Economic News for the New Year

Hi Arnold,
I actually think that the Mainstream Media will be able to have a small rebound in the next year or so. I think that they are going to learn a few lessons from the blogosphere and try to respond.

Granted, this response may fail as well. But at the very least I think we should see some more interesting coverage and reporting from the MSM.

Well, I hope.

I'm interested in the huge range between Kurzweil and the GAO (CBO)'s 1.4% in their longtermsim/modelchartsaugust2005.pdf, which yields a 50-year doubling-time -- even 3% is a 23-year doubling-time. Kurweil tracks curves like Moore's with a doubling time of only a couple of years, and unless these flatten out really soon, which I would think really strange, it's hard to see how they can avoid dominating the picture. I may well be biased, because I think Kurzweil is probably right about AI, so I'm floundering about, looking for confirmation or education: it seems to me that even if he's totally wrong about AI etc, that 3% is an extraordinarily conservative projection. The issue is increasingly automatic production of the means of (increasingly automatic) production. My thoughts, with consequent reduced interest in entitlement reform, are at my MistakesByTJM blog.

Tom:

Because of diminishing returns, productivity growth in a sector will in the long run tend to be limited by the growth in productivity of its least rapidly improving input. So if a sector is improving less rapidly than Moore's Law, you may conclude that raw computing power is not what is limiting that sector.

Take The Long Tail with a grain of salt, please. Chris Anderson, demonstrating the continued demise of Wired magazine, can't even figure out whether IBM has bloggers (http://www.thelongtail.com/the_long_tail/2005/12/announcing_the_.html#comment-12428499)
in spite of stuff like this page (http://smokey.rhs.com/web/ibm/hhbg2ib.nsf/web/index.html) and the fact that "IBM blogs" in Google shows their developerWorks site as the NUMBER ONE ENTRY.

Hell, one IBM blogger, Ed Brill, was featured in the recent Forbes cover story about blogs.

Wired has made itself a case study in the premise of the increasing irrelevance of MSM.

Robert points out that "Because of diminishing returns, productivity growth in a sector will in the long run tend to be limited by the growth in productivity of its least rapidly improving input."

That's certainly true, but I don't believe it affects my weakened-Kurzweil claim because I don't believe the inputs are (in the long run) a fixed set. On the contrary, when I talk about increasingly automatic production of the means of production, I'm assuming that inputs are replaced. For example, we did have productivity growth from old-style industrial motion analysis in which people were trained to handle shovels, looms, etc. better than they did before, and this was a real but limited improvement. It was overtaken in each area of production by more thoroughly automatic production, precisely because that production was improving at a faster rate. Ultimately we should see not only assembly-line workers but farmers and waiters being substantially replaced by "robotics" (I don't mean humanoid forms, of course), in each case with some changes made to the problems so that the problems are easier to solve with the available technology. I don't yet see a limiting factor to this replacement, other than "100% automatic, human labor cost of 0".

Of course, there may be jobs that can't be automated -- minister? And others whose automation may seem offensive to many -- psychotherapist? politician? If overall economic productivity is limited by the productivity of people with those jobs, then my weakened-Kurzweil claim fails. Otherwise, I don't see why it would.

Unfortunately, This has been a prediction that has been fairly consistently held in technology studies about solar power for the last 40 years. The pace of cost/competitive gains in solar power is slow. If you look over the forty year period, solar power has been slowly eating away at market share in places where delivered conventional power is very expensive, but it has not been rapid. I expect that to continue. One day this prediction will be true, however. One can hope. - AJ

3. Solar Power. According to the U.S. Department of Energy,

Within 10 years, photovoltaic power will be competitive in price with traditional sources of electricity.

Solar will probably arrive sooner as alternatives become more expensive faster. SoCal Edison is already contracting for a new solar plant. The only reason it isn't now is the toleration of pollution from conventional sources.

"3. Solar Power. According to the U.S. Department of Energy,

Within 10 years, photovoltaic power will be competitive in price with traditional sources of electricity. "

worldchaanging actually had a rather interesting idea of a bank that helps finance installing solar power and in return you pay your power bill to them. hence you pay the same for power, they collect the savings that go towards the installation costs over a 5 - 10 year peroid. There's other plans that sustianability groups are working on to finance solar. The effectiveness of the panels has actually increased more recently than in previous decades, but what's really making a big difference are solar concetrators i.e. schemes that take typical solar panelling and figure out a wait to increase their exposure to light through angling, mirrors, lenses, etc. anyway, check www.worldchanging.com to keep up with it. It seems like a combination of more solar (hyundai, honda, and sanyo are all producing solar panels now) with concentrators and better photovaltics are converging. We'll probably see widespread solar use in the next 5 than the next 10 I think.

peace,
A