Solar Wind Reach a Big Renewables Turning Point : BNEF - Bloomberg Business
I think I have posted on this before, but it is very important to understand how renewable energy can make fuel based energy more expensive. This is critical to understand how quickly the economics can turn against fossil fuels. So at the risk of repetition, let me set out an example.
The levelized cost of energy is a combination of fixed cost allocated over units of energy produced and variable costs directly associated with each unit of energy produced. The tricky thing is that the allocation of fixed costs depend on the assumed capacity factor prospectively and upon actual utilization retrospectively. So the retrospective is given by
RealizedLCOE = Fixed×CF÷Utilization + Variable
So let's consider a gas peaker which prospectively has a levelized cost of $185/MWh based on a capacity factor of 10%. Suppose also that the fixed cost at this CF is $135/MWh and variable $50/MWh, but the actual utilization is only 5% (and this is true of gas peakers in the US). So we realize the following levelized cost:
RealizedLCOE = 135×10%÷5% + 50 = $320/MWh
So the peak power that was supposed to cost 18.5c/kWh has blown up to 32c/kWh. We see that the realized cost is very sensitive to utilization.
So we all know about SolarCity's 14.5c/kWh PPA for a dispatchable solar project. Such a plant would already compete well with an new gas peaker at 18.5c/kWh or more. But realistically the utilization over the life of the plant is likely to be much less than the pro forma assumption of 10%. In fact with the encroachment of solar and wind over the next few years even a 5% lifetime utilization would be unsupportable. But optimistically at 5% utilization this plant costs 32c/kWh, and more conservatively at 2.5% utilization, the levelized cost reaches 59c/kWh. At this point, I do not see how any competent bank would be willing to lend for a new gas peaking plant. Even without Tesla bringing $250/kWh batteries to market, much more expensive batteries beat a new gas peaker quite handly.
So why is a dispatchable solar plant not subject to the same utilization problem that a gas peaker is? I see at least three factors. First, the solar power without storage can always be sold into the spot market. It can bid all the way down to zero if necessary. Second, battery packs are consumable over the cycle life. So long as cycles are not too infrequent, most of the cost of the battery is in fact a variable cost, not a fixed cost. Third, batteries are redeployable. If a facility was not utilizing cycles at a fast enough rate, surplus battery packs can be redeployed to another facility that will make better use of them. So most of the storage cost is truly a variable cost, while the fixed costs associated with solar are subject to underutilization owning to lower marginal cost competitions. So a dispatchable solar facility avoids nearly all the utilization risk that a gas peaker is faced with.
Once banks and bond investor figure this out, it will be lights out for any new funding for gas peakers. Moreover, any older plants in need of upgrades or major repairs will be met with the same financing problem. Batteries will find their way into the market. Once the transformation gets going, I think it will be quite swift.