Problem is — EIA misses the majority of solar energy produced. The agency's numbers only capture facilities over 1 MW, and even then, likely miss production as system owners may not know of their obligation to report this information, according to email correspondence with Christopher Namovicz of the EIA. EIA collects data for Electric Power Monthly through Form EIA-923, a mandatory report for all electric power plants and CHP plants that meet the following criteria: 1) have a total generator nameplate capacity (sum for generators at a single site) of 1 megawatt (MW) or greater; and 2) where the generator(s), or the facility in which the generator(s) resides, is connected to the local or regional electric power grid and has the ability to draw power from the grid or deliver power to the grid. 

It's difficult to say what percent of total installed solar capacity is represented by systems greater than 1 MW. Reviewing data available from Solar Energy Industries Association (SEIA), I calculate only 32 percent of the PV capacity installed from 2008 thru 2011 was developed for the utility markets, which seemingly represent the majority of systems larger than 1 MW. 

Table 1 (above) represents an educated guess at how much solar electricity was actually produced in the United States in 2011.  According to SEIA, about 2,100 MW of PV capacity was installed by the end of 2010. An additional 1,855 MW was installed during 2011, but mostly in the last two quarters of the year.  Accordingly, the adjustment factor column in Table 1 represents the remaining portion of the year for potential production. As a measure of conservatism, installations were assumed to be made on the last day of each quarter.  The analysis also assumes a 17 percent capacity factor for PV and 20 percent for CSP, relatively conservative values.  Finally, 503 MW of CSP installations were operating in the United States in 2011 (none were added during the year).  Even under the conservative assumptions applied, 2011 solar generation was calculated at 4,958 GWh, about 2.6 times the EIA value.

Of course, the actual production of distributed solar systems is generally not publicly available. Actual energy production and system capacity factor are tied not only to system location, but also how the systems are wired, oriented, and maintained. NREL's PVDAQ system tracks a small subset of systems, but much more data is needed to fully understand production trends over time.

In fact, accurate production data will be an increasingly critical component, not just to better understand actual solar production, but to mitigate investment risk. A favorite topic of mine is securitization, the process of pooling projects and enabling investment through a liquid, tradable security. The process is similar to how a mutual fund pools lots of stocks into a single tradable product enabling investment by individuals.  Solar securitization could allow pension funds and other money managers to invest in the industry with the benefit of a diversified portfolio and a tradable, market-priced product — important criteria to investors who don't have the resources or risk capacity to evaluate and invest in projects individually.  But pension funds and other investors need data, and lots of it, to understand risk in aggregate — of system production, customer default, and other facets — if they are to invest in a security of this type.

Accordingly, NREL is endeavoring to improve the datasets available to the investment community.  If you are interested in the effort, in either the design of the datasets or in providing your data (in an aggregated or otherwise fully confidential manner), please let us know by providing a comment or contacting us directly.

This article was originally published on NREL Renewable Energy Finance and was republished with permission.