2015 PV prices hit new lows in the US March 30, 2016Posted by Maury Markowitz in solar, Uncategorized.
Tags: grid parity, LCoE, solar power
Most of the data on PV pricing, like the Lazard numbers I prefer, tend to focus on larger installations, especially “utility scale”. In the US this isn’t all that useful on its own, because about half of all the panels going in are on residential rooftops. So that’s why this report is so useful.
Basically the price declines in larger installs remains as breathtakingly rapid as it has been for years, including a whopping 17% in 2015 alone. But those residential numbers are looking pretty plateaued to me. So let’s see what this means.
According to that graph, it looks like residential installs have stabilized at around $3.50 a Watt. To put this in perspective, that’s how much the panels alone cost in 2008, and the set I purchased in 2010 were already down to about $2.30 a Watt. Fully installed, my system cost about $7 a Watt, Canadian, so in the next five years the prices fell a little less than 50%. Still impressive.
Ok now what does all of this mean? I’ve done this before but review is always useful. First off, click on this link and place the new window beside this one. The page comes up set up for natural gas generators, so let’s reconfigure it for PV. Start here:
That’s the amount you pay in $/kW, but the number in the graph above is $/W, so simply multiply by 1000. So that $3.50 a Watt is $3500/kW. So in it goes:
So that’s CAPEX, now we address the OPEX, in these lines:
Basically there’s two parts here. The first line, Fixed O&M, is basic maintenance. If you click on the link (which I did, which is why it’s orange), you can see that a reasonable average value for small solar is around $20. This is certainly more than my own experience, but we’ll use their number because it’s based on real data.
The next three are all referring to the cost of running the power source. The Variable is how much more it costs to run it more, basically wear and tear. For instance, if you run a gas generator longer you’ll have to do more maintenance down the road. That effect doesn’t exist in a PV system, it’s zero. The next two are fuel costs, which are also zero.
So when you’re done, it should look like this:
Now we’ve skipped one field, and that’s the hardest to fill out, the capacity factor. This term is basically”how often did it run at full power”. The default value, 43.6%, reflects the fact that most gas plants don’t run all the time, they go up and down with demand (which is one of the great things about gas plants).
In the case of PV, this number is really just a measure of the sunlight you get, and that is different for everyone reading this. If you click on the link you’ll see they put it at around 20 to 30%, but I can say that that is a lot more than what I get on my home. For now let’s use the lowest of their figures, so that entire section should look like this:
And we’re done, it’s time to look at the bottom line – which is the bottom line on the page:
So this says that if you put panels on your roof at $3.50 a Watt, the power you get out of it is the equivalent of buying it at 14.5 cents/kWh. Here in Toronto, my average all-in rate (delivery costs, taxes, etc) is about 19 cents. And the line above it is important to, this is saying that if the cost of electricity is 12 cents today (set one line above that), if you consider inflation that means that over the lifetime of the panels you’ll actually pay an average of 16 cents if you buy it from the grid.
So in this case solar is a little cheaper than the grid. Here in Toronto where the price goes up and up, we’ve already got a reasonable argument to install PV. Which is amazing, if you think about it.
Ok, two tweaks. First up, change this field thus:
There’s a lot of debate about what this number should be. We know that panels installed in 1982 are almost all working today, at only slight degradation. So that suggests the number is something like 40 or 50 years. But they were built differently, we’ve improved things. Current estimates are that today’s panels will last anywhere from 60 to 100 years. But if you do the math, due to inflation, anything you make past 50 years is basically worth zero. For now, try 25, which is the typical warrantee period. And that gives us:
But now try playing with that. At 50 years you’re down to 8.9 cents, but even if the panels do last that long I’m not sure there will ever be a residential system that does, for a variety of reasons not the least of which is a lot of houses being built today are designed to be knocked down before that.
Ok, now the big one, that Capacity Factor. In the case of solar this is basically “out of all the hours in a year, how many of those have bright direct sunlight?”. This is semi-easy to look up. When I googled “Toronto bright direct sunlight” I got this page that says 2066. There are 365 x 24 hours in a year, or 8760, so that’s 2066 / 8760 = 23%. Calgary is 27%(wow!). Google your city and get the right number, here’s a nice list in hours per day, multiply by 365 to get the number we need.
Now that is the amount of sunlight, not the amount of power. They differ basically by the amount of losses in the equipment. In the case of modern PV systems, that’s about 15%, so the real number is 2066 / 8760 x (1 – 15%) = 20%, Calgary is 23%. Now I can tell you this number is high – my system gets something closer to 16% after losses, but I have a long wire run. So be sure to play around with these numbers! At 16% and 25 years lifetime:
The only other one worth considering is the Fixed O&M, which in my experience is closer to $10. That has a minor effect though, about 0.5 cents on the bottom line.
And I think this is a fairly accurate number. We’re at grid parity folks, even in Toronto, even on residential installs.
The long and short of it is that PV is about where it’s going to be for a while. The price of the gear itself is down around $1.50, which means the majority of that $3.50 price, $2 of it, are “soft costs” – paperwork, admin, advertising, etc. Even if the price of the gear drops to $0, the system will still cost you $2.
And what does that mean?
It means if you’ve ever considered installing PV, but were waiting for the prices to fall, it’s time to pull the trigger. From here on out the diminishing declines in install cost will be less than the amount you would have saved during that period by installing the panels. No, not all of the time, you’ll have to play the numbers, but that’s precisely why I wrote this article.