From the time the three big state-owned power generators were partly privatised in 2013, the possible closure of the Tiwai Point aluminium smelter was highlighted as a risk.
The biggest, Meridian, which supplies Tiwai, said closure of the Rio Tinto-controlled New Zealand Aluminium Smelters (NZAS) may result in a reduction in wholesale electricity prices and in electricity prices generally, and therefore a reduction in Meridian's revenue.
"In some circumstances, the impact on Meridian is likely to be severe," Meridian said then.
• Gavin Evans: Smelter review a wake-up call on energy policy
• Premium - Tiwai Point smelter says it needs 'tens of millions' in annual relief from electricity and transmission costs
• Premium - NZ sharemarket blindsided by Rio Tinto's review of Tiwai Point smelter
• Owners of Tiwai Pt aluminium smelter seek talks with the Government, raising possibility of closure
The two others, Mercury and Genesis, also mentioned Tiwai as a key risk in their offer documents.
The other generators - including Contact and TrustPower - are likely to be affected, one way or another, if Tiwai closes.
Rio Tinto said last month it had initiated a review and would look at all options, including closure of the plant, which consumes 13 per cent of the country's power.
New Zealand is not alone in facing this kind of issue.
Across the Tasman, US aluminium giant Alcoa is looking at shutting is Portland, Victoria, smelter, which consumes 10 per cent of the state's power. It's a similar dynamic there too; if Portland shuts, power prices are expected to go down.
Analysts are well versed in the kind of tactics that Rio Tinto use to bargain down power prices and transmission charges, so it could just be more of the same from the Anglo-Australian multinational, which takes its name from one of its mines in Spain.
But what will happen to the national power grid if the plug is pulled at Tiwai Point?
And how do you go about finding a home for power that Tiwai consumes - enough to supply half the greater Auckland region - if the plant shuts?
The answers are not simple, but it's clear that a full closure would cost hundreds of millions of dollars and would profoundly alter the supply/demand power equation in this country.
Proximity has its benefits.
Up to 5 per cent of power can be lost through its transmission over long distances, but Tiwai doesn't have that problem because it's powered by a direct link from the Manapouri Power Station, just 160km away.
There are also cost benefits of being a single customer in one location, as opposed to several in many, because there's no need to build an extensive network.
Unless some other big power can be found to set up shop at Tiwai Point, extensive investment in the power grid would be required.
The possible closure of Tiwai has been a key "what if" for the power sector for many years.
All the players are likely to have gone through the scenarios, time and time again.
Transpower, the state-owned power grid operator, has done a number of studies to understand what projects would be required if Tiwai were to close.
Two projects had already been completed as they provided additional benefits outside any changes with Tiwai, Transpower said in a statement.
There were several other projects that would be required to relieve any transmission constraints in the lower South Island and to shift the excess generation north, if needed.
Three of those were already approved, and the others would require regulatory approval, it said.
In a paper released three years ago, Transpower detailed the kind of money that would need to be spent on the grid if Tiwai closed.
Under its "what if" scenario, a closure or partial shutdown would require a fourth HVDC cable between the North and South Islands, which would take five years to build, at a cost of $150 million.
The Bunnythorpe-Tokaanu line would need to be "uprated", also taking five to eight years and costing $250m.
The Central North Island line would also need to be uprated, also taking five to eight years, at a cost of $100m.
"This plan will incur some hydro spill until all investments are in place, with the amount depending upon hydrology at the time."
"If Tiwai closes or reduces capacity there would be no security of supply concerns for the rest of New Zealand," Transpower said.
"In the short term, we may not be able to fully dispatch southern hydro lakes (particularly Manapouri) on the power system," the Transpower paper said then.
Rio Tinto has initiated a strategic review of the smelter, and one of the options includes closure. An update from the company is expected by the end of first quarter of next year.
Specifically, an exit in April 2021 would have a material negative impact on the power generator/retailer valuations by 6-12 per cent and reduce yields in the near term by an average of about 30 per cent, Grant Swanepoel head of institutional research at Craigs Investment Partners, said.
"This escalation of risk means we now adjust our base valuations for a 50 per cent probability of an exit occurring," Swanepoel said in a research note.
"If Tiwai exits, water is trapped in the bottom of the South Island, depressing prices for up to seven years," he said.
Until the constraints are alleviated, wholesale prices would likely fall to NZ$40/MWh.
"This has a negative impact on retail pricing, some water spill in the early years and capacity management.
"This could be a catalyst for converting coal-burning industries to electricity," he said.
Big changes ahead
John Kidd, head of research at Enerlytica, said profound changes lay ahead, should Rio Tinto follow through.
He said the most likely outcome from an abrupt exit would be the "stranding" of energy in the southern end of the South Island.
That would lead to very low prices in the deep south, relative to the North Island.
Kidd said a closure raised two big, separate, issues: power as a commodity and the transmission of it.
"Commodity-wise you have got 13 per cent that would leave instantly, assuming there was a full exit, and that would be fairly devastating for the market."
"In the 10 years since the Global Financial Crisis, we have not really seen electricity market grow at all - it's been flatlining," Kidd said.
"There are various reasons for that, but to see 13 per cent of the market disappear would obviously be a major hit, and it would require the market to rebalance to accommodate that."
Adding to the intrigue around Rio Tinto is the Electricity Authority's review of transmission pricing methodology, currently underway. At outcome is expected mid-way through 2020.
The methodology in its current proposed form provides NZAS with much lower-cost relief than it had been targeting.
"While there may be room to move between the parties, a solution will not be costless," Kidd said.
Then there is the "reset" of Transpower. As a regulated monopoly, Transpower has its revenue fixed by the Commerce Commission every five years. The next period runs from 2020 to 2025.
Tiwai's closure would be "demand destruction on a scale that the market has not seen before," Kidd said.
"What it means is that the most expensive thermal plant would exit quite quickly, and that really boils down to the Rankines," Kidd said, in reference to Genesis Energy's coal and gas-fired Rankines at Huntly.
Kidd said Contact Energy's gas-fired Taranaki Combined Cycle Plant would follow.
"If you look across the thermal portfolio, where does the supply response come from to meet what has happened to the demand? That's the question."
Kidd said there could be some "perverse" outcomes if Rio Tinto left.
"It will be very helpful towards achieving 100 per cent renewable generation policy, but it would take a hell of a lot of demand destruction to achieve that."
It would help the economics of electrification in certain industries. In dairy, could help reduce Fonterra's reliance on coal for its big dries in the South Island.
"But it's not happy days because there will be additional spending involved with grid augmentation to allow power to be exported from the deep south."
"In terms of transmission, the result is probably going to be negative for the rest of the electricity system because there will be additional costs involved that will need to be recover from those users that are left," Kidd said.
Analysts said Tiwai appeared to be cashflow positive at the moment.
The problem that producers faced last year was a spike in the price of the raw material alumina, while aluminium prices were flat.
That resulted in the collapse of the "smelt spread" between the two in the second half of last year, but there has been a recovery since then.
Kidd said the economics around smelters could get tricky.
"It is a very dark science to calibrate, or to reach a view as to what Tiwai pays in the market versus what the 'indifference' price is in the market."
The indifference price is the price that would be achieved on that volume of power if the smelter was not there.
"And how indifferent is Meridian and how indifferent is the market to these two scenarios? That's the fundamental question that the sector is always grappling with.
"They will be digging very deep at the moment to get to the bottom of their thinking as to whether or not they want to participate in any deal that may take shape from this strategic review."
Craigs IP's Swanepoel said Tiwai's closure would end up trapping about 1500 gigawatt hours of water at the bottom of the South Island, sending power prices lower.
"At the same time, you have two new windfarms [from Tilt and Mercury] coming on in 18 months time, bringing 1000 gigawatt hours on line," he said.
Swanepoel said if the plant was to close, a phased shutdown would be the best option.
S&P Global Ratings has continued to factor the Tiwai risk into its ratings assessment for entities in the sector.
Parvathy Iyer, senior director, infrastructure ratings at S&P Global, said Rio Tinto, once it had made its decision, needed to give 12 months notice of its intentions, which would prove important an window for the sector to adjust.
A 24-month or 36-month window would be even better, Iyer said, but she doubted Tiwai's future would be decided on the economics of the plant alone.
"It might depend on how the global supply of aluminium is looking - prices have been under pressure - and whether Rio Tinto wants to be in the business of making aluminium in New Zealand or not," she said.
All messed up
Craig's IP's Swanepoel said the effects of Rio Tinto's departure would be far-reaching.
"The signalling in the industry is going to be absolutely messed up, in terms of what we will actually need in the medium to long term."
"You would expect those with major hydro exposure in the South Island to hurt materially in the short term."
"The bottom line is that it's not up to New Zealand to kick Rio Tinto out," he said.
"It's up to Rio Tinto to leave."