Time Is Energy

Is solar PV too expensive?

A lot of people believe that solar will always be a niche power source (in our lifetime). The most common reason is that it simply costs too much.

While the following analysis is high level (i.e. it glosses over many details) and assumes many wonderful things which may not come to pass, it does show that solar could provide all the electricity generated in the US in 2005 by the year 2025. While not cheap, it would be affordable—at least in comparison to the ~$270 billion/yr spent on electricity in 2005 and a $25 trillion/yr and growing economy. We don’t actually HAVE to replace every watt with solar and I don’t expect we will…besides our energy needs will undoubtedly be greater in 2025 than they are today.

"Net generation of electricity increased 2.1 percent from 2004 to 2005, reaching 4,055 billion kilowatthours." (source US EIA)

4,000 billion kwh = 4 trillion kwh

Assume that every watt of solar capacity generates 1,500 watt-hours a year (i.e. 4.1 hours x 365 days/yr). While some spots in Arizona get about twice this amount, I live in Chicago.

The US would require 2.67 trillion watts of solar capacity to completely replace all generating capacity with solar panels.

At prices ranging from $3.5-$4 per watt, this would cost about $10 trillion. (About 5 months of US GDP to completely replace all US generating assets with solar power.)

But in ten years I expect solar to fall in price by about 50% due to economies of scale, greater capacity and the historical learning curve.

At $2 per watt to completely replace all US generating assets with solar will run $5 trillion dollars.

Obviously it will take time to ramp up solar PV capacity. I think we will reach 25 GW/yr within the next 10 years (from a global total of 2GW/yr now). At that point I expect solar will really grow fast (at half the current price PV is equivalent to grid power in cost without any subsidy), say 100% a year versus 50% a year growth today. If PV capacity could grow 100% a year for 10 years global annual capacity would reach 25 TW—so 2.7 TW in the US would certainly be feasible. Of course all this growth would drive costs significantly below $2 watt. [Any price below $2 watt means switching to solar saves the buyer money w/o subsidy.] On average I’d guess the cost would be about $1.5 watt, so ~$4 trillion to completely switch the US to solar PV.

I expect this could occur in the next 20 years (assumes US is the main solar market in the world for the next 15 years). Using $4 for the first 10 years, $2 for the next 5 and $1 for the last 4 gives:

$4 x 10 GW/yr average x 10 yr = $400 billion total ($40 billion/yr average)
$2 x 155 GW/yr avg. x 5 yr = $1,550 billion total ($310 billion/yr average)
$1 x 600 GW/yr avg. x 4 yr = $2.4 trillion total ($600 billion/yr average)

posted by Daniel @ 1:59 AM