Unless you have a strong familiarity with ISO/FTR markets this article will be pretty opaque.
The short version is that electricity delivery forecast is managed by a financial optimization process. Market participants (generators, line owners, and consumers) basically buy and sell their electricity in advance. Sometimes as far ahead as 2 years, all the way up to the day before (at which time the real time market takes over IIRC).
So in addition to the primary stakeholders you have financial participants that buy/sell electricity without having access to any apparatus for the generation/transmission/consumption. These participants sometimes have very manual processes for buying their contracts.
My guess would be some poor fresh grad was under supervised and transposed the source and sink address. Instead of buying at 203.40 they sold at 203.40 by reversing the source and sink.
Without trying to fall too hard into a potential late-stage-capitalism hole, what is the advantages on trading like this? Beyond the advantages of making money on futures.
I guess the buy/sell in advance lets generators etc fund projects ahead of time (maint, expansion, etc) instead of always living hand to mouth on "real time sold" and without having to land "huge multi-year contracts", so the market can be a bit more fluidly priced? (Maybe too fluidly in this case!)
Yes, that's fundamentally it. By making the marking more liquid, different types of entities can adjust for their specific predictability/risk needs.
Also note that though these markets are "free" they are still extremely highly regulated. I don't know much about the Finnish market but I've built trading algorithms for CAISO and PJM, and it is by no means a free-for-all. The rules about what types of entities can participate in which aspects of the market are quite strict and detailed.
Especially post-ENRON there is a strong focus on avoiding even any possibility of appearance of market manipulation. Basically the guidance from our lawyers was: every trade needs to have a strong, explicable economic rationale (apart from the market.) In our case we didn't even want to use neural-network based trading algorithms, since they (at the time) weren't sufficiently explicable for us to have a "rationale" for a trade that might later be called into question.
The market is mainly regulated by having some restrictions on pricing by requiring the sellers to be able to prove that their price roughly matches your their costs + some profit margin (to stop a big player from buying all the cheap electricity and relisting it for massive profits) and forcing producers to put money into escrow/buy insurance for the case that they fail to deliver what they sold.
This latter one led the government having to open a line of credit for tens of billions euros last winter when the prices were high just so the actual producers could afford to sell their (historically highly priced) electricity as the money you have to leave in escrow/insurance price is tied to the SPOT price.
The idea is to buffer changes in price. Started with crops originally iirc. Say I know I need 1 ton of seed for planting. It normally costs $5,000/ton. I am concerned that the rat epidemic might cause a seed shortage. It does. 1 ton of seed next year is worth $20,000. Because I purchased the future of 1 ton of seed for $5,000, I am not affected by this.
Made up numbers obviously, as I know someone is just waiting to “well akshually” me.
Question: how does that rat case actually work out followed through slightly more? If enough players bought at $5k/ton and the seed shortage increases prices due to scarcity, what happens when there's only enough seeds to meet 3/4ths of demand?
1) If the contract is physically settled vs financially settled.
2) The existence of a clearinghouse - which is a middleman in charge of making sure both parties deliver on their contractual obligations. Virtually all futures contracts go through a clearinghouse.
Lets assume that it is physically settled with no clearinghouse - what happens is the same as any other contract where the other party does not deliver on the terms of the contract (also known as FTD - failure to deliver).
In theory price v demand. Prices would increase until demand was reduced to the point that there was enough seed for everyone willing to pay the price.
It also depends whether you’re trading future that settle for cash (you buy for 5k and they give you 20k cold hard cash) or whether they’re actually going to send a semi truck to your farm, in which case they should have been holding enough collateral for such an incident.
Realistically, I believe most futures are cash settled and you’d probably be better off pocketing the extra 15k and spending 5k on planting a different crop this year.
Because there is effectively no way to store energy at scale, prices are extremely volatile, like from $1 to $1000 in five minutes.
If you want to buy a solar panel with a useful life of 10 years there’s basically no way to estimate how much revenue it will make. But if I promise to pay you $10 for that energy all year, you have some idea if your purchase makes sense.
> Without trying to fall too hard into a potential late-stage-capitalism hole, what is the advantages on trading like this? Beyond the advantages of making money on futures
Trying to view things as “late stage capitalism” isn’t a good way to really understand anything. It’s a cynical Reddit trope that assumes malice from the start, which doesn’t leave much room for a real explanation.
The reality is that trading like this is a type of contract and it takes two parties to enter into the contract voluntarily. Nobody is forced to trade energy contracts this way, but it can be helpful for both parties to do things like lock in more predictable pricing, hedge certain types of risks, and agree to mutual contracts that benefit both parties’ businesses.
Energy prices going negative sounds like something evil or malicious or “late stage capitalism” to the uninitiated, but even without trading errors there are legitimate reasons to have negative energy prices at certain times of day. For example, certain types of power plants like nuclear can’t quickly ramp up or down as demand changes. A power company might come out ahead by pricing power negative (that is, paying people to consume it) for certain night time hours if it allows them to keep the power plant at a higher, continuous output in preparation for peak times. Energy hungry industries such as aluminum manufacturing or other industrial processes might gladly adjust their production schedules to operate during these times to take advantage of the low or negative pricing. Everybody wins.
There are numerous dynamics like this at play. It’s not a “late stage capitalism” thing, despite how some people like to sneer at anything they don’t understand. There are legitimate reasons to operate this as a market and let market participants work out optimal deals.
The other main example of negative price offers into the market is renewables which receive some types of subsidy. For example, if you have a solar farm and receive $40/MWh subsidy, you might bid into the market at any price above -$40/MWh.
If your output regularly drops to low / negative levels - invest in storage and sell when the price is higher. If it occurs grid wide - interconnect to neighbouring grids.
I'd like to add a bit to your comment. I certainly agree that negative prices in this context don't mean anything remotely nefarious, but part of your comment suggests that "it's all consensual contracts" is a sufficient ethical justification for a system — which may or may not be what you actually think — and I wanted to touch on that.
Markets come in all shapes and sizes, and in reality you can't write up any contract you like — nor should you be able to. Indeed, electricity markets must be tightly regulated due to shared infrastructure, which is an example of the fact that these markets don't arise simply from the application of a fundamental "allow people to enter into whatever agreements they like" principle, but rather requires pragmatic _design_.
Ultimately, we, as collectives, decide what the rules of the game are. In the case of electricity, the need for innovation and flexibility to decarbonise the grid are perfectly acceptable justifications for adopting market solutions (the only real alternative being to allow the State to monopolise the whole system, which would come with its own serious and disruptive trade-offs); but ultimately, this is a _system_ that we decide on, and it doesn't arise out of natural rights, and systems can be more or less well-designed. I think you touch on this in your comment as well.
Markets aren't evil, but nor are they naturally good. They are a tool in our toolbox for constructing resource allocation systems. And the success of a system, and indeed _the_ system as a whole, must consider all the interlocking economic and social consequences — and it's anxiety about those that the "late-stage capitalism" trope really reflects. It's not something to dismiss altogether.
Physical commodities cannot be delivered instantaneously, additionally there is no giant warehouse at the commodity exchange that stores the commodities. This means that what you are going to trade is a delivery contract.
If you play a game like Prosperous Universe, there is no delivery problem on the central commodity exchanges and the commodity exchange has infinite storage capacity. When you buy something, you instantly receive the commodity.
If you want to know why there are duration differences on the contract beyond the delivery itself, e.g. why someone would issue the contract 3 months or 6 months before delivery, then the reason is that they want to minimize income fluctuations or to pay off specific contracts. Similar to how businesses want to turn capex into opex at exactly the rate the money comes in, but this time the goal is for the income to arrive at the same time as the expenses.
Electricity market is interesting from a late stage capitalism perspective because it's so managed by regulations and transparent, and relatively pretty closed-world, and there's fast feedback. So many of the capitalism critiques don't work as well against it.
Of course the fundamental problems of unfairness and poor people paying more for electricity than big industrial users can also be attributed to properties of how the market is managed and regulated by the public (the classic "critique of free market" -> "not free enough, ergo not fault of free markets" trope).
Besides generation, this also enables pricing and funding of energy storage. Eg you could use historical data to see how profitably you could play production peaks and lows to charge/discharge battery storage, and use fairly well known risk management stats for the uncertainty factors.
Like all market activity, the high-level point of all that activity is to derive the real price of the thing being traded. Without falling too far into the late stage central planning thing, this is the mechanism that has been absolutely broken (due to been poorly done by pathetic humans) in all previous attempts at communism.
As a tangent, it would be interesting to see if non-pathetic modern computer systems could actually achieve to implement that mechanism successfully. But if sci-fi is to be believed, that could easily fall into another form of dystopia.
> As a tangent, it would be interesting to see if non-pathetic modern computer systems could actually achieve to implement that mechanism successfully.
> Fingrid has closed intraday trade in directions FI>SE1 and FI>SE3 to ensure system security and balancing capacities concerning the non-matching situation in the Finland bidding zone after Day-ahead trades. Update: Intraday trade in direction FI>EE closed
> Fingrid is prepared to initiate Intraday purchases, if necessary, to ensure system security and balancing capacities concerning the non-matching situation in the Finland bidding zone after Day-ahead trades.
It's only -€0.2/kWh, not that bad at all. I've been seeing spot price predicted to be -A$1/kWh in several states in Australia (which is the lowest possible bid allowed by regulation). Even in Europe, I recall seeing some post desperately trying to figure out how to curtail their solar export because the price was reaching -€0.34/kWh. Not sure an error of -€0.2/kWh is anything newsworthy TBH.
That's the average for the whole day, the max negative price is -€0.5/kWh which is limited by regulations. It starts from 15.00 and continues for the rest of the day. Other than those hours, the price is normal, so the average is not as drastic.
Depends what the market rate was at the time and how the exchange handled the error.
People unfamiliar with markets are often surprised how often there are serious issues. It’s not uncommon for trades from a specific time period to be invalidated.
Are there really large solar PV installations that can’t curtail, continuously, all the way to zero? An MPPT ought to be able to operate arbitrarily far away from the maximum power point.
Or is there some other limitation that prevents this?
I think it's mainly because of large amount of rooftop solars which are either unable or don't have incentive to curtail, and then there are coal generators already running at their lowest output level.
In Australia, we’re moving towards dynamic operating envelopes for solar and other “distributed energy resources”. That potentially includes EVs and ACs.
The idea is the solar inverter is sent a dynamic stream of operating constraints by the distribution business based on network conditions.
It’s already being piloted in places like South Australia, which has very high solar penetration. Regulations are also moving towards requiring smart meters at all households (already the case in Victoria) and all new inverters be network connected.
The regulation applies in the opposite side as well. I think the highest bid allowed is like A$16.8/kWh or something like that.
It's not fully clear to me how the market work as a whole. I learnt that generators may have swap contracts with retailers and big users, which, based on my understanding, can distort the price to arbitrarily low, because generators get paid for the difference from the other side regardless of spot price, so they can set the lowest possible to ensure they can dump their power to the grid.
Also when it's predicted to be this low (or that high), there is likely some operations (market or not) behind the scenes to level the price. Usually when it gets to the time, the actual spot price has been in a reasonable range.
On the stock exchanges in the USA, they do that because we didn't use to and trading bots that were mis-configured went nuts and broke the markets for a while.
Now they stop trading for (15?) minutes to let everyone make sure it's real and not their trading algorithms being stupid.
It's not about not letting the stocks crash to far, it's about making sure the pricing is accurate.
The Federal Reserve however is a different story. :)
If you are guaranteed by law to buy it for less than you can sell it, you can price it mighty low to attract customers and still make money.
It's funny, your comment reads like a business has some kind of right to make a profit, rather than letting the free market figure out if it should exist or not.
> If you are guaranteed by law to buy it for less than you can sell it, you can price it mighty low to attract customers and still make money.
That's not a price floor, it would be a price ceiling on the second level supplier, and they could just choose not to sell. There's no way this scenario could happen except for subsidies and mandated production like the Defense Production Act.
> It's funny, your comment reads like a business has some kind of right to make a profit, rather than letting the free market figure out if it should exist or not.
This is a very terse note. Is there simply not more information available than "some price went negative on the market because someone made a mistake", or am I missing something that's implied here?
The market price is based off the cost of the marginal MW for some of these auctions. They tried selling power, but someone accidentally said they'd sell for a very large negative euro per MW hour amount and that set the price for the entire market and obligated them to pay the market to sell power. This is pretty rare, but not unheard of.
The seller (Norwegian Kinect Energy) will need to negotiate with other major marker participants to bail them out. But it’s likely in the interest of everyone of not to let them collapse.
The markets would be in disarray if there would be undo button for a trade. Preparations need to be made by large industry participants for every daily allocation.
It's Scando, so i would imagine that if the relevant ministers in FI SE NO all sign off, and no major market players (like Fortum) object, then they'll unwind it and fall back to replaying the previous day's bids. Or is this impossible ?
It's impossible to undo because a lot of bids are build on the top of other bids. You would need to undo tons of history. A lot of legit players and traders would lose money. They would sue for market manipulation.
Let one who makes mistakes to pay for them.
However this does not prevent doing other trades on the top this trade to other direction. It is the question who catches the falling dagger and pays for the mistake of others.
This is pretty clearly NordPool's fuckup. There's no way in hell this bid should have gone thru without human review. Seems clear that their UI is to some degree - in some critical spot - fragile, user-hostile, ill-designed.
You can say that in analogous cases, finance firms have been put out of business by bad bids caused by slipped fingers. Granted. But in this case you're throwing a wrench into a power grid, with the potential to cause a system disruption, in winter no less.
There are many market participants, some of whom start trading on this information immediately. The Finnish grid is connected to the rest of Europe so a market participant could in principle flow this cheap energy from Finland all the way to Spain if the cables had capacity. I would guess it was too late the moment it happened.
Sometimes it's a mistake, sometimes there really is a pile of power that has been contracted for, and is coming, and enough sinks that needed power which don't need any more power. There are power storage plants here and there - hydroelectric aside from small batteries. No doubt if one were on the path to this it might sink some of this power but they might be full or the grid in between the two spots might be full or they might already have their own contract in place sinking some nuclear power or who knows.
If it were on purpose I'd still expect the bid to not be so extreme. For example, maybe -5€/MWh and not the -250€ or whatever the number was. They could have still secured the sale without losing so much.
They prob didn't notice until the market closed. Even then, you can't just undo a bid like that unless they have some kind of emergency clause to fix mistakes like this.
I think it's all there. The company in questions bid provides ~5800 MW for 24h, paying 203€ to those that use that power. Other companies will have accepted that bid (automatically) and will now tomorrow expect that power to be delivered (and get money for it), while other providers are planning to do something else with their power. The company mentioned may or may not have that much power available.
> "We are working with other market parties to solve this extreme situation," it said.
Sounds like they are working on an agreement with the others to avoid their mistake turning out to be as costly as that.
It’s probably in the interest of all of the participants that this is resolved with a less costly outcome. After all, who knows which of them will be the one making a similar mistake next time. It’ll be good for them then, if they played nicely now as that can inspire how future happenings will play out.
this is similar to how knight capital lost 440 million us dollars one afternoon in a trading bot bug and had to get merged with one of their competitors. this isn't quite that big
Well, electricity consumption in Finland will very likely spike tomorrow. People with a certain type of contract for their electricity will actually earn money for using a lot of electricity. Don't know what the implications for the network are, but I don't think such unnatural spikes in consumption are good for it.
My current transfer fee is 0,03€/kWh. There is also the electricity tax (0,028€/kWh) and energy sales commission (0,004€/kWh). With hourly energy rate of -0,50€/kWh any usage will lower my energy bill for the current month even considering all the above costs.
Ah thanks. The rule of thumb "you don't earn" applies to "ordinary" negative energy prices caused by favorable weather. Obviously here the price was so extreme that old wisdoms no longer apply.
Myself I enjoy the comfort of a fixed rate contract. No efforts in trying to optimize consumption profile over days and weeks.
At first I thought your math was wrong, because that doesn't match the numbers in the article. But then I checked the day-ahead prices at epexspot[1], and it's indeed showing -500 euro/MWh in the afternoon, from 14h00 to midnight!
It is an auction and everyone pays the same price. During the afternoon tomorrow every household in Finland will be paid significantly for their power usage unless they have a fixed price agreement, which is not that common
According to Finnish Energy Authority (energiavirasto.fi), at the end of 2022 only 13,7% of customers had a contract where pricing is tied to the hourly market rate. It is much more common to have either a fixed price contract (usually for a term of 12 or 24 months) or a contract where the pricing is set by for example monthly averages of the market price.
I'd suspect that actually fixed-price agreements are very common, and represent the overwhelming majority of contracts for residents/individuals.
I know that after the war started spot-prices rose sharply and a lot of people made the jump. Some, unfortunately, locking themselves into relatively expensive prices.
I understand put options and short selling, but how on earth can the price of a commodity sink below zero? How can you make a „sell“ order that is negative? Can someone explain please?
Happens frequently in CAISO midday during shoulder months (https://www.caiso.com/PriceMap/Pages/default.aspx). Renewable generation produces a lot (sunny, decent wind) and electricity usage is so low, that supply exceeds demand. Electricity has to go somewhere to keep the grid balanced, so the utilities will wholesale electricity at negative rates.
How would you dump it at this large scale? The point of “selling” it at negative price is to encourage consumers to connect loads on the grid in order to burn this extra energy to maintain grid balance.
As a Finn, probably just send out a message on some social media platform that people need to start turning on their saunas for the stability of the network. (Alternatively, like fingrid did, send a notification to my phone saying to "consume energy normally, as the pricing issue does not reflect energy availability")
Even a fairly small electric sauna is ~10 kW, so that's be just 10000 homes solving that problem.
Very few people get paid during this market anomaly. Most people have fixed rate contracts, and the few that do have variable rates, even fewer have ones where the negative price would be passed on to you.
Transmission cost and taxes will still be present, so this is more of a "hey it's going to be a societally good thing to go to the sauna today instead of tomorrow, consider it!" (Also, the saunas are definitely in use, don't worry, nobody's going to pass up the opportunity to say they went to the sauna for a good cause.)
Or, via lower price - negative if necessary - you let someone else go to the trouble, via facilities that they themselves invested in. Also in systems like this, there is not necessarily anything special about "zero" or "negative". It's just "less than the previous number" or "more incentive this way than the previous number".
The idea that negative prices are something weird is pretty silly.
A negative price indicates that the person sellers disposal technique is worse than the buyers disposal technique. In other words, the market is working as intended.
Nobody is complaining about the cost of garbage disposal where people are paying for having less (!) stuff.
There get to be points in some commodities where providers temporarily do want to pay for people to take their product. Electric power is one where supply is not particularly elastic and does need to be disposed of when there is too much for grid stability.
This has happened a couple of times with oil pipelines as well.
Markets tend to figure these things out quickly though.
Commodities contracts that settle in the actual commodity and not in cash are actual obligations on both sides to deliver and take delivery. At some point in order to get rid of the obligation to take delivery at some point people are willing to pay to get rid of their contracts. Sometimes this happens when speculators make crazy bets during instability.
Happened with crude oil in 2020. If you’re still going to produce it but have no where to store it you need to pay people to receive a shipment of 10,000 barrels.
its like if you booked a prostitute to come over on Friday, on Friday you found out your partner was coming over and you need the full service sex worker to go somewhere else but have nowhere to send them, so you attempt to pay anyone to take the booking and everyone else is in the same predicament and eventually someone’s going to take the loss of having the prostitute around their partner and mess up their social situation but will be paid handsomely for it
you’re right, sex workers take deposits for this specific reasons and have autonomy and agency, it’s weird and outdated to deny [probably] women that
thanks for pointing that out it’s a big plot hole and very much unlike the energy market. although maybe deposits would be an improvement to the energy market
no, because it's not your typical cost of carry problem. Oil barrels will not spontaneously combust if you leave them at the plant, and you can control the rate at which you produce them. And the problem is contained at the point at which a seller is left holding the goods.
The electric grid will always distribute the load _somehow_, but the way that happens if there's not enough buyers will have far reaching consequences that are uniuqe to this utility. AND you do not have full control of the production of electricity, which means you don't have the means to react ro a surge unless you have enough sinks distributed _across_ the network.
Nowadays it is mostly renewable subsidies without a clause for negative price, marginal cost is 0, if they are paid at a fixed price they will produce and dump it on the market, in Nederlands they big glasshouse turn on the light when the sun is at its peak and earn money. And there is also inflexible baseload, that can't cycle up and down at the rate of changing demand or renewable ramp.
This is in fact a very good feature. When power is cheap or negative it's because there is more generation than load in some segment of the grid. To fix this you can tell people to stop generating, you can tell people to start loading, or you can just drop the price and let both happen on their own.
There are targetted ways this is done when the difference is so big it threatens grid stability, but its often better to let the market handle the gross imbalance on longer timescales when grid stability is not threatened.
There are some things markets are very good at. This is one of them.
> There are some things markets are very good at. This is one of them.
Ironic in a thread about the market making up a totally imaginary generation surplus.
10-20% of Finnish retail customers are on spot market based contract, and stand to save up to something on the order of 0.40€/kWh consumed (net after taxes and fees) off their electrical bill for the reminder of the day. I suppose every market-rational consumer would have a 10kW electric space heater outside melting snow on the ground, maxing out their supply.
It's actually particularly important that prices be able to go negative in an electricity market, because it's physically impossible to store electricity: it must be generated and consumed at the same moment. "Electricity storage" is shorthand for "conversion to and from a different kind of energy", which from the perspective of the grid is just a kind of consumption or generation. And if there is too much unused power on the grid, it will damage it.
No the opposite is the case. How would you expect a zero minimum price to be enforced? All waste products would have to be disposed by the government for free and the thing is that this costs money. Negative prices signal a desire to get rid of something.
Most Finnish households have a contract which limits the effect of this pricing anomaly (fixed price contracts are quite popular as people see the hourly market rate as risky and unpredictable). However, I do have a hourly market rate based contract myself and tomorrow afternoon/evening with rates of -0,50€/kWh any electricity consumption is going to lower my energy bill - even considering the energy sale commission, taxes and energy transfer fees which I will of course have to pay.
There are two fixed fees households pay, an energy tax of 0,02253€/kWh and a security of supply fee (0,013 euro cents per kWh). These are not affected by fluctuations in the market prices of electricity.
https://www.fingrid.fi/sahkomarkkinat/sahkojarjestelman-tila... shows a roughly +1GW (~12GW -> 13GW) rise in consumed power during the first two hours after the price hitting the negative pricing floor (-500€/MWh) at 15:00. Seems like there's some consumer elasticity there, but not enough to make the grid fall over.
This is after the national grid operator planned to intervene as necessary and imposed some restrictions on trading [1], but those seem to have been lifted and the intra-day market is working (at 10× normal volumes).
> Fingrid is prepared to initiate Intraday purchases, if necessary, to ensure system security and balancing capacities concerning the non-matching situation in the Finland bidding zone after Day-ahead trades.
> Update 24.11.2023 12.:12 Market situation has normalized due to Intra day trading. Fingrid resumes normal operations. Fingrid asks BRP to manage their balance normally.
> Fingrid has closed intraday trade in directions FI>SE1 and FI>SE3 to ensure system security and balancing capacities concerning the non-matching situation in the Finland bidding zone after Day-ahead trades.
> Update: Intraday trade in direction FI>EE closed
> Update 24.11.2023 12:25 Intraday trade is open to all directions
As a nuance, the grid falling over would not yet result from too much demand, absent other snafus. Rather you have rolling blackouts when loads get disconnected by the grid operator, or if there are failures in implementing those in a orderly way, less orderly blackouts for parts of the grid. These kinds of managed blackouts are common in some parts of the world, eg South Africa, the recent Texas situation, etc.
The frequency of the system is a clear signal about the demand/load match, and if it starts to drop more below some threshold of deviation, loads get disconnected from grid branches.
Of course there can be "interesting times" leading to cascading processes in the multiple interacting automatic failsafes, you can read up on post mortems of grid failures about what kinds of things can go wrong, eg https://en.wikipedia.org/wiki/Northeast_blackout_of_2003#Seq... .. which is why really want to preemptively do the controlled rolling blackouts instead.
>Seems like there's some consumer elasticity there, but not enough to make the grid fall over.
This has been widely reported as a market error, so many consumers might understand that the cheap electricity is not because there's a massive amount of electricity up for grabs but rather because some Norwegian company fucked up.
Perhaps if this was a real situation where there is actually a huge energy surplus that more consumers would actually "waste" electricity.
All trades are solved at the same time for the next 24h day in a single algorithm run for all Europe at 12h and taking about 15min to solve. You notice at the end and if it converge they won't re-run it for you, it's done.
… Wait, how do you think modern markets _work_? Hint: it is not an elderly man writing numbers on a ledger with a quill pen. There is not generally a human in the loop.
Other regulated commodity markets have what is called a «reasonability limit» avoid this exact kind of scenario. It is very concerning that the exchange didnt have any automated alerts that could trigger for example a phone call to verify that the order is correct.
The limits are there. -500€/MWh and 4000€/MWh. It did hit the lower limit for 10 hours in the row from 14:00 till midnight. Before 14:00 they were not the last bidder, so prices are higher there. Price on the market for everyone is set by the last bidder who fits in the predicted consumption pool.
They have price limits, but I think GP means volume alert limits. Someone unexpectedly bids a third of the Finnish production capacity, you call them and ask if they are sure.
The short version is that electricity delivery forecast is managed by a financial optimization process. Market participants (generators, line owners, and consumers) basically buy and sell their electricity in advance. Sometimes as far ahead as 2 years, all the way up to the day before (at which time the real time market takes over IIRC).
So in addition to the primary stakeholders you have financial participants that buy/sell electricity without having access to any apparatus for the generation/transmission/consumption. These participants sometimes have very manual processes for buying their contracts.
My guess would be some poor fresh grad was under supervised and transposed the source and sink address. Instead of buying at 203.40 they sold at 203.40 by reversing the source and sink.