Earlier quoted context omitted.
> First, providing the government with the ability to stabilize market prices in the short term when volatility strikes. Which is the thing they don't really even do, because their existence is not a secret, but then knowing of their existence discourages anyone else from setting up a reserve because they expect the government to unload right when they'd be trying to recover the costs of operating it. Then the market…
People aren't as stupid as you appear to think. Yes, there are second (and third, forth, ...) order effects. Typically these sorts of systems will settle into an equilibrium. A reasonably competent government agency will account for that where necessary. It's strange. You object to the government here yet expect private industry to fill the same gap. Why do you believe private industry would navigate these issues bet…
Profit-seeking actors have the direct incentive to balance risks and rewards. It's popular to hate on speculators, but "build a storage facility so you can buy a commodity when it's cheap and sell whenever the price is high" as a means to make money is actually pretty legitimate. And then they have the right incentives to manage costs and keep realistic inventory levels because they're spending their own money instead of someone else's. Whereas the government's incentive is to give lucrative contracts to cronies or hoard a ridiculous amount of the commodity because they're spending someone else's money and get blamed if there's not enough but not if there's too much.
There is also an advantage in diversity. Government tends to monoculture. How much does the price have to go up before the government starts unloading inventory? How much does the answer depend on politics? Things are better when instead of one essentially monopolist with a massive tank, you have a thousand independent entities with small ones, because then you get a smoother curve with less relationship to the election cycle. And you get different people trying to solve the problem in different ways. Speculators build tanks, entrepreneurs develop recycling systems, buyers make contingencies to use a substitute, but none of that happens if everyone is expecting the government to guarantee the price.
> Industry is notoriously bad at making short term sacrifices for long term risk management.
Middle managers in large bureaucracies are notoriously bad at this, because enormous conglomerates insulated from competition and subject to the principal-agent problem are not subject to a good set of incentives in many ways. It's why we're supposed to have antitrust laws.
Markets as a whole are pretty good at it, because "price goes up when supply is low" is a predictable opportunity to make money.
> Would you rather the government force them to maintain their own reserves via regulation?
The whole point is to stop having the people who don't pay the cost of doing it be the ones who choose how much there should be and what kind.
> Rapid price fluctuations are one issue. Essential resources are an entirely separate problem. Volatility and starving to death both involve price movement but are otherwise very different things.
They're the same problem because the problem in both cases is supply less than demand and then you're left with the same question of how best to contend with that.
Notice also that the government doesn't keep a multi-year supply of food and that doesn't seem to be any kind of a problem.
> So if the reserve is run by the government it's removing slack and reducing supply, but when run by private industry ... ?
When it's run by private industry it costs less, and more to the point costs the people who want the buffer instead of strangers without the bandwidth or domain knowledge to know if what's being done is cost effective or even necessary.
> No amount of regular slack is ever going to be able to compensate for a tail risk that blocks the import of an essential good. Take oil for example. No company is ever going to voluntarily warehouse enough to keep the entire US economy going for any significant amount of time. It's a crazy small tail risk and very expensive to counterbalance.
The US is a net exporter of oil and oil is widely traded global commodity with significant price elasticity of demand, so you don't get actual shortages unless you try something foolish like price controls. Instead people pay $4/gallon instead of $3 which causes the people who drive the most to switch to electric cars or hybrids, other suppliers to increase production, etc.
> Why would that be? Filling and emptying shifts demand but doesn't create additional.
Filling creates additional demand but if you're using a large enough reserve to be at low risk of ever running out then by design the emptying never fully happens.
> Anyway you seem to be arguing that private industry should do this for themselves. So whatever the effects are they will be present either way.
Private industry would size the reserve according to the risk instead of having the incentive to be excessively risk averse because they're spending someone else's money.
> Why do you expect disproportionate expenditures? The cost is that of warehousing.
Suppose you have a reserve which holds X amount and there is an average annual withdrawal and refilling of 0.5X, once every ten years you would use the full X amount, and once every 50 years you would use 5X if you had it.
The 5X reserve requires five times as many tanks and requires you to eat the time value of money on five times as much of the commodity, but only gets used once every 50 years instead of being mostly used every year. It's not worth having; it's better to eat the higher prices that year than to pay even more to prevent them. There are some risks it costs less to buy insurance against than to mitigate. But risk-averse people spending someone else's money will be more inclined to do it anyway, or to build a 10X reserve "just to be sure".
The government also uses government contractors which do not have a good record for cost efficiency.