Hacker News Reader

Comment by Ericson2314 | Hacker News Reader

> but all else being equal (ie: assuming total production cost stays constant), it’s better to have a larger fraction of your electricity costs be variable, so that if demand drops then production cost drops as well.

This is how capitalism gives us scarcity when there is no reason for scarcity. Infrastructure like trains and nuclear that absolutely swamps current demand, and whose costs cannot be adjusted very much (even if one can run fewer trains, remove some fuel rods) is good. But woe unto anyone that floods the market utopia-style under capitalism --- you will just get paid nothing and your investors will not repeat such a project again.

View on HN

Another way to interpret that sentence is “it’s better not to spend too much on construction, and it can be better to spend on fuel if it means spending less on construction.”

Building things you don’t need is waste of labor and resources, regardless of the economic system you use or who paid for it. This should be accounted for somehow.

What is needed changes over time.

It is better to spend more on capital than operational costs if one actually wants the world to change not just ideal along.

The best cost control is in large part doing lots of cookie-cutter work, hence the focus on small modular reactors. It is better to do those, and then ramp up the "small" part over time.

Likewise, we should construct lots of rail simultaneously with ramp up to improve those supply chains too.

But at no point do you want to spend more on fuel in any global optimal sense. That is just wasteful. It's not like we are actually uncertain that we won't need way more electricity production on a societal scale.

I agree that we are going to need more power generation as electricity replaces other forms of energy.

But for any given level of electricity generation, doing it in a cheaper way is still better than doing it in a more expensive way. For one thing, it allows capacity to be increased for the same cost.

The pricing is non-linear, both in terms of total output (due to economies of scale if done right) and especially in terms of share (are their fossil fuel PPs to cheaply fill in gaps).

If your costs are all fixed, and demand falls, you have to put prices up to stay alive. That's the issue with Nuclear: If it costs $1bn a year for the plant, it costs that whether you generate 1bn kWh (at $1 each) or 1 kWh (at $1Bn each).

What you call artificial shortage, others call "not making something no one wants and then forcing them to pay for it"...

There is no non-depressing future where demand for electricity doesn't go way up as fossil fuel is phased out. An uncertainty is a shit situation we should work to prevent, not have a contingency plan for.

This is where the Keynesian "socialization of investment" stuff comes in. Private markets get skiddish over change even when our economies are fully capable of dealing with the issues. Don't be held hostage to private capital playing "nose goes" when the solution is obvious.

Abundance really does mean everything is too cheap to meter on margin. There is no other definition. To get there we have to make people more risk tolerant, and the only way to do that is guaranteeing consumption.

It's not so much that your points are wrong, as that they fail to see a bigger picture.

When you say there is no no depressing future without increased electrical demand, that's true. But who guaranteed you a non depressing future?

And the same was true in 1990, but we've made almost no progress to removing fossil fuels. So if you'd started nuclear projects then, you'd be just coming online. And have no new customers. And in the mean time the price of gas and solar and wind would have collapsed in comparison. And you couldn't afford to shut down when the price crashed like they can. You're fixed costs would rapidly drive you bankrupt.

That's the problem here.

The future isn't nice. And it's not predictable. Even over relatively short periods.

So it's much better to plan 6m ahead and then do it again 40 times, than to try and plan 20 years ahead in one go. That makes it much much better to run small incremental short-term projects like gas plants and renewables.

And if you do have to make some sort of medium term plan then it needs to be very flexible. Like a gas plant that can shut down for a year and avoid 90% of its costs.

Nuclear is long term, fixed cost. And that's terrible.

But go ahead, by some shares in a nuclear operator (or someone else in the supply chain etc). Make your fortune being right in 2045. Just don't sign up the rest of us via the public finances please!

This has always been one of the issues with control economies: people vastly over estimate their ability to predict the future, they don't appreciate flexibility and they don't manage risk. That's how the USSR ended up producing enormous amounts of steel and no washing machines: no one asked what people wanted, they just said 20% more of the same compared to last year! China are doing the same thing right now with electricity targets and no financial services...

> But who guaranteed you a non depressing future?

Why contingency plan?

Your USSR analogy is apt but I think it makes my case. The USSR was planning the steel production and its uses. Electricity is the single most useful good with the most diverse downstream uses. I am not saying we should command all the uses of the electricity! But command-economying the electricity production itself is the a no-brainer.

One 6m and then 7m and then 8m all the way up to 20m. that is building increasingly bigger reactors with an efficient supply chain.

You're over-using markets and doing so in a circular way with your argument. You are confusing actual uncertainty around the world with the uncertainty that comes from the decentralized, loosely coordinate market-governance mechanism itself!

Some things are uncertain, and market are really great for small-scale experiments or "annealing" society. But or the brain-dead obvious things like "we need tons more electricity, rail, and dense housing ASAP!" there is no reason to rely on markets.

A good society has a socialist foundation and capitalist superstructure: guarantee the basics, but have fun with the rest. That is already how rich people enjoy capitalism (wealthy family entrepreneurs are not so leveraged), let's just scale that out to everyone. We can't all be top X% wealth, but we can all live with less uncertainty.

Cow's milk in the United States seems to be an exception to this

Reply on news.ycombinator.com