This is a tough concept to grasp, but very important in understanding the economics of renewable energy. Many in the PV industry like to claim that the energy payback of the PV panel is quick, but this is misleading since the value of the energy opportunity costs of the non-energy inputs is ignored. Let me try to explain. I hope you can take the time to understand this concept, I had to hear it a few times before it sank in to my thinking.
The non-energy inputs required to make a solar panel can include; aluminium (frame), glass, copper, labor, and capital. But all of these non-energy inputs could be used to make or save energy in other ways. The aluminum in the frames could be used to make vehicles lighter and save fuel. The labor making the panel could be used to change an air filter in a car or air up its tires, or add insulation to a house. The copper could be used to increase the efficiency of an electric motor or reduce the electrical losses in a transmission line.
When non-energy inputs such as labor, materials, capital, and land are used to make PV panels, they are no longer available to be used to save energy in other ways and their opportunity to make or save energy is lost. This is the energy opportunity cost of non-energy inputs.
The next logical question is "how can all the inputs be used in a more energy optimal way?" Should we use the copper in an electric motor or in wiring? An engineer might be tempted to calculate the marginal efficiency gain in the motor or in the wire and determine which is more optimal, and she might be correct. But maybe the analysis forgot to consider copper's use as a conductor in a heat exchanger to improve the efficiency of a refrigeration system, or even copper's use as a decorative object? How is this trade-off made? The answer is surprisingly simple; who will pay the most for the copper? This determines its optimal use.
So how does this apply to solar panels? If the sum of the costs of the inputs to make a solar panel are less than the value of the power it produces, a market should develop for producing solar panels. But what if the return is negative, which is the case with PV panels without any subsidies? Then the numbers tell us the labor, copper, aluminum, glass, energy, capital, and other resources used to make the panel could have/should have been used to save or make more energy had they been allocated to other activities, like making compact fluorescent light bulbs, or wind machines.
The other result of this logic can be the most disturbing, that is we will deplete our resources of oil and put more CO2 in the air faster by making panels than if we did not. Or even worse, given the negative net present value of PV systems not including the cost of the panels, we would be better off having made the panels to simply take them directly to a land fill rather than using the additional resources to actually connect them to the grid.