When we ask about the uncertainty in the “typical” costs of wind projects in the future, we are not asking for the full range of costs that might be seen for all projects that could be installed in that year. Just as with the 2014 values, many of the differences in costs between individual projects will be driven by project specific factors: understanding those project specific factors is not the goal of this study. Instead, we are trying to estimate general cost trends due to changes in wind technology, wind markets and wind policies. Thus we would like you to think about the range of costs for wind projects in 2030, and how that range might change under different future scenarios that you consider possible or likely. For each potential future scenario, individual project costs may still span a wide range: we are asking you to consider the “typical” costs for that scenario, defined as project costs that are in the middle of that range.
The figure below illustrates this concept. Assume the set of points on the left side of the graph represents the LCOE for projects newly installed in 2014, with the red marker indicating the “typical” (or median) cost of projects. The LCOE of individual projects installed in future years can vary from this distribution in a variety of ways. The four sets of data on the right of the graph, for example, could represent the LCOE for projects installed in 2030 under four different future scenarios. In these examples, the LCOE of individual projects can span a wider or narrower range (while the typical value stays the same), or can shift systematically in a higher or lower direction (with the typical value moving up or down). Project costs could change in other ways as well. As indicated in the figure, however, what we are interested in for this survey is the red marker – the typical (median) LCOE. Because there is uncertainty in this typical LCOE estimate (driven by uncertainty in wind technology advancements, wind markets and wind policies) we are interested in understanding the range of values: the span of the median costs shown in the orange box. We ask for a “low” estimate such that you think there is only 1 chance in 10 that typical costs will be lower (a 10th percentile estimate): in this example, you might provide a low estimate based on Scenario 4. We ask for a “high” estimate such that you think there is only 1 chance in 10 that typical costs will be higher (a 90th percentile estimate): in this example, you might provide a high estimate based on Scenario 3. Finally, we ask for a median estimate, which might be towards the middle of the scenarios in this example. You should, of course, consider the set of scenarios that you think are possible and/or likely, and how the costs of typical wind projects will evolve under each of those scenarios.
In this survey, the “typical LCOE” and “typical costs” designations are used to describe the costs of a (potentially hypothetical) project with an LCOE in the middle of the range of individual project costs. This is equivalent to the median of the distribution: half of all projects will have costs higher than “typical,” and half will have costs that are lower.