with policy positions. They start with an overriding goal -- lower taxes, let's say. Then everything that occurs becomes a reason that we should pursue that goal. Did the economy go well? Then we can have lower taxes. Did it do poorly? Then we need lower taxes to make it do better. Did a hurricane hit? Lower taxes to help people recover. Did we have an oil price spike? Then we need lower taxes on oil to keep the price stable for consumers. Etc. You have the goal, and any event can be made to be a forerunning event that's going to require that goal.

Over the long period of time in which we've written things, I've written over and over that you seem to want to remove things from the realm of politics and not address it. And I think that this is a case in point. What you're really talking about is a political conflict. And you seem to want to recast it as an inevitable resource shortage, and therefore more or less a physical event. The physical event may occur, but that's not going to really put the constraints on the politics that you seem to think are going to be there.

Michael Tobis said...

I don't disagree with the prospect that we can handle this badly. Nor do I disagree with the suggestion that what we see nowadays is the result of malice, some of it quite clever.

Where we seem to disagree is with my claim that the present recession may never end if the policies don't account for resource constraints. Even if they do, we are talking about a decade of low employment and/or financial instability.

If I read this right a US president cannot fix it, even with a majority. Nor can the whole G8. Nobody can. Humpty may already be dumped. I suggest we have a new normal on the economic front as well as the climate front. If so, we collectively need to get much much smarter and fast.

manuel "moe" g said...

MT's ultimate point is sound, but MT didn't do himself any favors by quoting Libertarian-in-name/Glibertarian-in-practice economists. There is nothing I can add that Rich P. didn't say better.

I think MT is saying:

[1] the historical precedent for a reasonable workable solution is FDR's New Deal + LBJ's Great Society, paid for with inflation.

[2] The voting majority of today's conservatives are more bat-sh*t than conservative, so [1] is politically impossible. (Ol' Nixon was the last Republican who actually cared about welfare of the entirety of citizens, and Reagan introduced the worship of the fraud called supply-side/trickle-down economics and low inflation trumping everything).

[3] You could borrow your way out of the problem before

[4] Two key physical limitations:[4A] peak-sweet-crude and [4B] the fact that our current rate of use of cheap fossil fuel will boil our grandchildren off the face of the earth.

[5] The last crash made creditors skittish, and the obvious future implications of [4A] & [4B] mean creditors will never again take on debt at the geometrical rate of increase required for [3]

[6] Automation was supposed to increase leisure. In fact, it labeled a huge chunk of workers as superfluous. (I risk derision from King of the Road by bringing this up ;-)

This was my best guess at MT's argument. Because of my mental limitations, I *know* I got it wrong. Any time taken to lift me out of my pit of ignorance would be greatly appreciated. {{wink}}

Steve Bloom said...

Here, Michael.

See also this (and search "rentier" there for more).

Michael Tobis said...

Thanks, Steve. Although there is a point in all that, there is an underlying ideological attachment to a "functioning economy" that seems to me particularly obvious and desperate in the Udall piece. That is, if there is not full employment, something is "wrong".

I am trying to make the case that the universe in fact makes no such guarantees, and rather than trying to "undistort" the economy into a simulation of such a guarantee, we should undistort it into a world view where there is drastically less work to do than there are people.

Pangolin said...

There are several issues here.

One is that "work" is commonly defined as "that which you are being paid in cash for." It's a ridiculously narrow definition and excludes all activities of personal or social value for which people do not receive cash income.

A second issue is that the personal or social survival value of a task has little do the level of it's renumeration. Parenting is "worthless" or paid a poverty wage in the case of state-paid foster parents. Many farmers clear under $30k/yearly for a service without which we would all starve. Financiers are paid vast sums to bankrupt and dis-employ thousands of people at a time.


There are clearly no lack of needed tasks to be done but freeing the resources from the clutches of economic elites who control money flows is another issue. There is no lack of "work." There is inequality enforced by those who control resources by controlling money.

Steve Bloom said...

To clarify, Michael, I only linked the Udall piece because it had all the Krugman links re structural unemployment. I agree that we need a new paradigm.