Fixing FITs some more

I noted yesterday that the UK’s expensive feed-in tariff (FIT) is not structured to support the development of the highly efficient technologies (turning 30%+ of the energy in incident sunlight to useful electricity) that the country will need in the future.

Here’s a more concrete example of what is happening. If we look at the web-site of one reputable supplier, Spirit Solar, we see that they offer several types of panel. The stand-outs are Sanyo’s “Hybrid Mono” products, with the HIT-N240SE10 achieving 19% efficiency. These are still just silicon based (I understand that best results will eventually be obtained with layered panels, each layer using a different technology to capture light of particular wavelengths), but optimised for efficiency. The Sanyos are just a tad more expensive than the bog-standard panels at 14-15% efficiency. But here’s what Spirit Solar say:

“The Sanyo are the most expensive per Watt of power generated. Their one big advantage is that by using a different ‘hybrid’ technology they give approximately 25% more power per square metre than any other panel. …the Sanyo gives 190 Watts per square metre, compared to between 134 [actually none are this low – the figure is presumably out of date and the text should read “141”] and 149 Watts per square metre for the other panels. The downside is that they currently cost around 50p more per Watt – so for a 2 kWp system you will pay around £1000 more if you choose Sanyo. So if you are wondering whether to buy Sanyo or not, ask yourself what are you trying to maximise – the total power you can squeeze out of your roof or your financial return? If your objective is simply to maximise the power output from your available roof space without regard to budget, then you should choose the Sanyo hybrids. Don’t choose the hybrids if you want to maximise your financial return.” [my stress]

You have to email for a quote, and PV prices are currently all over the place, but that extra £1000 is on a system costing at least £5000, probably somewhat more. That is, for 20% more, tops, you get 25% more power for a given area. No-brainer.

But, as discussed yesterday, the FIT scheme does not provide an incentive to maximise output per unit area.

In fact, the FIT scheme seems to have been devised with no thought as to the behaviour it will encourage. The sole goal seems to have been to “bung up a few solar PV panels”.

Another page on Spirit Solar’s rather informative site illustrates another problem. Scroll through some of the pictures of the company’s installations. There are a number of examples of roofs where far more area could have been covered with panels.

Why wouldn’t customers want to cover their whole roof? Because there’s no incentive to do so, of course. Many installations will be of 4kW capacity (4kWp, where p stands for “peak”) or slightly under. Why? Because larger schemes receive a lower tariff. In fact, as I understand it, and using the tariffs to be put in place from now on, there’s no point at all in putting up a scheme of 4-4.8kWp. That’s right – you get less for a 4.8kWp system than for one of 4kWp. (The generating tariff for a scheme up to 4kWp is 21p/kWh, whereas for 4-10kWp it’s only 16.8p/kWh, so schemes of 4-4.88kWp will earn less revenue than those of 4kWp or just less!).

Of course, this problem could be simply fixed by qualifying the first 4kWp for the higher tariff and only additional capacity for the lower rate. This would at least mean that solar PV generators have an incentive to cover the whole available roof area.

Another solution would be simply to have more bands. The worst thing about the whole scheme is that you need a large unshaded roof area to reach the optimal 4kWp (about 25m^2 for standard panels, 20m^2 for the Sanyos), so there should be bands for smaller roofs (say 1kWp and 2kWp, but this should be done on installation area as discussed yesterday) and to ensure larger roofs were fully utilized (say 6kWp, 8kWp as well as 10kWp, and more subdivisions up to 50kWp).

There is another problem, though, which is that the export tariff is only 3p/kWh. That is, if you use the electricity, you displace electricity (or gas) you would otherwise have had to buy for around 10p/kWh (or 5p/kWh for gas). Since most households would be hard pushed to make use of more than 4kW for any length of time (though I don’t see why solar PV generators don’t invest in a 20kW battery to store energy for the hours of darkness), this means the drop-off in revenue above 4kWp is even worse.

Surely the policy is eventually to generate electricity for sale to the grid. If this is the case, then shouldn’t the export tariff be somewhat higher? I gather the 3p/kWh is based on the wholesale price for electricity. But solar PV electricity is more useful than bog-standard electricity. It’s renewable for a start, so should qualify for ROCs (renewable obligation certificates) which otherwise cost money. And it’s at a peak time (except perhaps at weekends). One problem is that you don’t want the export tariff to exceed the retail gas price, because then solar PV generators have an incentive to use gas rather than electricity (since they can sell the electricity to pay for the gas and still be quids in), but there’s still scope for an increase to, say, 4.5p/kWh. This would make it a little more worthwhile for people to cover their whole roof with panels.

It may be worth DECC thinking through the FIT scheme a little more thoroughly, since these panels are going to last for decades – they’re typically guaranteed to provide at least 80% of their original output in 25 years. Once they’re up, it’s not going to be worth taking panels down – especially those qualifying for FITs – to replace them with more efficient technology.

Why don’t we try to get something right first time for once?

Fixing FITs

Regular readers will be well aware of my long-standing scepticism as to the value of a policy of generous Feed-in Tariffs (FITs) to support small-scale solar PV installations. I see that my blog category on the topic of FITs now has no fewer than 10 posts (11 with this one), dating back to February 2008 – and that first entry included a letter sent to the Guardian in December 2007. So I’ve been documenting this fiasco for 4 years – so far.

DECC have launched yet another review of FITs. It seems like just yesterday I responded to their fast-track review. Don’t try and read this in DECC’s amazing non-scrolling spreadsheet (downloadable from their FIT fast-track consultation page) but I see I ended my answer to their question 2 (which mostly consisted of expressions of astonishment that they had no effective mechanism for controlling the cost of FITs) by saying:

“If the intention is to exclude investors altogether and encourage only those pursuing a low-carbon lifestyle or self-sufficiency, then the domestic PV tariff should be reduced by at least 50% immediately.”

which is only what they’ve gone and done!

Subsequent to sending off my response to the fast-track consultation back in May, I realised that the most efficient way of setting FIT rates would be by auction (it seems some people made this point in their responses back then). The latest consultation is another opportunity for me to make this point.

But there are even bigger (fitter?) fish to fry. Just in time (normally abbreviated to JIT, so maybe I can say “JIT for FIT policy…”!), the Royal Society held a two-day discussion meeting on solar power in mid-November. As you’d expect there was a lot of discussion of new whizzy technologies. But a couple of more prosaic points struck me:

1. Solar power generation in the UK is limited by the available space, David MacKay emphasising this point in particular.
2. New solar PV technologies – that might convert a greater proportion of sunlight into electricity – are finding it difficult to get to market. As costs are lowered for these technologies, “traditional” silicon-wafer based PV continues to build scale economies.

In other words, FIT policies and other government subsidies around the world are driving prices down for the wrong technology for Britain. We don’t want 10% efficient panels, we want to turn 30-40% of the energy in sunlight into electricity.

Now, if you have a roof of a limited size, then, once there are no subsidies and the cost of solar PV is competitive with other forms of energy, you’ll choose the most efficient technology that continues to yield cost savings compared to buying electricity. That is, you’ll consider panels of 10% efficiency, 15%, 20% and so on until the cost per additional kWh over the lifetime of the panel exceeds the value of the electricity generated.

But subsidies screw things up.

As usual.

Specifically, in the UK FIT scheme, you’re limited by the banding to a certain size of installation. For example, the tariff for an installation of up to 4kW capacity is 21p/kWh. For 4-10kW it’s only 16.8p/kWh. So installations are typically of just under 4kW capacity. If your roof can accomodate a cheap 4kW panel there’s no incentive to instal a more efficient panel. Or even to use the whole roof. The UK’s solar resource is being squandered.

There are many ways to address the problem. You could devise all kinds of rules and regulations to ensure people used more efficient panels to drive down the costs of the best technologies.

One simple approach, though, might be to base the banding on installation area, not capacity. That is, the highest tariffs would be for installations that used just a few square metres. This would provide an incentive to squeeze as much energy out of the space as possible. It would also be more equitable as it would favour the owners of large roofs much less than does the present system. Most important, though, such a scheme would provide a market to help bring down the cost of the technologies we need for the future.

The present system ensures that all the available FIT income will go to owners of large roofs. Because more efficient panels cost more per Watt, the Government will keep lowering the tariff in response to a surge of installations of less efficient panels as their price comes down because of scale economies.  The FIT will never be high enough to justify installations of more expensive, more efficient panels on smaller roofs.

Tariffs for installations starting at (say) just 5m^2 roof space (compared to around 25m^2 for typical panels in a 4kW installation today) would ensure there is a market for the more efficient panels.

The approach of FIT banding by installation area rather than capacity could be simply combined with auctions – that is, prospective PV generators would bid the lowest FIT they were prepared to accept for one of a quota of solar PV installations of a given area – up to 5m^2, up to 10m^2 and so on.