As one of the few organisations still involved in solar PV development, we're seeing substantial falls in the price of installed solar PV. And that's not at field scale – we operate in the 10-200kWp range.

Excitingly, we're seeing installed prices down by more than 25% from six months ago, and sometimes even more. We have quotes for new roof-mounted systems at £740 per kW; quotes on the same systems in 2015 were more than £1000 per kWp. Costs drops are not surprising in a rapidly diminishing solar PV market, but prices are falling in surprising ways. System design is advancing, the supply chain is changing and components are becoming cheaper.

A significant cost-down is through vertical integration. Roof-mounted solar PV is the end product of a tortuous chain of suppliers and manufacturers. A UK installer, for example, might buy kit from a wholesaler, who gets his stuff from someone in Rotterdam, who receives his merchandise from China. Here, the supply chain is strung between some of the 5000 factories that manufacture PV bits and pieces. A module manufacturer, for example, buys silicon wafers from a wholesaler in Guangzho, who gets his silicon from someone near Beijing, who finally buys the stuff from someone who's dug it out of the ground.

We're seeing prices fall as companies link bits of this together through vertical supply chain integration. That's particularly true of the chain links outside China. Companies that both manufacture kit parts in China and carry out UK installations are getting cheaper than installers who simply buy from UK wholesalers.

Popularly, the falling costs of solar PV have been attributed to a key component – PV panels – getting cheaper. For the past ten years this has precipitated the 'terror dome' effect whereby installation prices tumbled. New production in China initiated a massive over-supply, and companies involved in module manufacture vertically integrated. The result is installation prices at a fraction of the cost a decade ago.

That period is, however, over. Panel costs have more or less stayed the same for the last twelve months in Europe at least. But there's more to a solar system than just its panels. A report by the Solar Trade Association last year showed that – at field scale at least – solar panels were less than 45% of the total installed cost.

Most of these are the boringly-titled 'balance of system' costs.

GTM Research's latest report estimates that global PV system prices will fall an average of 40% by 2020, driven by balance of systems cost reductions. I would argue that is conservative.

Inverter prices fell too, and by 26% in 2015 alone thanks largely to Chinese manufacturers entering the market. Previously most inverters were made in Germany, Austria or Switzerland. Certain bits of the inverter market are falling in price faster than others. Small inverters (10kW-50kW) are hugely in demand as system designs in big markets (America and China) have switched from central inverters to string inverters (i.e systems that contain many small inverters rather than one big one). Consequently demand for string inverters has gone up, and prices have come down.

A major PV innovation in recent years has been increasing the power at which systems operate at. By increasing a panel's capacity you need less framing, cabling and man-hours for the same output – a 12 x 280W system of panels will cost you the same as a 12 x 250W, for example.

Increasing panel output is a result of changing the design of a PV system’s architecture. A few years ago manufacturers begun offering panels with 1500V (up from the standard 1000V), meaning they can produce a greater output over the same area. There are gains to be had in the inverters, too. Higher voltage inverters reduce system losses and increases efficiency. And moves are underway to increase the voltage of the system architecture even further – to 2000V.

FiTs are a regressive subsidy design to self-immolate as prices fall. The government's FiT cuts were, in part, to force the industry to develop their practices, and – while grossly over-exaggerated – they appear to have done just that.

Whether PV can remain viable going through 2016 is dependent on installers innovating in both their supply chain and systems' architecture. Many are using the current period of building systems that have been pre-accredited/ registered to continue business as before. These are likely to go out of business as larger rivals appear with better developed supply methods.

When pre-registration/accreditation ends in September, we are faced with the difficult task of making new FiTs stand up financially. The price falls in the last six months represent only 60% of the falls required. That's a great start. But funders like us need to keep pushing suppliers to be ever cheaper.