Monetary Realism

Understanding The Modern Monetary System…

Solar is about to Change our World

I’ve been getting into Solar lately – the fall in prices has been absolutely shocking over the last 2-4 years. We are seeing price drops closer to 20% per year after several decades at 6% price drops per year. 6% year is a fantastic rate of decreases, but 20% is simply astonishing.
20% is an impressive number, but putting it into context will make your jaw drop with astonishment. My calculations show that if solar maintains 5 more years at current 23% rates per year price drops, solar power will be cheaper than using existing coal plants. That’s right – it will be cheaper to build new solar plants than to use existing coal plants.
It sounds absolutely crazy. But it seems true looking at the data.
First, look at this paper showing the levelized cost levels for using Coal and Natural gas. In this chart, we can see the Levelized cost of electricity for different types of energy. Photovoltaic Solar has a high initial cost, but after paying for the system, the cost goes to the lowest among any source of energy. Great news, but the initial cost for solar is quite high. It’s high enough to make almost anyone not want to use solar.
You can see the price drops for every single source of energy drops after 20 years, because the capital cost of building the plant is paid. After 20 years, the only costs are maintenance and fuel costs. It’s important to note after 20 years, coal drops to about $.05 per watt. That’s the cost of the fuel and maintenance for the plant.
But wait!  These prices for PV solar are from 2010. We are in early 2013. You’d think three years would not make much difference, but you would be wrong. Prices have been dropping at over 20% per year since 2010.
As far as I can tell, installed PV solar prices have been falling at 22% per year over the last several years. Those are huge, huge drops in price over a few years. If you think 22% is too high, take a look at this chart from Forbes.
I did the calculations using estimates of the price drops shown in this chart for utility level solar. Q1 2011 is $3.80, and Q3 is $2.40.
You can put this into Wolfram Alpha and it will give you the answer right away: 23%
The recent First Solar announcement expects prices of $.65 per watt in 2013 and $.48 per watt in 2015. Put this into Wolfram Alpha and we get a price drop of 14% per year.
Bloomberg shows a huge price drop in 2012 of nearly 50% in 2011, and 20%+ in 2012.
Let’s use 18% as the yearly price drop as an estimate for the yearly price drop since 2010, and then plug this into the levelized cost of energy used in the Zwiebel paper. The current prices for PV solar are probably close to $.09. This is still higher than coal for existing plants, but much lower than in 2010.
But the more remarkable data point is that we can expect solar to be cheaper than existing (not new) coal is in just a few years. We can expect some solar to be cheaper than existing coal in 2016. That’s when the levelized cost of newly installed PV solar should be cheaper than using an existing coal plant. That’s not far away at all.
And then just a few years after that, PV solar could become much, much cheaper than coal. Imagine 10 years at 18% drops in price. Where would the price of PV Solar be then?  It will be about 50% of the price of coal.
I hate to speculate, but imagine PV solar drops at 18% per year for 20 years?  Solar will be about 1/10th the cost of coal.  Basically, people will start begging for Solar power in just a few years, because it will be so much cheaper than coal power.
Even if we use the First Solar numbers – which is almost certainly a low ball estimate  – we are still talking 15% per year price drops in installed PV Solar. Even using 12% a year, Solar becomes cheaper than coal in under 10 years.
18% is a pretty big number, but we have seen much, much larger decreases in price over the last 2 years. Over the last 5 years, we’ve seen far larger drops than 18% per year. The pipe line for putting new PV solar discoveries into production makes it pretty clear we can see 18% for the next 5 years at least – and if that is the case, PV solar will be cheaper than coal.
But that’s not even everything. Something to notice in the Forbes article is the lowest number for installed utility solar. The lowest number for installed PV Solar was under $2.00 per watt in Q3 2012. This lowest number is going to drop at a healthy clip too – and that is the number some people are going to be looking at when they consider their own projects. In some places, PV Solar will be cheaper than coal in under 3 years.
Then, there is this Edison Electrical Institute report to consider. (Update Mike: Something is strange with this link here it is –
One prominent example is in the area of distributed solar PV, where the threats to the centralized utility business model have accelerated due to:
  • The decline in the price of PV panels from $3.80/watt in 2008 to $0.86/watt in mid-20121. While some will question the sustainability of cost-curve trends experienced, it is expected that PV panel costs will not increase (or not increase meaningfully) even as the current supply glut is resolved. As a result, the all-in cost of PV solar installation approximates $5/watt, with expectations of the cost declining further as scale is realized;
  • An increase in utility rates such that the competitive price opportunity for PV solar is now “in the market” for approximately 16 percent of the U.S. retail electricity market where rates are at or above $0.15/kWh2. In addition, projections by PV industry participants suggest that the “in the money” market size will double the share of contestable revenue by 2017 (to 33 percent, or $170 billion of annual utility revenue);
In the first bullet point, they note that Solar has gone down in price by about 22.5% per year for the last 4 years up to mid 2012. Using 22% going forward, we can expect to see Solar competitive with coal across much of the country in just a few years.
In the second bullet point, they note that PV solar is cost competitive for 16% of the market today. And it will be cost competitive with 33% of the market by 2017!  
This is absolutely huge. In just 4 years, Solar will be competitive with current power for 100 for a huge part of the market.
But this is not all from this report. The real banger comes in that shifting even a small amount of generated power over to PV solar will cause coal prices to go up for consumers still using coal.

“Assuming a decline in load, and possibly customers served, of 10 percent due to DER with full subsidization of DER (distributed energy resources)participants, the average impact on base electricity prices for non-DER participants will be a 20 percent or more increase in rates, and the ongoing rate of growth in electricity prices will double for non- DER participants (before accounting for the impact of the increased cost of serving distributed resources). The fundamental drivers previously highlighted could suggest even further erosion of utility market share if public policy is not addressed to normalize this competitive threat.”

This is probably too high of an estimate, but they say coal/natural gas power prices could go up by 20% if Solar takes over 10% of the market, and the rate of future price increases will double. Ouch!
I did a quick chart in R to figure out how quickly Solar will be cheaper than even pretty cheap coal electricity, with a variety of cost per year reductions. It turns out solar will be cheaper than even cheap coal in 20 years as long as the price reductions are greater than 4% per year.  Solar will be cheaper than cheap coal in 10 years as long as price falls at more than 15% per year.
Remember, we’ve seen price reductions of more than 22% per year for the last 5 years. Solar will be cheaper than cheap coal if this keeps up for another 5 years. These price reductions add up very quickly. If we can keep up the 20% per year for 15 years, we are talking about essentially free power, because the cost of solar will be 3% of its current levels, and 1/10th the cost of current coal.
It’s remarkable. And it has implications for our monetary system.


Expert in business development, product development, and direct marketing. Developed strategic sales plans, product innovations, and business plans for multiple companies. Conceived the patent pending Spot Equivalent Futures (SEF) mechanism, which allows true replication of spot and swap like products in the futures space.

View all posts by

40 Responses

  1. Strategic1 says

    This is sounding like alot of politics without economics.

    If solar companies are going out of business because prices decline below marginal cost, who is going to build the cheap solar systems you envision? You don’t think market prices are impacted by the level of competition? You seem to think that PV prices are set by some “law”, not by market forces.

    It is important to understand that sustained declines in price result from investments. If there is no probability of earning a satisfactory return, the investments will not be made and the price declines you anticipate will not happen. Right now, prices are falling fast because of oversupply. Even the article you cite admits that the biggest single fall in prices is due to this factor, and anticipates price INCREASES in solar installation, as overcapacity wanes.

    Note that in the paper you reference, most of the demand drivers are subsidies – they are not natural market forces. Absent the subsidies, the case for rapid installation of distributed systems disappears.

    The ability of distributed solar power to sell back to the grid at prices HIGHER THAN THE COST of normal electricity, and a host of other tax schemes. When these schemes were pulled in Germany (a very “green energy” oriented country) demand for INSTALLATION of distributed solar systems evaporated overnight. This is even as prices were jacked to pay for the closure of Germany’s nuclear reactors. If instead of selling back at high prices, DER solar was forced to sell at something like average cost to the utility of power generation, the cost per watt has to drop to levels that are a long way off, if achievable at all.

    Moreover, the paper ignores the many potential sources of falling conventional power sources: e.g. nat gas. Solar may be chasing a falling “fossil” footprint.

    You have to estimate the probability that such subsidies will be removed, particularly in light of new evidence that global warming is not happening and that climate change models are so erroneous that there may be no threat.

    Instead of exploring the potential limits to your theory, you just project linear falling costs forever. , you are sounding like a shill.

    • Michael Sankowski says

      A combination of Swanson’s Law, China, and subsidies will keep PV prices falling. They may not fall 20% per year, but they will fall rapidly.

      You fail to take into account many factors. Factors like Chinese pollution. The reason China is big into solar is because their people cannot breathe in the cities.

      China uses an entirely different way of looking at the world than we are used to here in the Unites States. Their government takes an active role in shaping what industries get funds.

      So we get things like China opening up new major plants for polysilicon production:

      The authors of this article leave out some very important points – namely, China did this with steel too.

      It’s easy to criticize China’s decision. But perhaps making money in every steel mill is not as important to China as having very cheap steel for the country. These are not capitalistic decisions they are making – they are society based decisions. They probably end in economic misery for some people, but if they can jumpstart solar for the world – who cares?

      The united states did the same thing for their steel, oil, and car industries. We did the same for computers. We threw many billions of USD at making computers for a few decades, then all of a sudden we have something like the internet explosion.

      I’d imagine the people who are thinking about this from a long term perspective think the waste today = savings tomorrow. As it stands, solar is a marginal energy source. But they see the 15% declines and do the math too.

      Will this be a straight line decline? No. I’ve traded for many, many years and it makes me laugh to be the subject of an extrapolation of a linear trend critique.

      First, it’s not linear. It’s an exponential trend. That’s what % increases and decreases are – exponential.

      Second, I wrote this post for the readers of the blog here – people who understand covering every aspect of a phenomenon is impossible in 2,000 words.

      Third, these people know I know about the nature of price series. I designed futures contracts and have traded for decades now. Price series data- even aggregated on monthly or quarterly periods like solar PV or polysilicon – shows a tremendous amount of heteroskedasticity and persistence. These are technical terms, not just words. Look it up if you have never seen the words, but I’ll explain it in plain english – volatility of a price series has a volatility of its own, and prices trend sometimes.

      If you think I dont’ understand this, please see my site I have a very deep understanding of how prices work in the real world.

      The drivers of Solar are like everything I’ve seen in the commodity world. Processes get cheaper over time, sometimes by significant amounts every year, as these processes are adopted.

      Here is a chart of Solar PV:

      This gives a 7% decline per year. But I want you to notice something extremely significant about this chart. Look at the prices beginning in 1996. Since 1996, the prices on this chart have been well below a 7% decline per year except for the years of a polysilicon shortage! Once that shortage was solved, PV prices started on a much higher % per year price declines. We can estimate those yearly price declines which have extended over 15 years now with wolfram alpha, and projected prices for 2016 of .5 per watt.

      12% per year price declines for the last 15 years and next 3. That’s much higher than was expected in the year 1995.

      I did give a huge graph of how long it would take at different levels of price declines to reach coal and put it into the article. I wanted to show almost any amount of price declines per year end up with solar beating coal in a rather short amount of time. To me 15 years is a short amount of time. My son will be 25 in 15 years.

      The chart shows that at 12% per year – something we have hit for the last 15 years – solar will be competitive with coal in about 12 years. 12% is the light blue line in the last chart in this post.

      Swanson’s law states that solar prices decline 20% for every doubling of capacity.

      Last year, we saw about a 90% increase in installed solar in the United States.

      By swanson’s law, we should expect about a 20% decrease in prices, and that is what we are seeing. Germany might be tailing off their solar installation, but the United States is ramping up – and the United States:

      has 3.5X as large population
      has huge populations in very sunny areas like California, Texas, and Florida.

      You need to remember something very important – Solar does not need to compete with coal everywhere. It just needs to compete in California for Solar to explode in the United States.

      If only California:

      was Extremely Sunny
      had a large labor force used to working construction
      had a concentration of solar technology companies
      had a huge amount of VC money to invest
      had extremely high traditional electricity prices
      a predisposition to embracing new technolgies

      – only then could we expect the Solar energy prices to fall over the next few years. But this is exactly what California has, so I fully expect the price trends to continue.

      Will it happen every quarter, every year? NO. That is not how prices work in the real world. Do I feel extremely confident 10 years from today installed PV prices will be 15-25% of what they are today? YES.

      • Michael Sankowski says

        And here is a very smart guy who says Solar already beats new coal plants, right now today. Solar plants can be deployed in 3-5 months. Coal plants take 5-8 years to complete. Essentially a coal plant started today is competing with solar in 6-7 years, and by then solar prices will beat coal.

        Solar is already cheaper than a new nuclear plant.

        The unsubsidized costs of residential solar are going to compete with residential solar prices in just a few years.

        • beowulf says

          That’s very interesting. I guess of the four Ps of marketing… its the last two that will determine if you’re to be the Amazon of solar or the :o)
          Place (distribution)

      • Michael Sankowski says

        Here is scientific america showing basically the same chart as I posted, but with less possible price paths.

        Note the average price paid for electricity is $.12 per kwh- this is what he uses for the cost of traditional power in 2011.

        If that is the case, we are far closer to parity than I imagined.

  2. beowulf says

    “we need to be able to put solar on land that is not otherwise useful (think landfills)”

    Uncle Sam owns an unbelievable amount of land out West. Unless there was a nearby water source for irrigation, the Feds literally couldn’t give it away to homesteaders.

    Sooner or later the US and Canadian governments will cash in with a joint venture (a la the St. Lawrence Seaway) to build Roy Tinney’s Central North American Water Project to redirect water headed to the Arctic Ocean towards the Southwest and then auctioning off the newly fertile govt land to farmers and developers. I guess before we get around to that we could put solar panels on those wastelands in the meantime.

    • Michael Sankowski says

      I would imagine you could sell “solar rights, which gives a power company the right to mine solar on your parking lot or rooftop. Putting up some simple roof type installation over the parking lot at a costco or target seems like a no brainer.

      Walmart is doing solar not just for the energy – they are finding solar installations reduce the amount of cooling they demand. I would imagine at some point pretty soon, we are going to start to see these installations pay off when factoring in the reduction in the demand for Air conditioning.

    • Michael Sankowski says

      It always makes me chuckle when I have to approve your comments. 😉

      • beowulf says

        Sometimes I make The Man a little nervous. :o)
        I forgot I had to log in first to post two links. Danke.

  3. Highgamma says

    The panel keep getting cheaper, but they are also declining as a percentage of the cost of a solar installation. The main cost of solar that is not dropping is installation costs and land costs. People need to make solar cheap to put on your roof (and make it feasible to repair a roof once solar has been installed) or we need to be able to put solar on land that is not otherwise useful (think landfills). The cost of the inverter (which must be replaced every ten years) doesn’t seem to be going down and is a big expense.
    Just because one part of a cost has gone down a lot doesn’t mean that the total cost will continue to decline.

    • Michael Sankowski says

      Installation costs can only get cheaper over time. The support component costs will probably go down 5%-10% per year too, just like we’ve seen in electronics in every other market.

      Distribution is probably incredibly expensive today. I cant’ even imagine how much this costs.

      There are many parts of the value chain in Solar which will benefit from more attention. The cost of the component parts is only one of many which will see 3X reductions in price over the next 10 years.

      For example, the entire homebuilding industry is basically a bunch of smaller players. there are large companies, but all of those workers could be installing solar in 10 years, and there would still be demand for more! Right now there are probably some specialty guys who are nearly electrical engineers doing this work. In 10 years, dudes will be doing this out of pickup trucks in Arkansas. That’s cheap labor.

      The DC/AC conversation gear is probably hugely expensive. Just wait 10 years when they are churning these systems out by the container load.

      I could easily see this Solar coming in at 1/15 of the cost today. You are correct – I haven’t done the detailed research yet. But given what happened with plasma TVs and computers, we can expect this stuff to be almost free in 10 years, and low-skilled labor putting it in.

  4. Mary Leonard says

    Re storage, check out the liquid metal battery, and this:
    It’s happening. In fits and starts, but then, so were PV panels not long ago.

    • Michael Sankowski says

      I am considering starting up a heat storage company. I have the technical expertise to be able to ship heat storage units tomorrow, and we are industry leaders in the construction of some of items which would be in very high demand.

  5. Strategic1 says


    thanks for the article, unfortunately, you haven’t really investigated the drivers of the cost reduction. Instead, you have simply projected a trend into the future, sort of like the people projecting AAPL to rise from $700 to over $1000, or the people who were writing books like DOW 36,000 in 1999.

    The cost is not being driven by efficency gains, but by oversupply. As you have noted, prior to a few years ago, prices fell around 6% per year, as photovoltaic yield was increased.

    What has happened recently is driven by other economic factors, most of which happened outside of the US, so maybe you missed them.

    The German government passed huge subsidies for property owners to install photovoltaics on their buildings / homes / farmhouses. The subsidies were so massive, that they created huge demand and many companies, particularly in China, created capacity to sell tax-subsidized installations to Germans (who have some of the most cloudy weather anywhere).

    The subsidies cost the German government so much money, that they terminated them early, leaving lots of companies without the customers they anticipated. Of course, they had made the investment in capacity, and now needed to entice new buyers who didn’t have German government subsidies into the market.

    Even with massive price reductions, many firms are going out of business. This will reduce capacity and allow for price increases, or much slower price reductions.

    Please understand that at current prices, returns on capital are negative. Some companies may have positive cash flows, if you ignore the depreciation expense on their [partly] idle factories. Even fully functioning factories are not earning enough to repay investors. This suggests that there will be limited new investment in physical plant, which also means that newer solar technologies will be slow to be deployed.

    Rapidly falling prices, then, are unlikey to create massive new demand, as you anticipate. Instead, they are likely to lead to slower investment in new solar technology, and slower rates of improvement in yield, possibly falling to rates of improvement not materially better than the ongoing 1-2% improvements in yield realizable from from carbon fuels.

    In short, solar is not poised to change anything fast. Nat Gas, however, might.

  6. Arfa says

    Thanks Michael. Good article. My question is this: I hear polysilicon price has fallen from $400 to $16 in a few years. Which means that incremental decline in poly prices will not contribute much to further declines in cost of producing solar power. What will fall in price from here on to make solar cheaper?


    • Michael Sankowski says

      Thanks Arfa,

      My take is manufacturing improvements, conversion improvements, supplemental gear, and installation costs will drive price decreases for the next 5 years. We’re at 15% or so conversion, but they are mapping out to 45% to take place over the next several years. Even if prices for materials remain the same, this amount of improvement means PV will drop in price substantially.

      Also, I bet the manufacturing process is going to become massively streamlined.

      Then, I’d imagine today installation is done by specialty contractors. That will change as all of the dudes who built houses in 2006 will be solar installers in 2017.

      Then the prices for the additional parts will fall dramatically. My experience with electronics is that this stuff gets vastly cheaper as the supply chain gets used to the product lines. Plasma TVs were like $10,000 in 2001. Not so much anymore though. I’d imagine we’re going to see the prices for the additional parts required for solar installations fall through the floor over the next 3-5 years.

      So I do think we could easily see installed costs fall at 20% per year for the next 5-7 years.

  7. John says

    Solar panel production price/performance show similarities to Moore’s Law.

    But the rate limiting step to solar application lies in storage and intermittency. Land sprawl is also an issue.

    European subsidies are slowly drying up as fiscal crisis deepens. Expect the same in the U.S. over time. A major part of the drop in price is a drop in demand.

    From an environmental standpoint solar panels are likely most needed in China.

  8. Sandy the Swede says

    My niece and her husband installed geothermal in their new house (Minnesota). They are paying approx $30/mo. for electric heatpump backup. Can geothermal be rigged to generate electricity efficiently at the utility plant level? One more question – how does solar furnace power generation compare to PV solar in terms of efficiency and economics? It would seem that the initial investment in a heliostat would be considerably less than PV, but that is a pure guess on my part.

    • Michael Sankowski says

      Geothermal is an incredible technology. I have a buddy who installed a system and is extremely happy with his decision.

  9. Walter Adamson says

    I recently posted an even more alarming analysis explaining the unavoidable train-wreck of the power transmission and distribution companies due to solar. “The Ten Ways Solar Energy will Destroy the Electricity Network Operators and The Poor”

    This is being accelerated in Australia because regulators have allowed electrify prices to rise at huge rates, 70% in 5 years, and as you explain solar prices are falling.

    If there is a false argument in my reasoning behind the demise of the distribution and transmission companies please let em know.

    Walter @adamson

    • Michael Sankowski says

      I didn’t get a chance to address this in my article – your article is great. The prices for residential solar are not falling as fast as utility level solar, but this is going to change at some point soon.

      There is a problem with all of these monopolistic utilities and solar – solar causes their revenues to fall dramatically, but these companies have essentially fixed costs.

      This means prices for using these networks will grow rapidly as users leave the system – which then further incentivizes people to shift to solar. It’s a death spiral for these companies.

      • Khannea Suntzu says

        Utilities are dumbasses. In a free market I can easily envision a new player on the block that offers consumers who install solar a sweet delivery deal. Such a new “lean & mean* utility broker will out compete any Dinosaurs still left standing.

  10. Stephen Kriz says

    De-centralized green power is our only hope of avoiding a climate catastrophe. Solar panels, a wind turbine and geothermal heating/cooling on every home and business in the United States. We have the technology to do that today, just not the political will. Of course, the fossil fuel industry and large public utilities will fight this tooth and nail, because it harkens the end of their industries. You can’t meter the sun and they know it!

  11. Eideard says

    The same states capable of exporting solar-generated electricity to the national grid are often capable of adding to that quantity with wind power. New Mexico has measured that capability years ago – only self serving reticence from our power utility has held back expansion.

  12. Khannea Suntzu says

    I for one welcome our new solar power overlords.

  13. jt26 says

    In terms of raw cost, the recent declines have been helped by overcapacity. Are the costs really total costs (inc. land etc.)? Years ago I did a calculation, for my area, on what would happen if I constructed a solar installation over my whole city property; even assuming I could accomplish ideal charge storage, it wouldn’t even cover 50% of usage. Land costs are real. I did the same once for carbon offsets for every single flight in North America … the amount of land using trees is crazy … those carbon offsets would be 100X more expensive assuming farmland prices weren’t affected. Then the environmentalists will be against putting solar farms in Death Valley (because it spoils the views) … remember Kennedy and wind farms near his property? But all in all I think’s it great for offloading 10-20% of usage, but maybe rooftop microfarms would be a better idea (love fresh tomatoes!.

    • Michael Sankowski says

      I remember reading the sun give about 1.3 kilowatt per square meter. 1000^2 meters is then a megawatt.

      It’s more than that but just to keep the numbers simple.

      The average home uses 14,000 kwh per year. Thats 39 kwh per day.

      At 15% sun/electricity, we get about 1.3 *.15 = .195kw per sq meter. We need 39kwh per day.

      Assume 4 hours of good sun per day. 39/4 = 9.75

      9.75/.195 = 50 sq meters per house per day. Thats about 500sq feet.

      This is doable, but its a lot of ground to cover. In my town, there is no way this could happen. We simply do not have the 500 sq ft per house of full sunlight.

      It might be possible in other places, but in dense urban environments, no way. Now, if the conversion rate goes up to 40% – like they expect it to over the next decade – it is entirely possible.

      Still, this does not account for the weeks without sun across the entire upper 50% of the United States during the winter.

      Solar is huge and will be huge, but these are huge problems. Still, I think the economics will be so compelling for solar that the switch is going to happen very, very fast.

      • beowulf says

        As I understand it, the good thing about nuclear is also the bad thing, its always running at roughly the output. So either you have too much power at night or not enough during the day. Solar is very good at providing surge capacity because there’s positive correlation between sunny days and everyone running their AC full blast.

        • Michael Sankowski says

          This is something I didn’t have space/inclination to address in the article, but solar cuts right at the most profitable part of energy supply for utilities. Utilities make a large amount of their profits from supplying peak power. Solar will generate the most energy on very hot sunny days, and cut into demand for power during those times.

          I’d expect this to have a huge impact on utilities even at low levels of Solar power adoption. Just reducing demand by 5% during those times is a big deal for the bottom line of utilities.

        • Vboring says

          This is a misunderstanding of both how utilities work and how electricity demand works.

          Peak power is the most expensive because it is served by generation and transmission facilities with very low utilization rates. Serving it with solar energy would greatly reduce the cost of the electric system. But this does not mean that that it would reduce profits. Depending on the regulatory environment and rate structure of the specific utility we’re talking about, some utilities will save more money by retiring underutilized peaker plants than they will lose income from not serving this load. Profits can rise with reduced peak load.

          More importantly, though, in most places solar doesn’t actually help reduce peak load. Peak load varies by region. In cold areas, peak load is usually in winter. In hot areas, peak load is usually 6-9pm when A/C is still pumping heat out of the house and people are cooking, cleaning, watching TV.

          And that isn’t your most dramatic oversight. Your blind application of annual % price declines is flat out silly. Solar plants aren’t financial products unhinged from reality. They are constructions made out of stuff. Stuff has performance limitations. This particular stuff has to be large, flat, and made out of fairly pure components that have to be protected from elements to function. And it has to be installed by semi-skilled labor. We don’t know what the cost limits are, but we can be pretty sure that they do exist.

        • Michael Sankowski says

          I highly recommend you read the EEI report I linked to in this post. You’ll find it most illuminating on how disruptive solar will be for the power industry.

          “According to ThinkEquity, a boutique investment bank, as the installed cost of PV
          declines from $5/watt to $3.5/watt (a 30-percent decline), the targeted addressable market increases by 500 percent, including 18 states and 20 million homes, and customer demand for PV increases by 14 times. If PV system costs decline even further, the market opportunity grows exponentially. In addition, other DER technologies being developed may also pose additional viable alternatives to the centralized utility model.”

          We are well below $3.50 per watt installed already. I want to highlight this one phrase: “customer demand for PV increases by 14 times.”

          The demand for solar is going to absolutely skyrocket with the price decreases we will see over the next 5 years. Remember Swanson’s law? The price of solar decreases for every doubling of capacity.

          If the demand for solar is increasing by 14 times – or roughly 4 doublings – what will happen to the capacity? What can we expect to happen to the price? We can expect it to drop to 40% of current levels. Yet, this is just the beginning of the demand curve for solar.

          Then as to the extrapolation of further price increases – that’s a decent point, but I am afraid I have to disagree here as well. My guide on this is First Solar anticipating 15% per year price declines for their products.

          My and your information on solar pricing is nowhere near as good as First Solar
          s information. They anticipate 15% per year for the next 3 years. I think they are lowballing. They are undoubtedly lowballing. They have every incentive to lowball the price decreases. Why? Because they are in danger of going out of business if the price goes down more. As people in this thread have pointed out, many German solar companies are having problems staying in business because prices are falling so fast.

          The market does not care about the number of companies supplying PV solar, it only cares about the price. Consumer demand isn’t driven by the number of companies, it’s driven by the price of PV solar.

          And as for the price leveling off, you are assuming that we need labor to install this stuff. At some point solar panels will be so cheap, you’ll just hang them any old way, and who cares about the details! Everything we have seen with electronics shows us the materials are cheap while the brains to figure it out is difficult.

        • beowulf says

          “but solar cuts right at the most profitable part of energy supply for utilities.”

          Somebody’s profits is someone else’s costs. So I guess that’s a good thing.
          Carbon taxes are inevitable so nuclear is the future for baseline power. The trouble for utilities is the cost of each nuclear reactor is enormous and their liability in any accident is limitless (Congress has set a liability cap but as we saw in BP oil spill, the feds can punch through any cap with environmental laws).
          The more likely danger for utilities is this, they can can spend $10 billion or more on a reactor and then one fine day scientists announce a breakthrough in fusion (or something more esoteric) energy production. That fission reactor they expected to amortize over 40 years becomes obsolete overnight.

          There’s really only one entity that can drop billions like its lunch money, self-insure any catastrophe and write off any investment without hesitation when and if something better comes along . Alas, the idiots running said entity think its running out of money.
          The Administration has tried loan guarantees and I think they’ve had all of two takers, the guarantee is for the bond investors, its a still a “bet the farm” investment for the utility itself. As Professor Bainbridge pointed out a few years ago, considering the govt is the largest operator of nuclear reactors in the world (specifically, that corner of the govt called the US Navy), it’d be more sensible for the Navy to build and operate nuclear reactors in the same way the Army Corps of Engineers has built and operates dozens of power generating dams . If fusion reactors are invented next year or next decade, fine, the Navy can start building those too.

  14. JGF says

    The first link (“First, look at this paper showing the levelized cost levels for using Coal and Natural gas”) isn’t quite right… it should be

  15. stone says

    Very interesting and I’m looking forward to reading your links. First thought though was that solar power needs the sun to be shining. So you need the capital outlay both for the solar power and for the back-up source of power for when the sun goes in. Solar thermal electricity plants could potentially use molten salt stored up during the day to provide power overnight but photovoltaics don’t have anything like that do they?
    Also I thought much of the cost drops for photovoltaics were simply down to Chinese government subsidies keeping production going at an ever greater loss. Perhaps that will continue and escalate indefinitely but to some extent it looks like China paying the rest of the world to use solar rather than coal. Perhaps China wants to do that as a way of pegging their currency to USD instead of having currency pegging inadvertently paying for the vast US military. So China prints Chinese currency to exchange for USD to spend on high grade silicon for PV panels that they then sell at a loss to the USA ???

    • Michael Sankowski says

      “First thought though was that solar power needs the sun to be shining. So you need the capital outlay both for the solar power and for the back-up source of power for when the sun goes in. Solar thermal electricity plants could potentially use molten salt stored up during the day to provide power overnight but photovoltaics don’t have anything like that do they?”

      Yes, the energy storage is a huge problem. PV does not lend itself to easy storage. Solar thermal is less efficient by a huge amount in 10 years. I don’t know exactly how this will play out.

      “Also I thought much of the cost drops for photovoltaics were simply down to Chinese government subsidies keeping production going at an ever greater loss.”

      First Solar is doing pretty well. I don’t think the overcapacity is the reason for the price drops. PV solar wont’ be going up in price once the overcapacity goes away.

      Gotta figure that in about 5 years, these solar companies will be printing money, as they will be able to sell for cheaper than coal, but their costs are falling at 20% per year.

      • stone says

        Isn’t it the case that American PV manufacturers such as first solar nevertheless use polysilicon made in China from raw material mined in Norway in mines now owned by the Chinese government and that whole Chinese polysilicon industry has a full-on loss making ethos?
        If China can pay say $100USD for raw silicon in Norway; convert that at a loss to polysilicon in China and sell it for $50USD to the USA – then that is sort of reversing the trade deficit. It is an alternative to China buying US treasuries as a way to peg the Chinese currency. The consequent excess USD in Norway get spent on US stocks for the Norway sovereign wealth funds and that keeps up the value of the USD versus the Chinese currency.
        Basically China is maneuvering the USA into using the USD exorbitant privilege for subsidizing solar power instead of for military spending ???

        • Michael Sankowski says


          It looks like Silicon production is a pretty easy process. The purity level required for Solar is far less than required for semi-conductors. I would imagine the places which produce carbon can replicate the Silicon process pretty easily – they have the furnaces already.

          “Basically China is maneuvering the USA into using the USD exorbitant privilege for subsidizing solar power instead of for military spending ???”

          As you put it, it sure looks that way. It is a good trick. I’ll have to think that through a bit more.

  16. beowulf says

    So who’s going to capture the consumer surplus?
    Is the investment play in solar panel manufactures or those companies promising building owners fixed electric rates if they’d agree to a long term solar contract or, at the end of the day, General Electric?

    • Michael Sankowski says

      I don’t know. Installation involves men moving stuff with their hands. The installation business will be gigantic and will almost certainly turn around our entire economy in about 4 years. There will be a million jobs for solar panel installers in 7-10.