Hacker News Reader

Comment by Osiris | Hacker News Reader

If the market solves all problems why would the government even need to mandate winterization? Wouldn't smart businesses have done the risk analysis and done the work? Maybe the problem here is that loss of electrical power has a significant external cost that isn't absorbed by the utilities. In other words, the cost of winterization was more than the cost to fix the damage to their equipment but doesn't take into account the cost to everyone else.

Externalities are, in my opinion, one of the primary reasons for market failure.

View on HN

Strongly agree.

Free markets often fail to adopt low cost protections to tail events. E.g. the free market didn't get us 100% of new cars having seatbelts and airbags, government requirements did. And those standards were fought by the car manufacturers.

My point is we need regulations to mandate our energy system to be resilient to outlier events (heat/cold/flooding/high wind etc.) and that's a government function separate from the free market structure of ERCOT and actually upstream of the market ERCOT creates.

Said differently, I don't want people to critique ERCOTs market structure (a huge success) when the real issue is the government resiliency regulations in Texas (a huge failure).

I don't see how the free market could have ever succeeded here. Even if it somehow accounted for events of the last week: the lost lives, damaged properties, and economic impact, it was billed as a once-in-100 year event.

Which company is going to take a hit to their profits each quarter so they can prepare properly? The companies who don't will be more competitive and successful right up until a freak event actually occurs. And if that happens, the absolute worst case is the your under prepared company goes under. Unlimited upsides and very limited downsides.

If generators were required to be winterized in order to access the market, that would have solved the problem.

No individual company will take the profit hit without a regulation forcing all the actors to do it.

I agree with your point, but want to point out that this last happened in 2011, and before that in 1989. Certainly not a 100-year event, and likely to become more frequent.

Having a power grid that is heavily reliant on burning fossile fules is a pretty absurdly bad externality too.

What if I've got a prepper bent and am OK with losing power for a few days in a 1:100 year event? Can I still get cheap, slightly unreliable power?

Not to make light of the events in Texas - it sounds scary and they need to be prepared for next time. But the answer isn't automatically "they should have had a gold plated the grid!". The answer might be "they should be have been prepared for a few days without power in extreme conditions".

Every new 9 gets added to 99.9...% reliability figure costs a lot more than the 9 before. At some point, it is more effective for rare events to be handled on a case-by-case basis.

This is a week about every 10 years best case, right now, where I’m at (near Austin). We lose water for twice that long every 5 years. The reliability is probably less than 99% for power, and 95% for water.

Fellow Austinite here. To what events are you referring? I have no recollection of multiday power or water outages in the last 10 years.

We had a boil water notice that lasted for a couple weeks after bad storms just a year or two ago, and now this one. I don’t think we’ve had power issues like this recently.

Do you live near a hospital? I have seen reported in some news articles that people who live near hospitals have been less affected.

thechao is probably referring to the 2011 outage - dunno if that was multi-day though

Your personal value for reliability isn’t relevant. The electric grid is providing power for hospitals with back up generators and office building which don’t. The value of each of those 9’s to society and the overall economy is what matters and business being forced to shut down lose vastly more money than it would cost to prevent this issue.

The market based solution is to add a very large fine for blackouts on a per customer bases. Grid operators can still select how many actual 9’s they want, but the externalities for doing so are priced in.

I'm not sure that works because the operator can just go bankrupt in that situation?

They are relevant. The electrical utilities could load-shed based on priority, so that places that need highly reliable power pay more and get special guarantees. Technically infeasible on a house-by-house basis, but they could do it for hospitals and such. They probably already do.

Your solution will cause power prices to go up. I'm sure a lot of consumers would be happy with that trade, but why should the ones who are unhappy with it be dragged along for the ride? There is no need to force the grid to provide reliable power to people who won't pay for it.

> Free markets often fail to adopt low cost protections to tail events.

Especially when they can foist off the costs of black swan events onto their customers, investors, insurance, and/or the government. In the mean time, they will optimize for profit in the average case.

I can't think of a reliable mitigation against "heads I win, tails you lose" such as seen in a privately owned rent-seeking utility, but clearly the existing system is going to fail more and more.

There are certainly valid criticisms of government operated public utilities, but in places I've lived under their umbrellas there weren't any life-threatening fiascos due to under-investment in expert-recommended tail risk mitigations. Maybe the operating cost of power / water was more than elsewhere, but I didn't notice (or have the ability to shop around).

A market approach may be to let utility customers buy service with a distribution of costs including claw-backs if there is a utility failure. Such a contract may incentivize utilities to install whole-house UPSs for customers who have a "$500 per hour of downtime" service contract. One can imagine the government mandating that in exchange for the last-mile monopoly and access to customers you must offer a full range of SLAs (ranging from a griddy-like "you pay for wholesale price + vig" to "you pay more but have a fixed cost but may have service interruptions" to "you pay substantially more but get clawbacks if the utility fails to meet delivery SLAs".

Likely over the next decade we'll see a continued 3rd worldization of more public goods such that those who are wealthy can just buy a powerwall and those who aren't will freeze / boil every 8-15 years.

A less markety approach is to operate the utilities as a quasi-government (IE not-profit seeking) organization with clear goals and governance; such things have worked in the past and also failed in the past and lots of fair-weather zealots will complain that it is government overreach.

There is every reason to predict that we will continue to send the cost of failure to the government/taxpayers and send profits (plus what should be normal infrastructure upkeep) of normal operation to whoever is well enough connected to operate the utility.

Based on what I've read, my understanding is that the separate Texas grid and ERCOT exist because Texas politicians wanted (and still want) to avoid federal regulation. The parts of Texas that were not on the separate Texas grid, i.e. El Paso didn't lose power.

Are you saying that the rest of the US does not have a power market with appropriate regulations? Because as far as I can tell, this crises could have been completely avoided if we didn't have our own isolated grid.

As far as I can make out, this narrative is simply not true, at least not with regard to power generation. Texas is subject to the same NERC regulations as the rest of the U.S. and Canada.

https://en.wikipedia.org/wiki/Texas_Reliability_Entity

That they are subject to some regulations doesn't mean they're subject to all.

https://www.nytimes.com/2021/02/21/us/texas-electricity-erco...

> One example of how Texas has gone it alone is its refusal to enforce a “reserve margin” of extra power available above expected demand, unlike all other power systems around North America. With no mandate, there is little incentive to invest in precautions for events, such as a Southern snowstorm, that are rare.

https://en.wikipedia.org/wiki/Deregulation_of_the_Texas_elec...

This makes sense, though I'm still not sure I understand exactly how it works. Thanks!

If the externalities are on the national level, and the fed was seeking to regulate and own the responsibility, whole texas sued to foist them off, who has responsibility for the externalities?

I don't think anybody said the market solves all problems. It's completely possible to have a free market with some regulations, you just say that to participate in the market you must meet certain reliability standards, like being able to withstand low temperatures for a long period of time. Private businesses can count on being bailed out when once in a few decade disasters happen (which makes some sense, imo, since it's not efficient for each business to prepare for disasters), so the only way they will prepare is if they are required to.

Many people do, in fact, insist the free market solves all problems. Many of them are in public office, many of them in Texas.

In practice, there is nothing your typical free-market evangelist likes better than a government-subsidized monopoly, but preferably without any pesky regulations.

I generally agree that externalities are a common cause of problems from free markets, perhaps the most common.

But I would suggest that free markets generally create good outcomes, and when we identify bad outcomes we should correct the rules of the market.

In this case, it probably would make sense for there to be a penalty that providers had to pay if they failed to deliver service after it was committed.

My understanding is that the people who put the market together expected that the huge rate increase which would accompany an unusual storm or other peak demand event was expected to be sufficient incentive for providers to harden their equipment so that it would be able to cash in on those peak rates.

This event showed that the incentive is not enough for events that are extremely rare. It also showed that a lot of the supply infrastructure to the gas plants failed, and hardening that infrastructure is really up to the producers who aren’t directly part of the energy market.

So regulation to increase preparedness standards for gas plants supply infrastructure might make sense. But it may be possible to accomplish the same thing more efficiently via fines and penalties for market participants who go off-line during severe weather.

My point is simply: the market itself isn’t really the problem, it’s creating the right set of incentives and regulations around that market.

One last thought: living in Austin and having lived through this whole ordeal (still living it), I can’t say that I would necessarily want to pay twice as much for energy every single month to avoid one crappy week per decade. I’m thinking that I could probably get myself a backup heat and power source, and do some winterization on my own house that would be more valuable to me than paying twice as much for “the grid” to be upgraded to handle events of this rarity. I admit, we don’t know if these will become significantly more frequent due to changing climate, and that makes the decision more difficult, but I don’t see it as flat out wrong that a significant natural disaster damaged the power system, nor do I think having an “invulnerable power grid” is necessarily a realistic or achievable goal.

I haven’t done the analysis myself, but I highly doubt winterization would double the cost of energy. It’s probably some small portion of the capex to build the plant in the first place. Would you pay an additional 5% or 10%? Probably, especially if you get to avoid your home being flooded by burst pipes, rendering it potentially unlivable for much longer than the week of power outage.

"But I would suggest that free markets generally create good outcomes, and when we identify bad outcomes we should correct the rules of the market."

the big players in a market are the ones who "correct the rules of the market", or in other words regulatory capture. so go ahead and identify bad outcomes and solve the problem with new regulation proposals until you're blue in the face, without power it's just people posting online. as long as actual regulatory power is concentrated in a few people, the regulatory capture is easy, and any discussion about it is a pastime.

Living in Austin myself, I agree completely. A little bit of resiliency is better than large increases in power costs. As it is the mass-migration into Austin is driving costs up for low-income families and fixed-income elderlies here. Driving up their power costs further is another hit they don't need. I'd rather see the city spend some of the property tax windfall they are having helping low income get some resiliency and improving the cities ability to respond to the rare events.

Free markets seem to do a bad job with resiliency. Redundancies, winterisation, warehousing N95 masks just in case, etc is treated as inefficiency by a free market and eliminated.

These aren’t externalities - is there an economics name for this? I’m sure it’s been noticed before.

Free markets do account for huge shortages, but we’ve eliminated the motivation for actors to do something about it with anti price-gouging regulations.

There would be tons of stockpiles if businesses could appropriately upcharge when there was mass panic buying. Instead we stupidly lock prices, which wipes out most of the point of paying extra storage costs compared to a just in time competitor.

The problem is we’ve as a society don’t like when prices go up with demand so we’ve crippled markets in their ability to deal with these types of events.

You'd need to make a strong argument to convince me that situation would actually be much better than the one we have currently. At least with the current price-gouging laws, we're able to semi-select for the people who are willing to spend the most time pursuing goods, which has a much better correlation with actual need than who's willing to spend the most money.

Furthermore, I'm not entirely convinced that removing price gouging laws would really prevent these sort of events. Capitalism has a tendency to optimize for the short term, and to select for firms that are perfectly optimized for the current environment, however unrepresentative that may be. Sure, counterculture firms exist, but there's rarely enough of them to make a big dent, and the markets often remain irrational longer than they can remain solvent.

The problem is that markets, controlled or free, are at a level of complexity far beyond our capacity to effectivel control from a top down perspective, there have been countless examples with top down legislation have had unforeseen consequences (or overtly controlled markets in Soviet Union). The idea with free market capitalism is that as much as possible the transactions/pricing between parties should allow for systemic optimisation.

There are two main problems, a) not all transactions are captured appropriately, b) we fuck with the system from a top down perspective so much as to completely undermine the premise of the low level optimisation. Oftentimes it's in relation to (a) that we do (b), in (b) we should be aiming to provide recompense to the side of the hidden transaction via the government; unfortunately the government oversteps here and tries to optimise the system themselves.

There seem to be two cases. Rare, but on human scales regularly reoccurring disasters (hurricanes in the east, fires in the west, etc ...) and more-rare "black-swan" events, basically 100+ frequencies.

The more regular disasters seem to be dealt with in markets. Insurance companies will have requirements to reimbursement, some companies will stockpile goods, multi-region firms will have DR plans etc. Alternatively governments regulations can require preventative measures and have stockpiles in place. In practice both happen to varying degrees of success. Price gouging allows for more inter-region support on a for-profit basis, but probably also disincents some charitable actions. On net we probably should allow "entrepreneurial gouging" (like a guy buying a water truck in a non-disaster area, driving it to a disaster area and charging high prices) It might be distasteful, but if it is providing net-benefit of otherwise unattainable resources that is still better. But companies organizing and planning around this, seems much more sinister. Like the water company charging more for potable water, rather than spending their time and effort fixing the system for everyone. That type of thing is probably better prohibited and regulated.

But the more rare events, nobody seems to handle well. Governments and private enterprises alike tend to only prepare for the last crisis. I'm not sure there is a government vs private comparison, so much as a human failure. I suspect good government regulation is going to be better, because it can sustain longer-term initiatives, but getting the policy to be "good" and not corrupted by grift is difficult, especially with out the feedback loop of reality.

There is no way it would be effecient to perpetually store 1000X your regular demand just to upsell in an unprecendented crisis, regardless of price gouging policies.

This is super predictable behavior in storm areas like Texas. Hurricanes come, everyone goes for gas, places start to run out of gas, nobody can raise rates, they run out of gas. Even a double normal price could easily pay off a much larger tank.

This is true, but it would be just as true whether you did it in a free market or through some government program because the inefficiency comes from having to make and store that stuff. A free market just gives us more ways to quantify that inefficiency.

I don’t think what you’re saying jives with the reality of what happened in Texas. The price of electric and natural gas skyrocketed during the shortage. A company that winterized their equipment stood to make a killing. And yet, despite the massive profit incentives, large swaths of Texas were blacked out due to lack of preparation. Besides that, the economic damage of the outages far outweigh how much winterization would have cost. The power companies are unlikely to bear the burden of that. All together, that’s not a good recipe for a pure free market system.

People also find the idea that only the rich can afford to have heat so that the pipes in their walls don’t burst distasteful.

>The price of electric and natural gas skyrocketed during the shortage.

AFAIK the price was capped.

https://news.ycombinator.com/item?id=26160546

Wholesale prices are capped. They contribute to but don't limit the price paid by retail customers.

For context, the cap is nearly 100x the normal price. In colder states power companies manage to be profitable at around the normal price of electric in Texas. The price of winterization isn’t that high, and a company that did it could have made a year’s worth of profit last week. The cap was very high and did not contribute to the outage.

In your other post you say this is a once in a century event. It really isn’t. This is the 3rd time since 1989 that Texas has had outages due to insufficient winter prep. It’s more like a once in 10 years event.

there is also another externality, due to perception (regardless of whether it is true) that it affects non rich people more means that it build distrust. trust (and in particular systemic trust, e.g. rule of law) reduces transaction costs; i.e. you can rely on others to do their job properly without having to continually monitor them and build in unnecessary contingencies.

This is a fringe event. Large scale grid storage is at least 10 years away. It's not a market failure. The irony is that all the complaints about grid storage will disappear once we have it because running a grid without grid storage will be considered insanity by that point.

The technology isn't there. The real solution would have been to cut demand to non essential uses.

I don’t see how that is a technology problem. The technology to run power grids that are resilient to cold temperatures exists, otherwise Canada, Alaska or any northern country wouldn’t have power. It’s just expensive and any player investing into cold-proofing their grid would have been priced out of the market in Texas long ago.

Yes. And then the incentive will be to buy up all available capacity and create artifical scarcity. If you believe that we are dealing with profit maximizing actors then this is the likely outcome in the absence of any price control. Lets say you believe that a competitor may come up to provide the service as soon as the price becomes attractive. Power generation is not something that can be stood up (at least not yet) at the drop of a hat. There are far better solutions to this problem and they all involve more and not less regulation.

> And then the incentive will be to buy up all available capacity and create artifical scarcity

This makes no sense. Which market is it you think can be cornered here? Someone buying up all of the natural gas in the world? All of the generators? What?

The thing about market prices is any time someone tries to corner a market and drive up prices, it incentivizes more sellers to enter the market.

If you’re talking about small periods of time, like cornering it for a week, that’s not really easy considering futures traded against this market would smooth the price like they do for every other commodity.

Same thing applies to oil, oranges, natural gas itself, coal, refined copper, etc.

There is nothing special about electricity other than the fact that people feel entitled to have it at all times and cheaply despite the fact that generation costs vary widely.

> There are far better solutions to this problem and they all involve more and not less regulation.

The California power system is a bastion of regulated electricity and it’s an unmitigated disaster. The prices and reliability are worse than the one in Texas. You just don’t notice because the fallout (other than burning down half of the state) doesn’t freeze people to death.

Vertically integrated government owned and operated utilities have imploded too- notably Ontario hydro failed due to costs related to their nuclear plants, and the rate payers had an extra line item on their bill - $10/mo? - for I don’t know how many decades to pay down the stranded debt.

I don’t think markets are the right tool for electricity generation transmission and distribution - too much coordination is required between the different functions, central planning optimizes the outcomes, and the lead times are so long on significant projects that a high $/MWh for a week in 2021 isn’t going to result in a new nuclear plant.

There are natural monopolies in distribution and maybe transmission as well.

100% regulated even if there are private players is the way.

The Chinese model is pretty spot on. Just let the government own voting shares in private companies. The question isn't whether public vs private is better, it's about how to harness the best parts of both. You want the government to step in and prevent unfair and counter productive behavior and you want the private company to run the infrastructure as efficiently as it can.

This would have gone a lot better if you'd said "the French model", or even the "British pre-1970s model". China is a country without a free press; it may appear effective but it's also highly corrupt.

In a truly free market, Texas would have simply received power production from nearby states that were connected into their grid.

This isn't a free-market situation because the state has gone to great lengths to keep this power system isolated.

The other states that were struggling with blackouts? No. Being on the same grid wouldn’t have solved this, as far as I can tell.

Externalities are, in my opinion, one of the primary reasons for market failure.

This. Amongst other reasons (market distortions, gaming, profit incentives which misalign to the actual role of the markets target service)

The demand pricing thing, is entirely sensible inside the construct of market for energy. But, the system itself demands both load shedding and load returning and has transmission system dependencies. Price driven load shedding can (if understand it correctly) drive more things off-line when you need to keep the, online.

It's cascading failure stuff. But, not everyone's bidding model is tuned to optimise the overall goal of delivering power.

My conclusion of modern "free markets", as you crank the iterations of the, uh, market simulation, will result in cartel or monopoly. That's what practically all markets have collapsed into if there is any degree of entry cost due to technology or infrastructure.

Also, I've noticed a HUGE drop in that term from the various GOP politicians that used to parrot that daily back in the 80s/90s). That leads me to consider their monetary overlords (I'm not suggesting the other side doesn't have them, just that the GOP is a bit more eager to directly funnel their sponsor messages) DON'T want free markets, because they all have won free markets.

Which is really a bad state of affairs.

I've also realized that money in politics is way up in the same period, and regulation to allow more money into politics has risen at the same time. Monopolies get more ROI on lobbying to help control their monopoly market positions, and keep antitrust at bay.

It's tough because free markets result in predatory pricing in times like this. Heavy regulation (especially these days, see above with money in politics) is highly resistant to change and not necessarily reflective of the common good or even "reality" or even physical laws.

Anyway, Texas is modern america: a rapidly failing government, and a rapidly failing economic system, all being undermined by steady shadowy corruption that doesn't become apparent until a sudden external stressor appears.

> Wouldn't smart businesses have done the risk analysis and done the work?

That is complicated. How do you validate costly infrastructure changes that are only testable or necessary once about every 10 to 20 years? Proving winterization for Canada or Wisconsin weather is challenging if you rarely have Canadian weather extremes.

I suspect the parent is fully cognisant of this. They are illuminating one of the fallacies of the free market extremists.

Free market vs regulation are irrelevant to my comment. Neither make rare conditions required for testing less rare.

Apologies - I misunderstood the intent of what you wrote. For testing would it not be possible to use modelling instead?

E.g. in the way that you might fire drill recovery from a nuclear reactor instability without first melting part of the core.

Or how you might test a martian lunar rover before sending it off to Mars.

Even if modelling the entire system wasn’t possible with sufficient fidelity, why couldn’t you replicate the system in a geographic environment with the requisite conditions?

> If the market solves all problems

Straw man.

> Maybe the problem here is that loss of electrical power has a significant external cost that isn't absorbed by the utilities.

The utilities don't have to pay for the deaths caused by not providing power, or charging so much they might as well not be providing it.

"Free" markets don't solve these problems. But it's worth noting that regulated markets are still markets.

Socialize the costs, privatize the profits! The capitalist mantra!

My personal definition of a libertarian has been: someone who read a capitalism economics textbook but skipped the chapter on externalities.

> If the market solves all problems why would the government even need to mandate winterization?

But Texas has a government and it _hadn't_ mandated winterization. So it doesn't seem to have done better than the market in practice.

The party controlling Texas’s government openly opposes the use of government to solve issues like this.

Reply on news.ycombinator.com