Expensive Nuclear's Dirty Secret

In this piece, we delve into the GKG Grid Model's results for Germany in a bit more detail. One revelation is that the bastards pushing expensive nuclear are smarter than you might think.

The PDF of this article is here. In that version, the scaling of Figures 2 and 3 better matches the text.

Figure 1. German Power Cost vs CO2 emissions for a range of nuke CAPEX. The cost of nuclear increases as you move from the Southwest to the Northeast. The Social Cost of Carbon increases as you move from the Southeast to the Northwest.

At the electricity grid level, the basic trade-off is CO2 emissions versus power cost. Figure 1 summarizes the Gordian Knot Group's study of the German grid. Vertical axis is cost per MWh at the plant gate. CO2 emissions per kWh are on the horizontal. The range of CO2 emissions is large enough that we need to make this axis logarithmic.

These curves are based on multiple runs of the GKG Grid Model. In these runs, the basic rule is the program must supply the actual hourly German demand for electricity for every hour from the beginning of 1993 to the end of 2000. The peak hourly demand in that 8 years was 101 GW. The average over that period was 62 GW. In each run, the model comes up with the combination of onshore wind, offshore wind, PV solar, batteries, hydrogen, open cycle gas turbine, closed cycle gas turbine, coal or nuclear that minimizes the sum of the grid cost and the CO2 emissions cost, the total cost to society of providing the power.

Thanks to its astonishing energy density, nuclear's overnight CAPEX should be less than $2000/kW in current money. This was the case in the late 1960's, when nuclear was just starting down a steep learning curve. But the body politic opted to replace competitive market pressures with a regulatory system whose overriding goal is preventing a release of radioactive material. Under this misdirected system, nuclear's overnight CAPEX in the West escalated to the point where it is now over $16,000/kW. In the Germany study, we ran a range of nuke Capexes running from the should-cost of $2000/kW up to $32,000/kW.

Each solid line in Figure 1 was created by fixing the nuclear overnight Capex at one of the six $/kW numbers shown in the top of the legend and varying the dollar cost to society of an additional ton of CO2 emissions from $1600/ton CO2 at the left upper end of the curve to zero dollars per ton at the right lower end of the curve.

The official name for this dollar cost is the Social Cost of Carbon (SCC). If the Social Cost of Carbon is very high, the societal optimum for any given nuclear CAPEX is near the left end of the curve. If the SCC is near zero, the societal optimum is near the right end of the curve. For each nuclear Capex, we ran the eight CO2 costs shown in the bottom of the legend, creating the 48 points marked on the curves.

Nobody knows what the Social Cost of Carbon is; but that has not stopped people from making guesses at it. The US EPA has been using $51/ton CO2; but recently the EPA proposed increasing the legal SCC by nearly a factor of four to $190/ton.\cite{epa-2023} Other estimates range from negative to over $1000/ton. It is not difficult to concoct hypothetical scenarios that support either end of that range and anything in between. I repeat: we do not know what the social cost of CO2 is. Anybody who claims to know what the SCC is is either a liar or a fool. Therefore, we must consider a wide range of CO2 costs.

The dashed lines are constant Social Cost of CO2 contours. For example, the nearly horizontal reddish dashed line at the bottom of the figure is the zero Social Cost of CO2 contour. Along this line, the model uses either an "All" nuclear grid or an "All" coal grid, where "All" in quotes means "almost all", since these grids include some gas peaking. If the Social Cost of CO2 is zero, the model uses the "All" coal grid for all nuclear Capex's above $4000/kW. So all the solid lines in Figure 1 other than $2000/kW converge to the "All" coal grid at the right end. These two grids are shown in the left-most column of Figure 2.

Figure 2. Three nuclear should-cost grids (top) and three non-nuke grids (bottom). Zero Social Cost of Carbon on the left; very high SCC on the right.¹

In Figure 2, the installed nameplate capacity for each technology is in blue, and read on the left axis. The hashed portion of the blue bars is the amount of that capacity actually sent to the grid. The CO2 emissions are in red and read on the right axis.

As the social cost of CO2 increases, the SCC contours become more humped. At the left end, the SCC contours are tied to should-cost nuclear, and its relatively low grid cost, even at a very high cost of CO2. But as the nuclear CAPEX rises, the grid cost rises rapidly. Eventually the model gets to the point where maximizing societal welfare requires using little or no nuclear. At nuclear CAPEXes above that point, the best the model can do is accept a much higher CO2 intensity in return for a decrease in grid cost. At that point, the SCC contours turn right and downward.

Once the CAPEX is above $8000/kW nuclear, Germany's choices are pretty stark. They are shown in the bottom row of Figure 2. At one end, she can have $60/MWh and 717 gCO2/kWh electricity from a coal grid with gas peaking. At the other end, she can have $167/MWh and 37 gCO2/kWh power from a wind, solar, hydrogen, and gas grid.

The problem is slow moving high pressure systems, which in cold weather produce inversions. The result is protracted periods of little wind and little sun that the Germans call dunkelflauten, dark lulls. Energy storage, even with some pretty optimistic assumptions about green hydrogen, is so expensive that the program does not make a great deal of use of it. Instead it is cheaper for the model to install immense amounts of wind and solar and accept a great deal of curtailment.² In part, this is due to the time and power required to recharge batteries and H2 salt domes. The GKG Model does a pretty good job of modelling these lags.

To bridge the dunkelflauten, the total installed capacity is 541 GW, more than five times the peak hourly load. The total installed capacity is more than 5 times the peak hourly demand, resulting in a wholesale electricity cost of about 17 cents/kWh, three times the cost of the all fossil grid. The resulting curtailment pushes the solar capacity factor down to 0.11 and the onshore wind CF to less than 0.20. In a nearly all wind/solar grid, the standard ``average" capacity factors are both meaningless and misleading, since they ignore curtailment.

It could be worse. Suppose Germany swears off both nuclear and fossil. Figure 3 shows the model's results for two pure wind/solar grids; batteries only on the left and hydrogen only on the right. I've tried to adjust Figures 2 and 3 so the vertical axes are to the same scale. The hatched areas in Figure 2 and the top row of Figure 3 are about the same, but the solid blue, unavailable or curtailed, capacities are so different, it is difficult to believe we are talking about the same function.

Figure 3. Two pure wind/solar grids. Batteries only left; hydrogen only right.

The batteries-only option is a non-starter. Batteries are so expensive that the model finds its cheaper to install humongous amounts of wind and solar, in order to get the amount of batteries needed down to 30 hours of average demand. But this requires installing a total of 1.5 million MW's of nameplate capacity, 15 times the peak hourly demand. The resulting curtailment pushes both the wind and solar capacity factors below 5% and the grid cost to 57 cents/kWh.

What's interesting about this solution is it is not a particularly low CO2 grid. The CO2 intensity is 154 grams CO2 per kWh. The GKG Grid Model is smart enough to distinguish between fixed CO2 and variable CO2. Fixed CO2 are the emissions associated with manufacturing and erecting the plant. They are a function only of the install capacity. Variable CO2 are the emissions associated with the operations, mostly fuel. They are a function of the kWh actually generated. Wind/solar CO2 is almost all fixed. If their capacity factor gets low enough, wind/solar are no longer a low CO2 source.

Thanks to some pretty optimistic assumptions about synthetic hydrogen, and Germany's ample salt domes, the H2 storage option is far better than the battery option. The program opts to install 17 days of average load of H2 storage capacity and more than the average load in both charge and discharge capacity. The total installed capacity is down by a factor of 3. The CO2 intensity is a respectable 27 gCO2/kWh. But the grid cost is an economy crippling 21 cents/kWh. This will force all electricity intensive industries out of Germany and effectively prevent any electrification of current non-grid markets.

The no nuke, no fossil options lead to impoverishment, without doing a really good job on CO2.

Enough Virtue Signaling. Let's Get Real

But this is Fake Jack talking. You know the one that blathers on about societal welfare and pretends to be concerned about all the humans that have little or no access to affordable electricity. Real Jack looks at Figure 1 slightly differently. Figure 4 is Figure 1, except I've added nuke's penetration of the German electricity market to each of the intersections. Penetration here is defined as the ratio of nuclear kWh's generated to the total load kWh's over the 8 year period.

Figure 4. Nuclear's market penetration as a function of nuclear CAPEX and the SCC.

The scoundrels that are quite comfortable with expensive nuclear power are on to something. As long as the perceived Social Cost of Carbon is high, nuclear's monopoly on dispatchable, expandable, very low CO2 power will funnel more money in their direction the more expensive it is. Focus on the $1600/ton SCC curve, the largely vertical, dashed green line on the left side of the diagram. At the bottom of this curve, where nuke is pushed down to its should-cost of $2000/kW, Real Jack is barely surviving in a viciously competitive market. But at we move up the $1600/ton SCC curve, nuke's market penetration does not start really dropping off until we get above $8000/kW. Even at an outrageous $16000 per kW, nuke wins 53% of the market.

Which would the Real Jack rather have, bare breakeven at $2000/kW and 99.96% of the market, or 100.5% of the market at $6000/kW with almost all the extra $4000/kW going to the waste trough whence any numbers of pigs can feed? Economists talk about Consumer Surplus and Producer Surplus. This is the Swine Surplus. We know Real Jack's answer.

But wait a minute. How did nuclear garner more than 100% of the market? That's an easy one. At $1600/ton SCC, it pays the model to make and use hydrogen for some of the peaking. Synthetic hydrogen is an electricity hog with a round trip efficiency of less than 37%. Our market penetrations are based on marketed electricity; but the model has to generate more than that amount of power. Still more money for the Real Jacks.

Of course, this highly lucrative situation depends on an extremely high perceived Social Cost of Carbon. But the same phenomenon is true throughout much of Figure 4. For example, at EPA's new SCC of $200/ton CO2, nuclear's penetration is still over 60% at a CAPEX of $8000/kW, A large portion of the diagram has a nuke penetration above 90%.

There's two ways this ripoff falls apart:

(1) Society's view of the Social Cost of Carbon moves lower. Real Jack has to be careful. If you focus on the northeast portion of Figure 4, you'll see there comes a point where nuclear's market penetration falls off a cliff. The location of that cliff is critically dependent on the perceived SCC. Real Jack's job is to work to keep the perceived SCC high and stay on the high side of the cliff.

(2) Society comes to it senses, and makes the Real Jacks compete with each other under an evidence based, biologically reasonable compensation scheme for radiation exposure. As long as they are forced to compete on a level playing field with no barriers to entry and no wasteful regulation, these greedy bastards will push nuclear down to its should-cost, precisely because they are greedy.

Here's the cool part: (2) works regardless of the Social Cost of Carbon, perceived or actual.

1. In the top right hand corner of Figure 2, the should-cost, extremely high SCC run, almost all of what little CO2 is emitted is produced by nuclear. Further increasing nuclear and decreasing gas would increase CO2 emissions due to the very low marginal nuclear capacity factor. If the capacity factor gets low enough, gas becomes a low CO2 source.

2. The GKG CO2 emissions for these high cost grids is in part phony. The high cost of electricity will push customers away from electricity to direct burning. Electric vehicles become less attractive. More importantly, electricity intensive sectors of the economy will migrate out of Germany, often to high CO2 grids.