Both of these sentences appeared in major US newspapers, one during the Reagan administration and one just last year:
“Now, consider solar. Photovoltaic systems get the most attention. But for now, at least, they are very expensive, costing an estimated 39 cents a kilowatt-hour .”
“Costs for photovoltaic power currently range from $6 to $9 a watt, which translates to 15 to 20 cents per kilowatt hour.”
It’s reassuring, right? Solar power is still kind of expensive, but at least we can look back a quarter decade and see how much progress we’ve made, even if that progress has been slow compared to, say, the 10,000 fold improvement in desktop computing power. Except–as you might have guessed!–the Reagan-era quote is the one that has solar power costing 15 cents a kilowatt hour. The 39 cent figure is from 2008.
So what’s going on? Well, we were browsing through some old newspaper stories about solar energy and were struck by the fact that what was being said about prices and projected improvements a quarter century ago sounded–literally–exactly like what you might read in today’s paper. Even down to the price per kilowatt hour. If anything, the average price of solar energy (as cited in the press) seems to gone up a bit in the last two decades.
We’ve collected quite a few examples from each period–the first batch from 1985 to 1990, the second batch from the past year or so. There is a lot contextualizing and caveating that could be done, of course. But we find the simple juxtaposition of the quotations to be much more interesting, and will save our commentary for another day.
Welcome back to our solar future:
Then:
1985: Solar [is] two to three times more expensive than conventional energy sources, [which average] 5 cents per kilowatt hour on Southern California Edison’s system. -LA Times
1987: Costs for photovoltaic power currently range from $6 to $9 per watt, which translates to 15 to 20 cents per kilowatt hour. - St Petersburg Times
1987: “Solar electricity, at about 35 cents per kilowatt hour, is far too expensive for most applications in the industrialized world. The average cost of electricity is 6 cents per kilowatt hour. Solar supporters estimate that they can match that price in five to 10 years.” - Washington Post
1988: Electricity from photovoltaic cells now costs 30 to 35 cents a kilowatt hour, compared with the 6 to 12 cents a kilowatt hour charged by U.S. utilities, Surek said. But in the next five years, new technologies should push photovoltaic power costs down to about 10 to 15 cents, he added. - Chicago Tribune
1989: ”Right now, this technology’s at about 25 cents a kilowatt-hour,” said Robert H. Annan, director of photovoltaic research at the Energy Department. ”I think it will be driven down in the near term to 12 to 15 cents,” - NY Times
1989: Luz company officials say their solar thermal installations are already cost-competitive with the expensive sources of electricity used at peak hours, producing power for about 8 cents a kilowatt-hour. - NY Times
1990: Electricity from solar power still costs more than 25 cents a kilowatt-hour; residential customers in the United States pay an average of about 10 cents. - NY Times
1990: “[Photovoltaic] power now costs 30 cents a kilowatt-hour.” - NY Times
1990: The solar industry promises it can produce electricity for about 15 cents a kilowatt-hour by the mid-1990s. - Washington Post
Now:
2007: “For businesses, the cost of solar energy is about seven times that of the 7 cents a kilowatt hour they pay for power, the [Arizona Dept of Commerce] report says.” -Arizona Republic 2007
2008: Estimates from the California Energy Commission show that […] photovoltaic solar panels can cost 70 cents per kilowatt hour. And [thermal] solar plants such as Ausra’s, which use mirrors to boil liquids and generate steam, cost about 27 cents per kilowatt hour. - San Francisco Chronicle
2008: “Best case” for power from solar panels is about 25 to 30 cents a kilowatt hour. - NY Times
2008: “Southern California Edison estimates that electricity from the Fontana [solar] facility costs about 27 cents a kilowatt-hour, compared with an average of 8 cents a kilowatt hour from conventional generation.” -LA Times
2008: Natural gas, the cleanest and most expensive fossil fuel, costs 9.4 cents a kilowatt-hour, according to the California Energy Commission. […] Now, consider solar. Photovoltaic systems get the most attention. [...] But for now, at least, they are very expensive, costing an estimated 39 cents a kilowatt-hour .” - Miami Herald
2009: “China’s renewable energy frenzy has been accelerating recently, especially in solar energy. Last winter, winning bidders for three projects agreed to sell power to the national power grid for about 59 cents a kilowatt-hour.” - NY Times
2009: This is all possible thanks to state and federal government subsidies, and the declining cost of producing electricity from solar panels, down to $6 per watt. - Newsweek
2009: Under the program, the Ontario Power Authority will pay small solar producers 80.2 cents per kilowatt hour of energy they send to the grid. - CBC
UPDATE: To be clear and to reiterate: There are a lot of complexities and mitigating factors we chose not to address here. The point, to the extent there is one, is that a lot of what’s written in major newspapers about solar in 2009 sounds exactly like what was written 20 years ago, even down to the price of a kWh of solar generated electricity. Is that interesting and/or noteworthy and/or amusing? We thought so. Feel free to disagree.
It shouldn’t be mistaken for anything more than it is though — which is simply an exercise in “Hey look at this!” Some people might be surprised by the relatively slow pace of price improvements over the period in question. Others might see it as evidence of a technically clueless media. And so on.
October 19th, 2009 at 6:15 pm
You need to adjust for inflation.
October 19th, 2009 at 6:21 pm
Of course. Not pretending to make a rigorous price comparison here, so didn’t feel it was necessary to make inflation adjustments. For the record, though: 25 cents in 1990 is equal to 40 cents in 2008 dollars.
Jebediah
October 19th, 2009 at 6:45 pm
What you have here is a collection of old & new hype, out of context numbers and mismatched programs. You’re not just comparing apples & oranges, you’ve got raisins & grapes, plums & prunes & various odd bits of fruity nonsense.
The record according to newspapers & the like is almost totally useless, as most of the stuff comes directly from hyped-up manufacturer’s press releases or from journalists who don’t know a kilowatt from a kilowatt-hour……
If you take away gov’t subsidies, most of the solar industry would collapse (this applies to any country in the world). The only truly cost effective application of solar energy in North America is solar pool heating……
October 19th, 2009 at 10:21 pm
I recall reading in The Economist or hearing in a physics colloquia about the price jump in the 80’s relating to solar cells. The reason was that the original solar cells from the time were made with the excess silicon that say Intel would have thrown away after etching their chips or wafers that weren’t up to spec. The solar companies would buy these wafers at a reduced price. Once solar became a large business the chip makers jacked up the price and solar was forced to make it’s own wafers.
This may not be the whole story, but certainly could have contributed to the jump in price.
October 19th, 2009 at 10:29 pm
It does raise some questions, but they are pretty easily explained.
1) In the 80s, monocrystaline silicon was in lower demand than now. This is due to both lower demand for both microprocessors and for silicon solar cells. Now we have intense demand for both. Since microprocessors are using most of the silicon, and microprocessors create more value from the same raw materials, they can afford to bid up the prices of monocrystaline silicon.
2) With a new technology, strong fixed-price asset, depreciation schedules are unreliable because they are just predictions. Since production solar cells were “new” in the 80s, we didn’t know how fast they would deteriorate. Now we realize they wear much faster. This increases depreciation schedules, and as such raises the costs.
3) Some of the difference is made up in inflation. Probably not very significantly, but at least some of it.
October 19th, 2009 at 11:16 pm
I want my 5 minutes back.
October 19th, 2009 at 11:50 pm
The solar systems costs included solar panels cost, peripheral component cost, and installation cost. These three costs should seperated to discuss. According to PVinsights, http://www.pvinsights.com, solar panel cost should be much lower than that in 1980’s by efficiency improvements. Installation cost should be highly influence by inflation rate and labor cost. Hope any one can explain the peripheral cost. Peripheral components are inverters, charge controllers, back up batteries, and so on.
October 20th, 2009 at 8:52 am
This is why efficient small-wind wins in the end. Our Green Revolutions Windmill (http://greenwindmill.com) needs only 2.5mph to generate 5Kw at a current retail of $6K or about $1.20/watt installed. Depending upon the metrics used above, that is about $0.03/Kwh. We are producing an average 3200Kwh/month in a Class 1 wind zone with our windmill, which is scalable to megawatt capacities. We will install a 30Kw unit at a cost of about $12K to $14K in November in downtown Brattleboro.
Most small-wind devices cannot produce such efficiencies at low speeds yet some of us can. The advantage of ours is the simplicity of design and low cost of goods combined with a high efficiency of air-flow capture.
October 20th, 2009 at 11:58 am
[...] Was Solar Energy Really Cheaper In The 1980s? » INFRASTRUCTURIST [...]
October 20th, 2009 at 12:52 pm
[...] Was Solar Energy Really Cheaper In The 1980s? » INFRASTRUCTURIST [...]
October 20th, 2009 at 2:07 pm
[...] Was Solar Energy Really Cheaper In The 1980s? » INFRASTRUCTURIST [...]
October 20th, 2009 at 7:50 pm
My 3.0kW grid-tie system will end up costing me $8,000 after the $3 per watt APS rebate, the $1000 Arizona tax CREDIT (NOT deduction) and the 30% federal tax credit.
My 3.0 kW system will make me 100% efficient (no need to pay for any power from the local utility company). Now, if you work $8000 into a home mortgage that is just $48 per month, AND you have ADDED value to your home!
Also, this article doesn’t even adjust for inflation. In 1987 a 12-pac of budweiser cost about $5, now it is around $8.49. A move in 1987 was around $4, now it is at least $8.
And on and on…
October 20th, 2009 at 7:51 pm
BTW: My 3 kW system is just $2.66 per installed watt!!! That is cheap folks!
October 20th, 2009 at 7:53 pm
Without any types of government or utility subsidies it is exactly $6.89 per installed watt…and just $4.50 if I install it myself ($2.50 per watt is what you pay the company/dealer to install the system).
October 30th, 2009 at 9:04 pm
Yeah, these numbers simply illustrate that reporters do a really bad job at reporting *anything* related to energy. As far as I can tell, almost none of those articles were even accurate on their *own* terms, let alone comparing them to other articles.
So, as a study in the decline of the news media, possibly interesting. In regard to infrastructure? Not interesting at all.
As an interesting data point, having researched the details of solar, I found that things people don’t always think about make a *huge* difference; a cheap tracking mount increases output by 25%-100%, but almost nobody knows this, and the cheap tracking mounts are only sold in Australia!
Given wild variations like this, I do not believe *any* generic estimate of the “cost” of solar. The numbers are gobbledegook.