Inverter is the biggest single cost item in PV installation

While crystalline PV and semiconductor manufacturing both rely on silicon substrates, there are arguably only extremely gross process similarities in actual manufacturing practice. In both PV and semiconductor manufacturing, silicon ingots are sliced into wafers, and those wafers are processed into devices which are packaged into a product. PV wafers are not being suitable for semiconductor manufacturing.

Semiconductor manufacturing involves a tremendous number of steps (300+) to embed the integrated circuitry of an IC. Moore's law, which predicted the cost reduction, fundamentally takes advantage of the relative inefficiency of the initial capabilities of semiconductor manufacturing. Cost reductions come from putting more circuitry into smaller packages. TI's first IC in 1958 had 2 components on a 1/2" wide bar of Germanium. The industry can now place over a billion components on a single IC a fraction of that size.

As opposed to all that complicated circuitry in an IC, a PV cell can be modelled as a current source with a single parallel diode. Thus, much of the equipment in a semiconductor fab has little or no application or even any relevance in a PV fab. However, this does not mean that PV cell manufacturing is trivial or without cost reduction potential.

For the PV device, we are dealing with watts per square meter. But, I will have to argue as there is always some type of economy of scale/learning curve opportunity in virtually any manufacturing process. The learning curve for PV is arguably tied to how many square meters are produced - and how efficiently those many square meters of cells can be produced - as opposed to how many devices can be crammed into a tiny space.

Though it has not followed Moore's law, the solar industry has experienced one of the most aggressive learning curves (cost improvements) of any industry. In crystalline PV manufacturing, improvements at each process step are typically beneficial and accretive to device performance and/or overall manufacturing and cost reduction process. Thus, incremental change at each step can provide collective benefits in cost and performance.

The ability to broadly leverage simple changes is one critical reason why the crystalline process remains the dominant technology in PV installations.

Here again I must argue with the previous response as there is plenty of development focused on each of the many steps in PV crystalline manufacturing. Simple changes will continue to drive the learning curve.

At this point, though, the biggest single cost item in a PV installation is not the PV module - but to the inverter. Inverters do benefit from reduced components costs, and are also evolving in design that can be more efficient (i.e. produce more power) over the broad range of conditions experienced in a PV installation.