Revkin has been looking into a new plan to get to 80% carbon reductions in California. Here's the latest in his series...
There's yet more coming on real-world options and challenges for California and the rest of the planet, but here's a middle course: http://dotearth.blogs.nytimes.com/2011/11/02/more-on-energy-and-climate-paths-for-california-and-beyond/
no plus ones
Shared publicly•View activity
View 4 previous comments
- I should add that of course the #1 thing that affects the date at which SunRun etc. reach parity is the falling price of solar panels. Which as you know have dropped precipitously in cost. The fact that this has happened absent major technological breakthroughs -- that in essence it's happening because deployment is leading to economies of scale and a big push down the learning curve for manufacturers -- is the argument for Deploy Deploy Deploy, no?Nov 2, 2011
- Hi Chris,
That is the argument for Deploy, Deploy, Deploy. And it's wrong! "Learning curves" are simple correlations between cumulative scale and cost. But what does it tell you about the actual causal factors behind them? Very very little! So drawing a straight line into the future based on past learning curves makes it very easy for the SunRun folks to tell you "we'll be at grid parity in just a few years, just you wait." And it makes it really easy for Joe Romm to just say "Deploy Deploy Deploy (maybe do some innovation if you want) and Deploy" without putting any real thought into exactly what is going to get clean energy costs down to "grid parity."
But if you really want to know if and when solar or any other technology are going to get cheaper, you have to look inside the learning curves: what has driven the cost reductions we've seen in solar in recent years (hint: it's not just the build-up of cumulative deployed capacity or "experience"!). And if you want to know where things will go in the future, you need to know what the major cost components of a solar installation are (various materials, labor, finance costs, shipping, etc) and then look at where those cost components could change. Can cell efficiency keep improving to reduce materials needs for the same kWh? Can economies of scale in manufacturing continue to give you real reductions in cost beyond GW-scale plants? Can the price of silicon stay low, or will it rise again as demand picks back up? etc etc...
As I noted at National Journal this week: "Chinese firms have achieved recent cost advantages by simply scaling up yesterday’s solar technology, wringing cost declines out of gigawatt-scale manufacturing supply chains and capitalizing on both a temporary glut in refined silicon and lucrative Chinese state subsidies. None of these factors are truly repeatable, and technology and market analysts project that China’s solar cost declines will soon stall out well above the levels necessary to make solar power truly affordable and subsidy-independent."
I've talked to sharp market analysts who have looked deep into the cost structure of PV and the real causal factors behind recent deciles, and they don't see a straight line path to grid parity, especially not for conventional crystalline PV. PV module prices will probably bottom out around 65-70 cents per kWh sometime in 2012 (Q2 or Q3 if I had to bet). That's remarkably lower than prices just a few years ago, but once you add in the cost of installation, far higher than grid parity without subsidy.
If you want to take Joe Romm or SunRun at their word, by all means do. But I won't be investing any money on their stock tips!Nov 2, 2011
- "If you want to take Joe Romm or SunRun at their word, by all means do. But I won't be investing any money on their stock tips!"
But I don't have to take them at their word, because two of the smartest companies on the planet just put something like $300 million into the "stock tips" of which you're so dismissive:
GE Invests In Rooftop Solar, Competing With Google’s Recent Solar Play
"Earlier this month, Google announced that it is creating a $280 million fund to finance SolarCity's residential solar projects. And this week, GE announced a plan to invest in $63 million in developing commercial partnerships with 10 home energy technology companies as part of the $200 million Ecomagination Challenge. Among the companies is SolarCity competitor SunRun--the only home solar company on the list"
So what you're telling me is, if you were an investor, you'd short Google's $280 million investment in deploying existing solar technology? Google, a company that has repeatedly said that it only puts money in renewable energy projects if they provide a good return on investment, that simply being a do-gooder doesn't cut it for them when they evaluate these kinds of plays? (Google, the company that, after all, invented the term RE<C, which seems to be the core of the Breakthrough philosophy?)
Speaking for myself, as someone whose sole priority is innovation, I don't care how we get there. It just seems bizarre to me that there are two sides to this debate that seem so polarized. I mean look, I cover stuff way up the pipeline every day. In a solar trivia pissing contest I bet I've written about things so early stage that practically no one else has even heard about them; so I totally get that we need to spend like hell on R&D.
What I don't get is why spending on deployment at the same time is such a problem. Is the argument that the two trade off, that they're zero sum? Because that just doesn't seem to reflect the world in which we live, even a budget-constrained one. Is the argument that for some reason money poured into incentives for deployment is bad but money poured into R&D isn't?Nov 2, 2011
- Let me be clear: spending on deployment is not a problem! Spending on deployment poorly, while assuming all it takes to get cost reductions is "scale" "learning" and "experience" is a problem. We need to understand what's going on here to structure deployment incentives correctly to make limited public dollars go as far as they can and to accelerate the causal factors behind learning curves effectively. That's what I'm focused on.
(In fact, we need to spend a lot more on smart deployment than R&D, because deployment-stage tech is more costly than lab bench stuff, as you know)
RE the GE or Google investments, they are smart so long as the policies that make SunRun profitable stay put. And it's not like GE is only banking on learning curves to get solar to cost parity. They're investing in their own high-tech, innovative thin film solar product spun directly out of NREL research! So that's what I'm talking about: look at the real technology and business innovation path to cost parity, dont just draw lines from the past into the future and say, "Walla! Scale goes up, cost goes down." I see too many solar companies making promises like that (or renewables advocates or fans) they can't cash.
GE or Google may actually be able to cash those promises (or NREL or ARPA-E) because they get into what's behind the learning curves and where the technologies can go from there. And I don't think GE invested in SunRun because of the learning curves the SunRun execs showed them...
So just so sum up where I'm coming from:
1) virtually all clean energy tech is too expensive to compete without public policy support; that puts it at perpetual risk of market collapse and setbacks because of policy uncertainty and inconsistency (e.g. austerity and expiring tax credits)
2) getting to cost competitiveness isn't a simple matter of scaling up cumulative deployment. Economies of scale in manufacturing are key, but cost gains become diminishing returns here (e.g. for solar, beyond GW-scale manufacturing, how much more can you gain from scale?); you can't extrapolate forever. Business model and financial innovation is key here too, but it won't get you to cost parity alone. Real, technological innovation is needed. GE, Google, ARPA-E, NREL all know this.
3) some of that innovation can only happen if you have markets for these technologies. But as these markets are artificially created (e.g. by policy), we must take care to make the incentives created by those markets demand and reward continual innovation and cost reductions to accelerate the maturation and improvement of clean techs and ultimately free them from subsidy dependence. That has to happen before you can really "Deploy Deploy Deploy" at a scale that really matters to the climate. Remember that 90+ percent of all energy demand growth in the coming decades will be outside of the OECD nations, and they're not going to be able to afford to pay German or California prices for solar or wind or Tesla Roadsters.
That make sense?
( & will probably be interested in this...)Nov 2, 2011
- Thanks for breaking it down. We're at the precipice of my knowledge so I have nothing else to contribute, except for this hypothesis: we are spending so little on both deployment and R&D relative to any number of other absurdities (new highways, advanced weapons systems permanently in cost overrun, etc.) that any amount more of either we can get is only a net positive at this stage. If you have resources on what constitutes "smart" deployment I'd be interested in reading up.Nov 3, 2011
- If you're worried about the best method to encourage renewable energy .. isn't it a good idea to compare the results of R&D en Deployment in different countries? Has any other country performed as good as Germany on this? Isn't the rest of the world profiting from Germany scaling and driving down prices with its FITs that are being lowered year after year?Nov 3, 2011