As someone who regularly gets into arguments about this, I can say that he’s definitely right; you wouldn’t believe the amount of nonsense that can be disposed of simply by looking up the relevant numbers and doing a minute’s worth of easy arithmetic.
For example, I’ve heard some people recently claiming that a combination of solar photovoltaics, electrolysis to produce hydrogen, and these new Bloom box fuel cells are cheaper than nuclear fission. Look up the costs of solar farms; about $3 per peak watt. Their average power output is less; we can very optimistically assume that they run at 20% of capacity on average. Efficiency losses from electrolysis and fuel cells are about 50%. Putting it all together, this would cost about $30 per watt of average power delivered. Not including the cost of the fuel cells.
A little googling will show that the total cost of building two new AP1000 reactors in Georgia is about $14 billion, and they average at least 93% of their peak power, and transmission line losses bring their average power delivered to about 1000 MW each. So their cost is about $7 per watt of average power delivered, or about 23% the cost of solar.
There’s a lot of extremely harmful bullshit out there, and defeating most of it doesn’t take any advanced techniques; it just takes a willingness to look up some relevant numbers and do a bit of arithmetic.
up-voted, but I don’t think it’s simple arithmetic that they’re missing, there’s a lot of ideological baggage preventing the masses from seeing nuclear as the better alternative.
I’d caution against such under estimation of people’s mental capacities, if only they knew how to add and subtract almost entirely misses the point-and is too condescending, not good PR.
I never said that they’re incapable of doing the math, just that they don’t. For whatever reason. No further condescension is intended; just a really helpful suggestion.
You can make the calculation return any result you want, for example by including cost of millennia of nuclear waste storage in price of nuclear power; another thing—nuclear power gets massive federal insurance subsidies (but then coal gets free license to kill people by pollution etc., so it’s not exclusively nuclear problem).
If you know what result you want, you will be able to come up with it.
You can make the calculation return any result you want, for example by including cost of millennia of nuclear waste storage in price of nuclear power;
You can calculate arbitrarily high costs for anything if you try hard enough. What of it? We’re not going to deal with nuclear waste by sticking it in Yucca Mountain and guarding it for thousands of years; that would be silly. Here’s a summary of how to realistically deal with nuclear waste. We have more than enough money for this budgeted as part of every nuclear plant’s operating and maintenance fees.
another thing—nuclear power gets massive federal insurance subsidies
Not true (PDF warning). The nuclear industry runs its own insurance pool, paid for out of their own pocket. The regulations requiring this do say that the federal government may help out in extreme circumstances (i.e. something on the scale of Chernobyl) but to date the feds haven’t spent a dime on this. And I see no reason to believe that they ever will.
If you know what result you want, you will be able to come up with it.
If you’re motivated to play games with the figures, consciously or not, then sure you can. But I try to avoid that sort of thing, and it tends to be pretty obvious.
Note that I’m not accusing you of dishonesty—but I’m guessing that you ultimately got those arguments from someone who was trying to make the numbers fit his position, rather than the other way around.
Another thing to calculate on the cost of nuclear power:
photovoltaics don’t have evacuation plans, labled evacuation routes, large government monitoring safety boards, or National Guard/Air Force aerial defense concerns.
It’s hard to look up data on so-called “externalities” like that.
Solar power requires heavy industry to build, and that has loads of externalities. It takes up a lot of space and affects local climate and ecology. And then there’s the unreliability of the sun, which can have economic consequences.
As for the nuclear externalities you mentioned, the evacuation planning and government safety things are paid for by power plant fees, and budgeted into the cost of building and operating the plants. Defending the plants is something you have to do with all forms of power generation, and I actually think you’re miscalculating the risks by looking at the power plants themselves, which (in the case of nuclear) tend to be pretty beefy and well-guarded. Attacking the transmission lines would be much easier, and much harder to defend against. This goes double for wind and solar farms that are located far away from everything and have to use longer power lines.
(And really, what are the odds you’ll ever have to use those evacuation plans? I’d worry more about crossing the street. No water-moderated reactor has ever had an accident that made evacuating people nearby a good idea, even after all these decades of operating them, and there are good theoretical and practical reasons to believe that it never will.)
And while we’re looking at externalities, consider this: nuclear is the only option that’s currently competitive with coal on a cost-per-kWh basis. Very cautious safety regulations, by holding nuclear power back, are responsible for a lot of coal emissions—which are far more dangerous than anything people are talking about for nuclear plants. Paradoxically, our worries about nuclear safety have made us much less safe. What we have here is a widespread failure to shut up and multiply.
I really like this as a test-case for rationality, because it’s important and we really can look at it probabilistically for insight.
As a point withiin the greater whole, I don’t think that the security requirements of photovoltaics are the same as those for a nuclear reactor. Also, security difference between “spent” photovoltaic cells v. spent nuclear fuels?
In any case, the overarching point: “you wouldn’t believe the amount of nonsense that can be disposed of simply by looking up the relevant numbers and doing a minute’s worth of easy arithmetic.” turns out to not be so simple, because there are a lot of issues involved. There’s the issue of disposal, which you show a link to, but you don’t seem to have incorporated those numbers. There’s the issue of how much taxpayer money goes into scrambling jets near Indian Point each time alerts are raised, etc. The calculation clearly isn’t easy in the present, and also does not incorporate the cost/benefits analysis of focusing on photvoltaics because ultimately they will almost certainly be more efficient (your efficiency/production rates for existing infrastructure doesn’t reflect changes and advances in technology, of which there are many). In short, “easy arithmetic” isn’t always so easy.
I won’t bother looking up figures as I’m not terribly interested in long term nuclear waste, but you’re wrong about insurance (at least your argument is wrong, not necessarily the conclusion).
Implicit or explicit guarantee for extreme cases are worth trillions. There were some papers measuring how much even implicit guarantee was worth for Fannie Mae/Freddie Mae, and this was enormous amounts by letting them raise money far cheaper than would be otherwise possible (and taxpayers eventually paid, but it was beneficial to Fannie/Freddie long before that).
Do you consider the law regarding car liability insurance to be a subsidy? It requires you to carry liability insurance up to a finite amount, despite the fact that you can do much more damage than that with your car, and then bankruptcy law will shield you from paying the full amount.
This is the same kind of insurance nuclear plants have: they’re require to have an insurance on up to $X of damages, and then “someone else” bears any cost beyond this.
Nuclear plants can’t be insured for the damages in a meltdown. Not because the risk is so huge that it should never be done, but because any jury award would be effectively infinite, irrespective of the actual damage. There’s no point to buying insurance when the uncovered liability increases in lockstep with your insurance coverage. However, the actual meltdown risk is extremely small and even the required insurance is effectively overinsuring the plants.
This nuclear plant “insurance” can’t be compared to what FM/FM had because they are able to continue operation and making profits after a “meltdown”, while a nuclear plant would be over and done with.
If you don’t like the kind of uncovered liability nuclear plants have, they’re the least of your concerns—you really should be advocating an end to driving, since no driver can meet the insurance standard you seem to expect out of nuclear plants.
Now, with that said, you are correct that comparisons of green technologies to coal do conveniently leave off the damage that coal plants spill off onto other people and are therefore misleading. I’ve long railed against assessments of coal that ignore the cost of dumping toxic crap into people’s lungs. Example. (ETA:Better example.) Still, that requires an objective accounting of environmental costs, not just (as is often the case) assuming they’re infinite.
Making drivers not responsible for damages they cause is a massive subsidy, and without it we’d have far more investment in car safety (and I mean genuine kind like replacing human drivers with robots, black boxes, and compulsory alcohol testers before it lets you drive, not current air bag waste of time), and far fewer deaths and injuries.
Pardon, but I don’t see how this is responsive to my comment.
1) Drivers are made responsible for the damages they cause, up to the limits imposed by bankruptcy law; the law also attempts to ensure [sic] that each driver on the road is capable of paying up to $X in damages. What they are not made responsible for is arbitrarily large damage they could potentially do, but this is unavoidable—no one is capable of setting aside that much money, even solar power operators (or rich people).
2) In absence of “making drivers not responsible for damages they cause”, we most certainly would not have more investment in car safety; we wouldn’t have cars, period. (BrE: Full stop.) Or, without the multiple negatives: If everyone driving had to be capable paying all damages they could ever potentially do with their vehicle, no one would be allowed to drive, or use most technologies. I don’t think you’re understanding the implications of this requirement.
Yes, drivers—and nuke plants—should carry insurance. Maybe the required amount (in either case) is too high. Or too low. Or derived from the wrong process. But no one can insure unlimited liability, so the safety improvements you describe just wouldn’t happen if that were a requirement; the technology just wouldn’t be used. But once you accept that people should only have to insure up to a finite amount, and given the low, self-borne risk of nuclear plants, you must accept that they already meet this.
3) Arguably, the reason we don’t already have self-driven cars is precisely the phenomenon I warned about: uncovered liability increasing in lockstep with coverage. The average person who kills someone with their vehicle is typically required to pay a lot less than when it is done by a wealthy corporation. Given jury reactions to new technologies and wealthy corporations, if someone actually did offer self-driving cars, they could very well have to pay out more in damages, even if they were safer than 99% of human drivers!
You can’t run cars with power that comes directly through the power line.
You ignore the running cost of the nuclear reactors. You don’t price risk from blowups and you don’t price long term storage costs. Risk from peak uranium http://www.technologyreview.com/blog/arxiv/24414 is also unpriced.
If the word “average” would be meaningful in this context than you would simply compare solar cell productivity + transmission line losses to nuclear plant costs + transmission line losses.
Of course you can do simply arithmetic but that doesn’t mean that you are right when it’s not clear that you are using the right numbers.
You can’t run cars with power that comes directly through the power line.
No, but you can power people’s homes, businesses, and industry. Currently we’re burning ungodly huge amounts of coal to do that. Just because a green energy source isn’t the final solution to all energy needs is not a point against it.
You ignore the running cost of the nuclear reactors.
I was addressing the argument, often put forward, that the up-front costs of nukes are much higher per-kilowatt than other green energy sources. The operations and maintenance costs of nuclear energy are so low that they are seldom attacked. Here are some approximate numbers
You don’t price risk from blowups and you don’t price long term storage costs.
Those are both included in the O&M costs, actually. And by the way, name me a single light water reactor that has blown up. Or any modern reactor, for that matter. There are good reasons to believe that such an event is very unlikely or (in some cases) actually impossible.
If the word “average” would be meaningful in this context than you would simply compare solar cell productivity + transmission line losses to nuclear plant costs + transmission line losses.
Construction costs? O&M costs? O&M plus loan payments until the up-front investment has been amortized off? There’s more than one type of cost to consider, so I decided to focus on the particular argument that construction costs were too high.
As for “average” being meaningless, it’s true that I’ve ignored transmission line losses. Those are not high enough to significantly affect the calculation, and both nuclear and solar tend to have longer-than-average distances between the plants and the consumers, so I doubt there would be too much difference between them (unless you went with something like the idea of putting solar farms in the Sahara desert and sending the electricity to Europe).
You aren’t comparing the price of nuclear vs. the price of solar but the price of nuclear vs. solar + hydrogen.
By your own numbers the price of solar is 3$ per watt while the price of nuclear is 7$ per watt.
Your solar power plant that’s backuped with hydrogen produces the energy at different prices at different times.
While a nuclear plant can produce the same amount of power at night than at day it’s not possible to change the amount that gets produced as fast as you can change how much hydrogen you burn in fuel cells.
You aren’t comparing the price of nuclear vs. the price of solar but the price of nuclear vs. solar + hydrogen.
Oh crap, you’re right. I got this confused with another discussion. Sorry about that. Anyway, the latter is a more meaningful thing to compare.
By your own numbers the price of solar is 3$ per watt while the price of nuclear is 7$ per watt.
That’s cost per peak watt; a more relevant number is cost per average watt (assuming perfect energy storage at no cost). To get that, you have to multiply by the capacity factor. For new nuke plants, that’s about 93%. For solar, it about maxes out at 20%. So construction cost per average watt would be about $7.50 for nuclear and $6 for solar.
Of course there’s more to it. There’s the cost of storage and backup, and maintenance, and of course the plant lifetimes differ by a factor of 3-4, and both types of power will get significantly cheaper to build over the next decade or so. But as a first approximation, you could do worse than multiplying peak cost by capacity factor.
While a nuclear plant can produce the same amount of power at night than at day it’s not possible to change the amount that gets produced as fast as you can change how much hydrogen you burn in fuel cells.
Correct (for large light water reactors). Power grids do need the ability to adjust production to meet rapid changes in demand. What of it?
Very cool, although it needs to become a lot cheaper if it’s going to be competitive. I see a viable niche for solar in places like California, where the air conditioning needs in the summer cause the peak power to come at a time when the sun is shining brightest. This is rough on the power grid. Solar panels could be very useful for smoothing out the peaks there, if it can be made cheap enough.
Meanwhile, nuclear has a new wave of modular reactors coming. It’s going to get quite a bit cheaper, too, and it’s still nowhere near its full potential, as the LFTR folks can attest.
Anyway, nerding out aside, my point remains that simple arithmetic is necessary and often sufficient to discuss this sort of thing like grown-ups.
“He who refuses to do arithmetic is doomed to talk nonsense.”
-John McCarthy, on mainstream environmentalism.
As someone who regularly gets into arguments about this, I can say that he’s definitely right; you wouldn’t believe the amount of nonsense that can be disposed of simply by looking up the relevant numbers and doing a minute’s worth of easy arithmetic.
For example, I’ve heard some people recently claiming that a combination of solar photovoltaics, electrolysis to produce hydrogen, and these new Bloom box fuel cells are cheaper than nuclear fission. Look up the costs of solar farms; about $3 per peak watt. Their average power output is less; we can very optimistically assume that they run at 20% of capacity on average. Efficiency losses from electrolysis and fuel cells are about 50%. Putting it all together, this would cost about $30 per watt of average power delivered. Not including the cost of the fuel cells.
A little googling will show that the total cost of building two new AP1000 reactors in Georgia is about $14 billion, and they average at least 93% of their peak power, and transmission line losses bring their average power delivered to about 1000 MW each. So their cost is about $7 per watt of average power delivered, or about 23% the cost of solar.
There’s a lot of extremely harmful bullshit out there, and defeating most of it doesn’t take any advanced techniques; it just takes a willingness to look up some relevant numbers and do a bit of arithmetic.
up-voted, but I don’t think it’s simple arithmetic that they’re missing, there’s a lot of ideological baggage preventing the masses from seeing nuclear as the better alternative.
I’d caution against such under estimation of people’s mental capacities, if only they knew how to add and subtract almost entirely misses the point-and is too condescending, not good PR.
politics is the mind killer seems to be relevant here.
I never said that they’re incapable of doing the math, just that they don’t. For whatever reason. No further condescension is intended; just a really helpful suggestion.
ok, I see your point.
You can make the calculation return any result you want, for example by including cost of millennia of nuclear waste storage in price of nuclear power; another thing—nuclear power gets massive federal insurance subsidies (but then coal gets free license to kill people by pollution etc., so it’s not exclusively nuclear problem).
If you know what result you want, you will be able to come up with it.
You can calculate arbitrarily high costs for anything if you try hard enough. What of it? We’re not going to deal with nuclear waste by sticking it in Yucca Mountain and guarding it for thousands of years; that would be silly. Here’s a summary of how to realistically deal with nuclear waste. We have more than enough money for this budgeted as part of every nuclear plant’s operating and maintenance fees.
Not true (PDF warning). The nuclear industry runs its own insurance pool, paid for out of their own pocket. The regulations requiring this do say that the federal government may help out in extreme circumstances (i.e. something on the scale of Chernobyl) but to date the feds haven’t spent a dime on this. And I see no reason to believe that they ever will.
If you’re motivated to play games with the figures, consciously or not, then sure you can. But I try to avoid that sort of thing, and it tends to be pretty obvious.
Note that I’m not accusing you of dishonesty—but I’m guessing that you ultimately got those arguments from someone who was trying to make the numbers fit his position, rather than the other way around.
Another thing to calculate on the cost of nuclear power:
photovoltaics don’t have evacuation plans, labled evacuation routes, large government monitoring safety boards, or National Guard/Air Force aerial defense concerns.
It’s hard to look up data on so-called “externalities” like that.
Solar power requires heavy industry to build, and that has loads of externalities. It takes up a lot of space and affects local climate and ecology. And then there’s the unreliability of the sun, which can have economic consequences.
As for the nuclear externalities you mentioned, the evacuation planning and government safety things are paid for by power plant fees, and budgeted into the cost of building and operating the plants. Defending the plants is something you have to do with all forms of power generation, and I actually think you’re miscalculating the risks by looking at the power plants themselves, which (in the case of nuclear) tend to be pretty beefy and well-guarded. Attacking the transmission lines would be much easier, and much harder to defend against. This goes double for wind and solar farms that are located far away from everything and have to use longer power lines.
(And really, what are the odds you’ll ever have to use those evacuation plans? I’d worry more about crossing the street. No water-moderated reactor has ever had an accident that made evacuating people nearby a good idea, even after all these decades of operating them, and there are good theoretical and practical reasons to believe that it never will.)
And while we’re looking at externalities, consider this: nuclear is the only option that’s currently competitive with coal on a cost-per-kWh basis. Very cautious safety regulations, by holding nuclear power back, are responsible for a lot of coal emissions—which are far more dangerous than anything people are talking about for nuclear plants. Paradoxically, our worries about nuclear safety have made us much less safe. What we have here is a widespread failure to shut up and multiply.
I really like this as a test-case for rationality, because it’s important and we really can look at it probabilistically for insight.
As a point withiin the greater whole, I don’t think that the security requirements of photovoltaics are the same as those for a nuclear reactor. Also, security difference between “spent” photovoltaic cells v. spent nuclear fuels?
In any case, the overarching point: “you wouldn’t believe the amount of nonsense that can be disposed of simply by looking up the relevant numbers and doing a minute’s worth of easy arithmetic.” turns out to not be so simple, because there are a lot of issues involved. There’s the issue of disposal, which you show a link to, but you don’t seem to have incorporated those numbers. There’s the issue of how much taxpayer money goes into scrambling jets near Indian Point each time alerts are raised, etc.
The calculation clearly isn’t easy in the present, and also does not incorporate the cost/benefits analysis of focusing on photvoltaics because ultimately they will almost certainly be more efficient (your efficiency/production rates for existing infrastructure doesn’t reflect changes and advances in technology, of which there are many).
In short, “easy arithmetic” isn’t always so easy.
I won’t bother looking up figures as I’m not terribly interested in long term nuclear waste, but you’re wrong about insurance (at least your argument is wrong, not necessarily the conclusion).
Implicit or explicit guarantee for extreme cases are worth trillions. There were some papers measuring how much even implicit guarantee was worth for Fannie Mae/Freddie Mae, and this was enormous amounts by letting them raise money far cheaper than would be otherwise possible (and taxpayers eventually paid, but it was beneficial to Fannie/Freddie long before that).
Do you consider the law regarding car liability insurance to be a subsidy? It requires you to carry liability insurance up to a finite amount, despite the fact that you can do much more damage than that with your car, and then bankruptcy law will shield you from paying the full amount.
This is the same kind of insurance nuclear plants have: they’re require to have an insurance on up to $X of damages, and then “someone else” bears any cost beyond this.
Nuclear plants can’t be insured for the damages in a meltdown. Not because the risk is so huge that it should never be done, but because any jury award would be effectively infinite, irrespective of the actual damage. There’s no point to buying insurance when the uncovered liability increases in lockstep with your insurance coverage. However, the actual meltdown risk is extremely small and even the required insurance is effectively overinsuring the plants.
This nuclear plant “insurance” can’t be compared to what FM/FM had because they are able to continue operation and making profits after a “meltdown”, while a nuclear plant would be over and done with.
If you don’t like the kind of uncovered liability nuclear plants have, they’re the least of your concerns—you really should be advocating an end to driving, since no driver can meet the insurance standard you seem to expect out of nuclear plants.
Now, with that said, you are correct that comparisons of green technologies to coal do conveniently leave off the damage that coal plants spill off onto other people and are therefore misleading. I’ve long railed against assessments of coal that ignore the cost of dumping toxic crap into people’s lungs. Example. (ETA: Better example.) Still, that requires an objective accounting of environmental costs, not just (as is often the case) assuming they’re infinite.
Making drivers not responsible for damages they cause is a massive subsidy, and without it we’d have far more investment in car safety (and I mean genuine kind like replacing human drivers with robots, black boxes, and compulsory alcohol testers before it lets you drive, not current air bag waste of time), and far fewer deaths and injuries.
Pardon, but I don’t see how this is responsive to my comment.
1) Drivers are made responsible for the damages they cause, up to the limits imposed by bankruptcy law; the law also attempts to ensure [sic] that each driver on the road is capable of paying up to $X in damages. What they are not made responsible for is arbitrarily large damage they could potentially do, but this is unavoidable—no one is capable of setting aside that much money, even solar power operators (or rich people).
2) In absence of “making drivers not responsible for damages they cause”, we most certainly would not have more investment in car safety; we wouldn’t have cars, period. (BrE: Full stop.) Or, without the multiple negatives: If everyone driving had to be capable paying all damages they could ever potentially do with their vehicle, no one would be allowed to drive, or use most technologies. I don’t think you’re understanding the implications of this requirement.
Yes, drivers—and nuke plants—should carry insurance. Maybe the required amount (in either case) is too high. Or too low. Or derived from the wrong process. But no one can insure unlimited liability, so the safety improvements you describe just wouldn’t happen if that were a requirement; the technology just wouldn’t be used. But once you accept that people should only have to insure up to a finite amount, and given the low, self-borne risk of nuclear plants, you must accept that they already meet this.
3) Arguably, the reason we don’t already have self-driven cars is precisely the phenomenon I warned about: uncovered liability increasing in lockstep with coverage. The average person who kills someone with their vehicle is typically required to pay a lot less than when it is done by a wealthy corporation. Given jury reactions to new technologies and wealthy corporations, if someone actually did offer self-driving cars, they could very well have to pay out more in damages, even if they were safer than 99% of human drivers!
The motto of this book on sustainable energy is Every BIG helps
You can’t run cars with power that comes directly through the power line.
You ignore the running cost of the nuclear reactors. You don’t price risk from blowups and you don’t price long term storage costs. Risk from peak uranium http://www.technologyreview.com/blog/arxiv/24414 is also unpriced.
If the word “average” would be meaningful in this context than you would simply compare solar cell productivity + transmission line losses to nuclear plant costs + transmission line losses.
Of course you can do simply arithmetic but that doesn’t mean that you are right when it’s not clear that you are using the right numbers.
No, but you can power people’s homes, businesses, and industry. Currently we’re burning ungodly huge amounts of coal to do that. Just because a green energy source isn’t the final solution to all energy needs is not a point against it.
I was addressing the argument, often put forward, that the up-front costs of nukes are much higher per-kilowatt than other green energy sources. The operations and maintenance costs of nuclear energy are so low that they are seldom attacked. Here are some approximate numbers
Those are both included in the O&M costs, actually. And by the way, name me a single light water reactor that has blown up. Or any modern reactor, for that matter. There are good reasons to believe that such an event is very unlikely or (in some cases) actually impossible.
As for long-term waste management, I’ve addressed that in more detail here. It’s surprisingly straightforward.
Construction costs? O&M costs? O&M plus loan payments until the up-front investment has been amortized off? There’s more than one type of cost to consider, so I decided to focus on the particular argument that construction costs were too high.
As for “average” being meaningless, it’s true that I’ve ignored transmission line losses. Those are not high enough to significantly affect the calculation, and both nuclear and solar tend to have longer-than-average distances between the plants and the consumers, so I doubt there would be too much difference between them (unless you went with something like the idea of putting solar farms in the Sahara desert and sending the electricity to Europe).
You aren’t comparing the price of nuclear vs. the price of solar but the price of nuclear vs. solar + hydrogen.
By your own numbers the price of solar is 3$ per watt while the price of nuclear is 7$ per watt.
Your solar power plant that’s backuped with hydrogen produces the energy at different prices at different times. While a nuclear plant can produce the same amount of power at night than at day it’s not possible to change the amount that gets produced as fast as you can change how much hydrogen you burn in fuel cells.
Oh crap, you’re right. I got this confused with another discussion. Sorry about that. Anyway, the latter is a more meaningful thing to compare.
That’s cost per peak watt; a more relevant number is cost per average watt (assuming perfect energy storage at no cost). To get that, you have to multiply by the capacity factor. For new nuke plants, that’s about 93%. For solar, it about maxes out at 20%. So construction cost per average watt would be about $7.50 for nuclear and $6 for solar.
Of course there’s more to it. There’s the cost of storage and backup, and maintenance, and of course the plant lifetimes differ by a factor of 3-4, and both types of power will get significantly cheaper to build over the next decade or so. But as a first approximation, you could do worse than multiplying peak cost by capacity factor.
Correct (for large light water reactors). Power grids do need the ability to adjust production to meet rapid changes in demand. What of it?
Not to dispute your point, but solar photovoltaics should hopefully soon become much cheaper:
http://media.caltech.edu/press_releases/13325
Very cool, although it needs to become a lot cheaper if it’s going to be competitive. I see a viable niche for solar in places like California, where the air conditioning needs in the summer cause the peak power to come at a time when the sun is shining brightest. This is rough on the power grid. Solar panels could be very useful for smoothing out the peaks there, if it can be made cheap enough.
Meanwhile, nuclear has a new wave of modular reactors coming. It’s going to get quite a bit cheaper, too, and it’s still nowhere near its full potential, as the LFTR folks can attest.
Anyway, nerding out aside, my point remains that simple arithmetic is necessary and often sufficient to discuss this sort of thing like grown-ups.