George Monbiot told me last night he's going to publish tomorrow a critique of them (it's online now, here), arguing that when subsidising solar electricity (photovoltaic/PV) panels they are a waste of money and a rip-off to taxpayers.
As I have in the past campaigned for FITs, I was initially taken aback. But then I recalled I'd looked into the cost-effectiveness of PV systems before and agreed.
The truth is they do have very long cash payback times, even with a subsidy, which is an artificial way of reducing payback and doesn't reduce its actual real-world cost.
The question then arises - shouldn't we spend the money on other technologies which can have the same impact on reducing carbon emissions but more cost-effectively? George quotes the same McKinsey report comparison table Ive used before and which is in the Stern Report, to indicate which technologies do offer value for money.
What's The Evidence For Saying They Are A Waste Of Money?
Firstly: Here are the results of a 2006 study from Bartlett School of Graduate Studies, University College London:
This paper compared two solar systems, an actual building integrated, photovoltaic roof (BIPV) and a notional solar thermal system for a residential block in London, UK.
The carbon payback for the solar thermal system is 2 years, the BIPV system has a carbon payback of 6 years.
Simple economic payback times for both systems are more than 50 years. Calculations considering the current UK energy price increase (10%/yr), reduce the economic payback time for the PV roof to under 30 years.
The costs to reduce overall carbon dioxide emissions using a BIPV roof are lb196/tonne CO2, solar thermal individual systems at lb65/tonne CO2 and community solar thermal at lb38/tonne CO2.
Secondly: A BRE / DTI 2006 UK Photovoltaic Domestic Field Trail (PV DFT) made recommendations on installing PV but did not even look at payback.
Thirdly: Here's something I wrote in my forthcoming book on environmental refurbishment that was taken out by an editor:
Photovoltaics would not need the high level of financial support that the technology clearly does if there wasn't a problem with generating enough power in extreme latitudes to justify their installation.
Many people have had their enthusiasm for photovoltaics curbed when they find out exactly how much power they can expect to generate for their cash. Therefore it is important to be quite clear on this: you are unlikely to generate more than a fraction of what you need and if you're looking for quick paybacks, forget it.
The best way to demonstrate this is through a worked example.
Lesson 1: Solar panel manufacturers quote figures for the "peak power" and "installed capacity" of their products. According to industry standards these are the amounts of electrical output in watts that they would generate if one kilowatt per square metre of the sun's energy were to fall on them. But how close is this to the amount of sunshine at your location? These figures can be found out from the same source on insolation given in the section on solar water heating. For most of the latitudes that cover England and Wales, the summer insolation is a fraction of that figure. Even Europe's sunniest place, Limassol in Cyprus, only gets 325 W per square metre. London gets 198 and Edinburgh 172 in July. In December, the figures are 96, 22 and 13 respectively. So in the winter, it's a lot less -- and that's when you need more power because the lights will be on for longer.
On average, Edinburgh receives just 9% of the solar energy required by the panel to generate what it says it will on the box.
Lesson 2: Suppose you installed 30m2 of panels that were quoted by the manufacturer as having a peak power or installed capacity of 5.7kWp. Suppose they were installed in London, which has an average insolation figure over the year of 109W/m2. In that case you wouldn't get 5.7kW averaged over the year, but 0.109 x 5.7kW = 621W. However that is the average figure.
In darkest December they were generating just 125W, or enough to power 10 low energy light bulbs. In fact it might be even less than this, because of shading, downtime and other system inefficiencies.
Lesson 3: How much would this cost and what would your payback be? Here are some figures from 2004 for an installation on the roof of The Insolvency Service, in Bloomsbury London. (There aren't that many case studies where the figures are all available -- this one comes from a UK government report - Large-Scale Building Integrated Photovoltaic Field Trial: URN Number 07/1316, BERR, 2007:
Size Annual Output Project Cost Cost per peak W Cost per kWh Payback time
kWp MWh/year lb lb/Wp p/kWh (years)
25.4 11.8 318,760 12.6 108.2 237
237 years payback? This is a good example of how not to do it - and perhaps ironic given the purpose of the institution (if we did this we would go bankrupt).
The best performance figures in the report come from a water park in the Cotswolds.
These are from 300 building-integrated monocrystalline modules rated at 85W on a 150m2 sloping roof, with a yield of 51kWp. The system cost EUR397,500 in 2003 and the following year generated 44.3MWh. The figures in the report are:
51.0 42.8 2 65,000 5.2 24.8 54
54 years is still a long time to wait to get your money back, especially when the modules only came with a 20 year warranty from BP. And 24.8p per kWh is still twice the current average electricity price.
(These examples illustrate how to work out PV systems' cost effectiveness. To work out how much carbon emissions they save, simply multiply the megawatt-hours by the carbon dioxide emissions figure given for fossil fuel electricity, which presumably the panels will displace, given on page x - 550 kg/MWh.)
Lesson 4: Although the figures from the field trials report a mean/peak power ratio of over 7%, equivalent to an annual yield of over 610kWh/kWp, if we accept a grid-purchased electricity price of lb0.114/kWh (the 2008 UK average), and require a 5 year payback period, the break-even cost of a PV system - the total installation cost per peak watt - is lb0.35/Wp.
But the UK Energy Saving Trust, in its brochure on solar electricity published in 2007, quotes installation prices for domestic rooftop PV of lb5-lb7.50/Wp. This seems a bit steep, but: it is 14-20 times higher than the break-even value.
And you're only generating a fraction of the power you need.
