Tags

, , , , , , , , , , , , , , , , , , , , , , ,

“Modern” renewable energy sources made large gains in providing for global energy consumption over the ten years from 2009-19, according to a recent report, but that “headline” is highly misleading. So is a separate report on the costs of solar and wind power, which claims those sources are now cheaper than any fossil fuel. The underlying facts will receive little critical examination by a hopelessly naive press, nor among analysts with more technical wherewithal. Of course, “green” activists will go on using misinformation like this to have their way with policy makers.

Extinguishing Dung Fires

The “Renewables Global Status Report” was published in mid-June by an organization called REN21: Renewables Now. Francis Menton has a good discussion of the report on his blog, The Manhattan Contrarian. The big finding is a large increase in the global use of “modern” renewable energy sources, from 8.7% of total consumption in 2009 to 11.2% in 2019. The “modern” qualifier is critical: it distinguishes renewables that made gains from those that might be considered antiquated, like dung chips, the burning of which is an energy staple in many underdeveloped parts of the world. In fact, the share of those “non-modern renewables” declined from 11.0% to 8.7%, almost fully accounting for the displacement caused by “modern renewables”. The share of fossil fuels was almost unchanged, down from 80.3% in 2009 to 80.2% in 2019. Whatever the benefits of wind, solar, and other modern green power sources, they did not make much headway in displacing reliable fossil fuel energy.

I certainly can’t argue that replacing dung power with wind, solar, or hydro is a bad thing (but there are more sophisticated ways of converting dung to energy than open flame). However, I contend that replacing open dung fires with fossil-fuel or nuclear capacity would be better than renewables from both a cost and an environmental perspective. Be that as it may, the adoption of “modern renewables” over the ten-year period was not at the expense of fossil fuels, as might be expected if the latter was at a cost disadvantage, and remember that renewables were already given an edge via intense government efforts to subsidize and even require the use of wind and solar power.

The near-term limits on our ability to substitute renewables for fossil fuels should be fairly obvious. For one thing, renewable power is intermittent, so it cannot be relied upon for baseload generation. The chart at the top of this post demonstrates this reality, though the chart is “optimistic” in the sense that planners have to consider worst-case intermittency, not merely average production by time-of-day. Reliable power sources must be maintained in order to prevent the kinds of disasters like we saw in Texas last winter when demand spiked and output from renewables plunged. This is an area of considerable denialism: a search on “intermittent renewables” gets you an unending list of rosy assessments of energy storage technologies, and very little realistic commentary on today’s needs for meeting base-load or weather-induced demands.

While renewables account for about 29% of global electricity generation, there is another limit on adoption: certain jobs just can’t be done with renewables short of major advances in battery technology. As Menton says:

Steel mills and tractor trailer trucks and airplanes powered by solar panels? Not happening. … I think these people really believe that if governments will just do the right thing and require airplanes to run on solar panels, then it will promptly happen.

Cost and Intermittency

Again, we’d expect to see more rapid conversion to renewable energy, at least in compatible applications, as the cost of renewables drops relative to fossil fuels. And major components of their costs have indeed dropped, so much so that the U.S. Energy Information Administration (EIA) now says they are cheaper than fossil fuels in terms of the “levelized cost” of new electric generating capacity. That’s the average cost per megawatt-hour produced over the life of a new installation. The EIA’s calculations are distorted on at least two counts, however, as Willis Eschenbach ably explains here.

The EIA’s cost figures reflect a “capacity factor” that adjusts the megawatts produced to presumed “real world” conditions. It’s more like a utilization adjustment made necessary by a variety of realities (intermittency as well as other technical imperfections) that cause output to run lower than the maximum under ideal conditions. Eschenbach reports that the factors applied by the EIA for solar and wind, at 30% and 41%, respectively, are overstated drastically, which reduces their cost estimates by overstating output. For solar, he cites a more realistic value of 14%, which would more than double the levelized cost of solar. For wind, he quotes a figure of 30%, which would increase the cost of wind power by more than a third. That puts the cost of those renewables well above that of a “combined-cycle gas” plant, which uses exhaust from gas turbines to generate additional power via steam.

The true costs of renewables are likely much higher than nuclear power as well, based on earlier comparisons of nuclear to combined-cycle gas. The EIA does not report a cost for nuclear power, however, because the report is for new capacity, and no additions of nuclear capacity are expected.

The Cost of Back-Up Capacity

Eschenbach notes a second major problem with the EIA cost comparisons. As discussed above, the intermittency of solar and wind power means that their deployment cannot provide for base loads. Other “dispatchable” power technologies, on which production can be ramped up or down at discretion, must be available to meet power needs when renewables are off-line, as is frequently the case. The more we attempt to rely on renewables, the more significant the intermittency problem becomes, as Germany, Texas, and California are discovering.

How to account for the extra cost of dispatchable power required to smooth production or meet peak demand? Renewables are simply incapable of doing so reliably, and back-up capacity ain’t free! Meeting demand at all times requires equivalent dispatchable capacity in the power mix. It requires not just dispatchable baseload capacity, but surge capacity! Meeting long-term growth in demand with renewables implies that new back-up capacity is required as well, and the levelized cost should reflect it. After all, those costs won’t be saved by virtue of adding renewable capacity, unless you plan on blackouts. Thus, the EIA’s levelized cost comparisons of wind, solar and fossil fuel electricity generation are completely phony.

Conclusion

Growth in wind and solar power increased their contribution to global energy needs to more than 11% in 2019, but their gains over the previous ten years came largely at the expense of more “primitive” renewable energy sources, not fossil fuels. And despite impressive declines in the installation costs of wind and solar power, and despite low variable costs, the economics of power generation still favors fossil fuels rather substantially. In popular discussions, this point is often obscured by the heavy subsidies granted to renewables. 

In truth, the “name-plate” capacities of wind and solar installations far exceed typical output, so installation costs are spread over less output than is widely believed. Furthermore, the intermittency of production from these renewable sources means that back-up capacity is still required, almost always from plants fired by fossil fuels. Properly considered, this represents a significant incremental cost of renewable power sources, but it is one that is routinely ignored by environmentalists and even in official reports. It’s also worth noting that “modern” renewables carry significant external costs to the environment both during the useful life of plant and at disposal (and see here). It’s tempting to say all these distortions and omissions are deliberate contributions to the propaganda in favor of government mandates for renewables.

Note: This post also appears on The American Reveille.