Let's take these both in turn.
First the EPC requirement:
The principle that solar should be part of a package of energy efficiency measures isn't a particularly bad idea. The implementation however is full of flaws.
Having a fixed EPC requirement immediately rules out those houses which can never reach an EPC rating of D even if they implemented all possible efficiency measures - e.g. houses in areas which do not have a gas supply and have to rely on bottle gas, listed buildings which restrict what energy measures can be taken etc. It is also unfair - someone could be eligible for the FIT despite having no real efficiency measures in place; whilst someone else could have spent a small fortune on state of the art insulation etc. and still not be eligible. A requirement relative to the potential EPC of the building would have been far better. Ironically for many buildings which cannot achieve an EPC D rating, solar pv is pretty much the only practical way of improving their energy efficiency.
DECC have applied the EPC rating to anything that qualifies as a building. It appears that if you put solar panels on a barn which will feed into the grid directly (as opposed to feeding into a nearby house) you will still need the barn to have an EPC rating of D!
The EPC requirement adds to the complexity of selling solar pv - do you expect the customer to pay for the EPC survey, or do it for free as part of the survey. The former will put potential customers off; however, the latter will mean doing EPC surveys with no guarantee of a subsequent sale - this wouldn't be a problem if the profit margins for solar pv could accomodate this but the dramatic cut in FIT is also cutting the margin (and we aren't talking about massive banker-leve profits here, just being able to make a living).
Finally, the EPC adds additional time to the process, particularly if remedial work has to take place. This in itself is not a major problem apart from DECC are now proposing that they can reduce the FIT further should installations go above estimated levels with only 8 weeks notice. This means anyone requiring additional work could find themselves committed to this work when a FIT cut is announcement with insufficient notice for them to complete the work.
As regards the 21p:
This is pretty much the minimum level for solar pv to make sense. For some it is already too low for them to continue in the business and draw a decent wage. From the customers point of view the break even time is pretty much at the maximum that could be tolerated (and beyond it for many). It is already at the point where it no longer makes sense to take out a loan to cover some or all of the installation costs, which immediately restricts it to the few who have ready cash rather than the many. 25-30p would have been "bang on" to quote the OP. 21p is just a little too low, and not justified by the actual fall in the wholesale cost of materials (panels, inverters, mountings).
If it were to stay at 21p for 12 months, it might just work. However, DECC are proposing a further reduction to 16p a mere three months later, and a further drop of 5-10% three months after that, and then every 6 months with additional proposals that the FIT should not be index linked etc. If those proposals go through it pretty much makes solar pv a toy for the very rich only.
