If your job is to establish a stable market between producers and consumers, and the market price you set swings positive and negative, through 6 orders of magnitude in just 72 hour...you are not doing a good job.
There is quite a bit of elitism when programmers talk about real-life events but admittedly less here on HN than in many other forums.
I am sure running a power grid with thousands of producers and millions of consumers is incredibly complex. Did you know every wind turbine and generator is spinning exactly in sync across the whole grid. I am amazed that it works.
What obviously does not work is the libertarian fantasy of live market pricing for turning generators on and off. What are the chances of rolling blackouts on Monday as oil cracking plants come online to take advantage of free power?
So what do you recommend? Mandatory lights/heaters/AC on/off times, so the power produced is actually used, and the power needed is actually produced?
The british have the problem that during a football game half-time, a bunch of people start their tea kettles at the same time, and there is a relatively huge spike in electricity usage... this means, that the electricity providers have to plan for this and import electricity and/or shut down other cunsumers for those 15 minutes. Now imagine an unplanned half-time, without the time to prepare, and a magnitude higher changes, that happen suddenly (eg, three/ice brings down cables)... there's no way to predict everything and deal with every problem.
What variable pricing does is, that it gives incentive to consumers to use electricity when it's cheaper and not use it when it's more expensive. For homes that means washing clothes in the evenings and on sundays (cheper electricity here then) and not during peak usage hours, and for industry that means shutting down the highest consumers when there is not enough power available, and having them use up extra power when needed.
My understanding is Texas doesn't even have connections to other grids?
Historically the UK had "economy 7" usage, where domestic electricity was cheaper at night. It's not convenient, it stems back from the end of the socialist era of the 70s where individuals took a backseat to the good of society. Most users only use it to boil a hot-water tank and 'night storage' heaters.
It's a shoddy solution for a problem that shouldn't exist in an advanced economy.
Just because something is complex, it doesn't mean there is no one with the required expertise. Texas chose Ayn Rand levels of deregulation with its energy market, and it will keep struggling with black swan events since those are the types which the private industry tends to chronically underinvest for, since they are the easiest to cut and turn into profit.
Is it complicated? Of course. But evidence is pretty damning in this event.
The real question is does the spot price being able to go to $9,000 produce a useful market signal? Or is it just a pointless hazard to ensnare the unprepared? Normally this kind of question doesn't make sense because if there are still transactions at a high clearing price someone has decided they're getting $9k of utility. But in this case the market is a post facto settlement system decoupled from most usage, rather than something built on deliberate transactions.
Also, does anyone know who ends up paying for energy usage that cannot be attributed to any specific customer? Like if someone jumps their meter while the spot price is at $9/kWh, does the local transmission company just end up eating that or what?
However, I believe that in the last polar vortex incident, something like 20% of people paid on the capacity market failed to be available. Unless there are steep penalties built into these contracts for failure to produce when needed, there still might not be proper incentive to perform proper design and maintenance.
As we go forward, it will be interesting to see which incidents fall under force majeure, and which incidents result in actually breaking contracts.
This particular incident was foreseen: the 2011 FERC report on smaller similar outages in ERCOT recommended fixes to prevent this. And I know of at least one energy expert who was tweeting days before this happened that he hoped Texas had fixed their problems that were identified in 2011.
Personally, I favor regulating the ever living hell out of it, and having the state force carbon free electricity by 2035. I think that's the only way we will see movement towards more reliability and less carbon: guarantee that there's a market for such products.
If electricity was like gas the pump would refuse to operate since your wallet is too light.
To think this would not have an effect on energy demand, and thus price, is delusional.