The only thing we can be sure of about the future is that it will be absolutely fantastic. So if what I say now seems to you to be very reasonable, then I have failed completely. Only if what I tell you appears absolutely unbelievable, have we any chance of visualizing the future as it really will happen.

- Arthur C. Clarke (h/t Brin)

Monday, October 1, 2007

Nordhaus Seeks Magic Bullet, Will Pay

Nordhaus and Shellenberger have made a splash with their A Manifesto for a New Environmentalist, and there's been much reaction hither and yon, of which what I have seen has missed the mark.

As usual with economists, I was planing to wheel out my "wrong question, therefore answer irrelevant" spiel, but somehow it didn't quite fit. They aren't growth-obsessed, at least not in the arguments they present.

As I read it, their problem is here, in their core paragraph:
...this money ought to be used to create a new military-industrial-academic complex around clean-energy sciences, similar to the one we created around computer science in the 1950s and '60s. The transformation of Silicon Valley from a sleepy collection of apple orchards and small towns to the information technology powerhouse that it is today was the result of massive investments by the federal government into a set of interlinked military, industrial, and academic institutions in the region--a fact that is largely ignored by many high-tech executives, who prefer to imagine that it all started in Bill Hewlett's garage. Concretely, this means creating undergraduate and graduate programs in new energy sciences; post-graduate fellowships for scientists, engineers, and technicians; and training for the electricians, construction workers, efficiency experts, and installers needed to make the clean-energy revolution real.

This is new-age magical thinking, which is to say, not actual thinking. There is no "clean-energy revolution". No amount of expenditure on homeopathy or cold fusion will pan out.

Clean energy is, presently, expensive and limited. People are motivated to change this, but have so far not succeeded. Accordingly, the prognosis isn't very good. You can no more wish this inconvenient truth away than any other. It is conceivable that there is a fix, but there's no reason to expect that throwing money at the problem will pay off.

I will be cheering as loudly as anyone if something pans out, and I am certainly in favor of as much research in this direction as makes sense scientifically or technically. But there is a limit to what experts can achieve, no matter how much funding they get. Insufficeint funding slows things down, but excessive funding certainly does not speed things up. Ten scientists cannot get a decade's worth of one scientist's work done in a year any more than nine women can make a baby in a month.

If there's anything to this stuff they ought to offer some arguments of substance. "Some energy experts have calculated that an investment of roughly $200 billion would bring the price of solar energy down to that of coal." Names, details please? Solar advocates have long compared apples and oranges. Oh, and until storage problems are solved, coal will continue to be much cheaper at night.

No, the reason we are in trouble is not because we are too mopey. Yes, bad news is always a hard sell, but that's no reason to deliver imaginary good news instead.

Exactly which research programs are underfunded, and by how much? I am sure there are some. That's something we can talk about, and we can also talk about which programs haven't panned out and should be cut.

Throwing too much money in the general direction of a problem never solved anything.

Update: To clarify, I have no objection to further R&D on renewables. In fact, I encourage it. I object to the idea that we can, without diminishing returns, avoiding subsidies or taxes, count on decreasing the unit cost of renewable energy to the point where it is cheaper than digging up some foul gunk and torching it. That's wishful thinking taken to the point of insanity.

The collapse in prices per unit of information technology are a bizarre and possibly unique event in history. Relying on something similar happening in energy technology is where Nordhaus & Shellenberger lose touch with reality. And everything else they say is completely conventional.


In the end, we will be devoting a larger share of GDP to energy, and if we manage to prosper through our coming troubles, we will have plenty of it. This means the energy will be "more expensive". So?

I say let's get it over with. The disruption is tiny compared to the alternative, and investing huge efforts into magic pixie dust is probably not going to help. A gradually increasing tax on carbon emissions to the point where sustainable sources are competitive will do the trick. The magic pixie dust, maybe not.

11 comments:

Jed said...

Greetings. You wrote:

"No amount of expenditure on homeopathy or cold fusion will pan out."

I disagree. Cold fusion was replicated by hundreds of world-class laboratories such as China Lake, Amoco, SRI, Texas A&M, Los Alamos and BARC. Recent progress in joint projects by the Italian National Nuclear laboratories and SRI, and 100% reproducible transmutations reported by Mitsubishi Res. Center, Toyota and the Nat. Synchrotron Lab. lead me to believe that the effect may well pan out. Power density and temperatures are already as high as the core of a fission reactor, so if the effect can be controlled it will produce useful levels of power.

Based on your comment, I suspect you do not know much about cold fusion. Hundreds of positive, peer-reviewed papers on cold fusion were subsequently published in mainstream journals. I suggest you review some of this literature before commenting.

You can find over 500 full text reprints of scientific papers from all of the institutions listed above, and many others, at our web site:

http://lenr-canr.org

- Jed Rothwell
Librarian, LENR-CANR.org

Pico said...

All three of you are wrong. Going by this news item - Solar takes off with US power supply deal - construction is about to commence in the US on the first solar power station providing base-load power, 24 hours a day, at prices competitive with coal. It sounds like they are planning to become very very big in the electricity industry world-wide. Woohoo! At last we are going to see some action!

John Mashey said...

I posted some notes over on DeSMogBlog.

I know some very smart people at Stanford and Berkeley, which both do serious energy research, and certainly understand the role of government funding, and how it was an important ingredient for Silicon Valley.

They certainly understand the differences between:
a) research
b) development
c) deployment

and progressive commitment principle therein, and who needs to fund them, and how university/industry R&D has to work, especially now that the big industrial R&D labs (like Bell Labs or Xerox PARC) aren't what they used to be, or aren't, period.

I also know some very smart venture capitalists, who are certainly funding b) and c), knowing perfectly well that lots of the companies won't make it.

I know some big, serious companies like AMAT, who have been investing big-time for years in building technologies to help solar costs come down.

None of the ones I know are funding homeopathy or cold fusion.

I certainly know ones that are rather scared that Peak Oil is going to cause a lot of burning a lot of unsequestered coal.

Anyway, *none* of the ones I know think that some magic pixie-dust is going to make the *serious* problems go away, and none of them think there is one silver bullet, and lots think that appropriate carbon-taxes would help ...

Michael Tobis said...

John, well said. I agree with all of it and I'm not objecting to funded research.

I'm objecting to the idea that the more money you spend on research, the less deployment will cost, until it's magically cheaper than digging foul crud out of the ground and torching it.

That's the magical thinking part.

Take that out and they are saying nothing original at all. The only original bit is just wrong.

John Mashey said...

Yes, althouhg I might be more specific.

From my 10 years at Bell Labs, key lessons about R&D management were:

1) Progressive commitment, although this sometimes meant:
research
applied research
exploratory/advanced development
development
deployment

With the first sometimes being 20 years out, and with the last one costing $$$$, which meant you better know what you were doing by then, given the scale of the Bell System at that point (and Bell Labs was 25,000 people just itself.)
That meant there were a lot of little research projects, and a lot fewer, but much bigger development projects.

It is a common saying around the Valley that some startup failed because they were doing research, but were really doing development ... which is why VC's try to avoid funding research.

2) Like many things, return-on-investment for each phase was S-curve. I.e., there was some minimum amount of effort necessary to make anything happen, and then if it looked good, there was a good increase on return, and then after a while, return flattened out, and the name of the game was to cap it at the 2nd inflection.

3) And, finally, sometimes you could have great ideas, but implementing them required technologies that weren't there yet, and could NOT be accelerated effectively by any amount of prudent investment.

I can think of at least two big BTL projects that went 8-10 years and wasted a $B each ... ahh, for the days of monopoly.

=====
All in all, I conjecture that we are probably *not* investing enough in energy-related research, i.e., I think we're past the first inflection on the S-curve, but not as far up as we could usefully be, although this is not a strong-confidence assessment. I am involved with two wireless sensor net companies in various stages of deployment [nergy savings of one sort or another], and they couldn't have been done 3 years ago (technology wasn't quite there), so that's about right. But there are other areas that I think could easily have easily been accelerated that did *not* depend on things like specific semiconductor technology levels.

Jed said...

John Mashey wrote:

"None of the ones I know are funding homeopathy or cold fusion."

Then I renew my suggestion that you review the peer-reviewed literature on cold fusion. You will then know that dozens of corporations and national laboratories are funding cold fusion.

You cannot "know" something about a subject if you do not read the scientific literature. You have no basis whatever to judge whether the claims is valid or not. Snide comments equating cold fusion and homeopathy do not constitute a logical argument or an informed evaluation of the science.

Thousands of scientists have published cold fusion replications and theory papers. They include many distinguished scientists, such as three Nobel laureates, the Director of the Max Planck Institute for Physical Chemistry in Berlin; the director of BARC and later chairman of the Indian Atomic Energy Commission; three editors of major plasma fusion and physics journals; a retired member of the French Atomic Energy Commission, and many top researchers from U.S. national laboratories and U.S. Navy laboratories. These people know much more about this subject than you do, and you have no business ridiculing or dismissing them.

- Jed Rothwell
Librarian, LENR-CANR.org

tidal said...

When I see current discussions of "technology breakthroughs" and R&D in energy, more often than not it comes conflated with references to the space program, Moore's Law and Bell Labs, Bill Hewlett's gargage, inflection points, and the like (as it is here).

Obviously, the funding, training, infrastructure, venture capital, etc. parts of the story are apt analogies. But I think the analogies fail fundamentally because energy research is faced much more directly with the stark realities of the first and second laws of thermodynamics. Technology advances are going to be much, much more difficult to come by and, by definition, will face diminishing returns rather quickly (i.e. as you move towards the (unattainable) 100% efficiency, each percentage gain is that much more difficult to achieve). Contrast that to most of the history of Moore's Law.

Wish that wasn't the case... On a somewhat related hunch, I actually think that 50 years hence, thermodynamics and entropy will be part of the daily vernacular and deeply integrated into economic theory.

tidal said...

Apropos to my comment above, this guest post on grist today: http://gristmill.grist.org/story/2007/10/2/145328/975

Excerpts: "There is no Moore's Law effect for solar. Sure, there have been some incremental efficiency gains and prices have dropped a little (though not for the new super-efficient panels), but nothing like Apple's recent drop in price on iPhones or the steady decline in LCD television prices. Consumers of PV panels needn't fear obsolescence any time soon.

The media is doing the consumer and the planet a disservice by over-hyping new solar technologies...

Consumers need to be aware that there are no revolutionizing advancements on the solar front and that PV is a safe and secure investment today. The current solar installation subsidies offered by several major power companies are a huge incentive to act now. Waiting for technology advancements only increases the chance that the subsidies will be discontinued, forcing consumers to pay more to the utility companies in the long run, versus making an investment in generating their own clean power today. Spread the good word -- no news is actually good news!"

S&N might aver, for instance, that there will be "breakthrough" technologies elsewhere than solar, but I don't expect it.

John Mashey said...

Tidal:
I think you are conflating what I wrote with other things you may have read.

You can find lots of dumb things in the popular press, and of course startups pitching ideas often claim a lot :-)

Certainly, I've had to explain to a few people why PV people (and battery people) get excited about a % or two efficiency increases, and why Moore's Law scaling just doesn't apply to that, and that very different kinds of materials science have to be applied.
Sunpower (SPWR) has done real well by getting a few % [disclosure, I own a modest amount of SPWR, should have bought more].

Of course, the people who actually *do* this work understand it perfectly well, and have some passing acquaintance with laws of thermodynamics. Likewise, so do many good VCs [at least some of whom are pretty good engineers or scientists.]

Strangely enough, we'd heard of thermodynamics at Bell Labs also :-), although the Bell Labs references were more to the idea of a place with some effort dedicated to attacking long-term, difficult problems with no magic fixes.

Where Moore's law scaling does apply is in monitoring&control applications, i.e., like the wireless sensor net apps that I mentioned, which can save substantial power use, but in order to be wireless required a certain level of CMOS to get the power down far enough to make a battery last long enough. [Look at Dust Networks, for example, or UC Berkeley's wireless research center.]

Likewise, nobody serious (like say, Charlie Gay over AMAT) thinks solar cell manufacturing follows Moore's Law, but I've seen their projected cost curves, and while they are not as fast as CMOS improvements have been, they are not bad.

This discussion was about R&D management, and why one does research before one does massive development, and I explicitly said:
"Anyway, *none* of the ones I know think that some magic pixie-dust is going to make the *serious* problems go away."

As for economic theory, we certainly agree! Have you read Ayres & Warr 2003? I think the usual neoclassical economists ahve been ignoring energy.

tidal said...

@ John Mashey. I have read Ayres and Warr, thanks. And no, my comments were not directed at your post, albeit you did mention Bell Labs!

@ all: I think that Joseph Romm's post today does a great job of expanding on Michael's original points: Debunking Shellenberger & Nordhaus — Part II, Breaking the technology breakthrough myth

On the one had, it is unfortunate that Schellenberger and Nordhaus are taking this much flack - "we" are all basically on the same side, after all. But we can't afford to commit to policy that is, in Michael's words, "magic pixie dust." This excerpt is how Joe puts it: "If the fate of the planet rests on non-existent technology, we are in big, big trouble — because the thing about nonexistent technology, like fusion, is that it tends to stay non-existent, or like hydrogen cars, just has too many technical and infrastructure barriers to overcome — even after you spend hundreds of millions of dollars pursuing the technology. Fortunately, the technology to combat warming does exist, as I argue in my book at length (see also the “stabilization wedges” work from Princeton)."

John Mashey said...

Peace.
Needless to say, anyone who has ever done serious R&D management knows you don't
schedule" real breakthrough innovations, which also why VCs don't normally invest in pure research [5-7-year ROI expected].

I've seen lots of cost curve projections on different sorts of solar, and I didn't see huge discontinuities anywhere, just the usual grind of incremental improvements and manufacturing volume cost reductions.