Contact Energy - CEN Chart

[Snoopy]

Not covered in those calculations is the fact that Contact is heavily hedged by its retail book, and the number of ToU customers to the extent that it often struggles to meet that hedge position when prices are high. I am not sure how you will account for that hedging in your calculations.

I imagine you are talking about Time of Use customers, who make a commitment to only draw power at hours when the power price is low? But of course the weather gods do not always play by the book. So there may be occasions where Contact has to supply power at what were forecast to be 'low rate times', when the actual power demand was high (meaning cost of supply exceeds the price Contact have contracted to supply).

I am not sure if you are talking about Contact further hedging this risk. Or if you are talking about the supply agreement itself being the hedge (from a power consumer perspective).

My general policy is to take out all hedge deals when evaluation a company's profitability. My reasoning being that these are all 'zero sum games'. Thus while Contact may do well out of such deals in some years, the weather patterns may see the same deals backfire in other years. Thus -long term- the expected contribution of hedge deals to any company's profitability is a net zero.

SNOOPY

[Snoopy]

Having already done the exercise below for FY2022, I thought it worth repeating the process for FY2021, in case FY2022 was a 'rogue data year'. The fact that power shortages have been in the headlines this winter (FY2022) might suggest that it was merely unusual weather patterns that were causing unusual wholesale power price spikes. So what happened over FY2021 at Contact Energy?


Wholesale Power Price Effect of potential Onslow on Contact Energy (FY2021 re-modelled retrospectively)

Month	Percentage of Wholesale Power Priced > $111MW/h {A}	Total Energy Generated (GWh) {B}	Energy Generated > $111MW/h (GWh) {A}x{B}	Thermal Energy Generated {C} (GWh)	Hydro Energy Generated > $111MW/h (GWh) {A}x{B}-{C}	Onslow Lost Wholesale Value to Contact Energy @ $60MW/h
June 2021	90%	806	725	184	541	$32.46m
May 2021	95%	810	770	220	550	$33.00-m
April 2021	93%	662	629	185	464	$27.84m
March 2021	90%	593	534	77	457	$27.42m
February 2021	90%	532	479	54	425	$25.50m
January 2021	55%	626	344	22	322	$19.32m
December 2020	50%	578	289	27	262	$15.72m
November 2020	38%	604	230	54	176	$10.56m
October 2020	55%	659	362	81	281	$16.56m
September 2020	63%	801	505	179	326	$19.56m
August 2020	43%	869	374	251	123	$7.38m
July 2020	60%	866	520	278	242	$14,52m
Total						$250.14m

Calculation Notes

1/ The "Percentage of Wholesale Power Priced > $111MW/h" (table column 1) is arrived at by looking at the monthly "Distrbubution of wholesale market price by trading periods" (from each of twelve Monthly Operating Reports) 'bar graph' top three pricing categories: $111-135, $136-160 and >$160, for the Otahuhu electricity market power pricing node. Next, I add together the three 'bright red bars', representing the current month, by eye. Then I write down the 'cumulative total percentage figure' to an accuracy of about 5%. Next, I do the same exercise for the 'Benmore power pricing node'. I average the two results to get a final 'representative percentage'.

---------------

EBITDAF for Contact Energy over FY2021 was $553m. So taking $250m off that figure, reduces EBITDAF to:

$553m - $250m = $303m = DOUBLE OUCH!

This shows that the 'massive potential hit' that showed as an example of what might happen to Contact Energy profitability in the future (post 2290) was likely not a one off rogue event based on unusual weather conditions.

SNOOPY

[Jantar]

I imagine you are talking about Time of Use customers, who make a commitment to only draw power at hours when the power price is low? But of course the weather gods do not always play by the book. So there may be occasions where Contact has to supply power at what were forecast to be 'low rate times', when the actual power demand was high (meaning cost of supply exceeds the price Contact have contracted to supply).

I am not sure if you are talking about Contact further hedging this risk. Or if you are talking about the supply agreement itself being the hedge (from a power consumer perspective).

My general policy is to take out all hedge deals when evaluation a company's profitability. My reasoning being that these are all 'zero sum games'. Thus while Contact may do well out of such deals in some years, the weather patterns may see the same deals backfire in other years. Thus -long term- the expected contribution of hedge deals to any company's profitability is a net zero.

SNOOPY

ToU customers do not just draw when prices are low, bur pay a different price depending on whether they are taking energy during a business day, non business day, or at night.

But the real hedge is the retail book. Retail customers pay a fixed price irrespective of the wholesale price, and make up a very large part of any Gentailers energy sales. There is only a very small part of the company's generation that is fully exposed to the wholesale market. I don't know what it is now, but it used to be only around 10% exposure. There were many times that we would be scrambling trying to buy 50 or 100 MW on a short term CFD from another gentailer. Similarly there were many times that another gentailer would be begging us for a contract.

[Jantar]

Wholesale Power Price Effect of potential Onslow on Contact Energy (FY2022 re-modelled retrospectively)

Month	Percentage of Wholesale Power Priced > $111MW/h {A}	Total Energy Generated (GWh) {B}	Energy Generated > $111MW/h (GWh) {A}x{B}	Thermal Energy Generated {C} (GWh)	Hydro Energy Generated > $111MW/h {A}x{B}-{C}	Onslow Lost Wholesale Value to Contact Energy @ $60MW/h
June 2022	70%	760	532	165	367	$22.02m
May 2022	100%	692	692	141	551	$33.06m
April 2022	95%	597	567	139	429	$25.68m
March 2022	90%	637	573	156	417	$25.02m
February 2022	60%	566	340	37	303	$18.18m
January 2022	70%	606	424	48	376	$22.56m
December 2021	10%	664	66	22	44	$2.64m
November 2021	5%	666	33	15	18	$1.08m
October 2021	10%	710	71	25	46	$2.76m
September 2021	20%	696	139	36	103	$6.18m
August 2021	53%	763	404	76	328	$19.68m
July 2021	75%	911	683	186	497	$29.82m
Total						$208.68m

Notes.

1/ The "Percentage of Wholesale Power Priced > $111MW/h" (table column 1) is arrived at by looking at the monthly "Distrbubution of wholesale market price by trading periods" (from each of twelve Monthly Operating Reports) 'bar graph' top three pricing categories: $111-135, $136-160 and >$160, for the Otahuhu electricity market power pricing node. Next, I add together the three 'bright red bars', representing the current month, by eye. Then I write down the 'cumulative total percentage figure' to an accuracy of about 5%. Next, I do the same exercise for the 'Benmore power pricing node'. I average the two results to get a final 'representative percentage'.

---------------

EBITDAF for Contact Energy over FY2022 was $537m. So taking $209m off that figure, reduces EBITDAF to:

$537m - $209m = $328m = OUCH!

This is a massive potential hit coming up for Contact shareholders (and all shareholders of gentailers in fact).

My modelling is somewhat simplistic and Jantar, in post 2288, has pointed out a couple of obvious weaknesses.

1/ There would be a consummate 'upping of supply bids' at the lower end of the power price scale, for pumping. Getting more money at the lower end of the power pricing cycle would be good for Contact Energy.
BUT (1 Counterpoint)/ If you look at the distribution of power pricing, in most instances of 'wholesale power pricing' it is the upper figure range of >$160MW/h that dominates. But in reality power pricing at the peak can go much higher than $160/MWh. So there is a good chance I have underestimated the real loss in wholesale power price peaking that Contact would incur, by using that $160MW/h figure, should Onslow become operational.
2/ If Contact is being paid less for their wholesale power, then the retail side of the gentailer business becomes more profitable. So some of that 'loss in wholesale power profit' would be clawed back at the retail end.
BUT (2 Counterpoint)/ Retail is a less capital intensive business to enter and competition is more intense than in the wholesale space. The reduction in 'wholesale power price spikes' should make competitor retail businesses easier to operate profitably and sustainably. Thus I would expect retail competition to intensify 'over the medium term', if Onslow goes ahead. And that means lower profit margins for the existing gentailers' retail arms over time.

At this point there are lots of factors to consider. But even if the real loss to Contact is only half of that my somewhat crude calculation assessment, a '$100m hit' to EBITDAF is still pretty substantial. This exercise is making me think again about what premium I should pay for having shares in these gentailers for their supposed 'certain future cashflows' going forwards.

SNOOPY

Snoopy, there is so much wrong with this analysis that it is not fit for any purpose. Some of this I have already commented on, but lets look at just the June 2022 report to show up a bit more of it.

Your figures give total generation of 760 GWh, which is correct. But look at the line of the report that says Contracted Electricity Sales of 751 GWh. That only leaves 9 MWh exposed to the wholesale market, a long way from the whole 760 GWh.

There is absolutely no reason to say that all generation above $111 per MWh should be capped at that price. Onslow would place a soft cap, not a hard cap on prices. Overall wholesale prices above the start price of Onslow generation would be lower, . At the other end of the scale, wholesale prices lower than the start price of Onslow pumping would be higher, but again would not have a fixed limit. You have not included that aspect in your calculations. Nor have you included the savings in gas purchases due to there being no thermal generation in the market.

[RTM]

Not sure if this has been posted before re Onslow:

https://www.google.com/url?sa=t&rct=...ZtjMqlbdqaGUDr

Thanks for posting the link. I had not seen it before.
Really interesting.

[Snoopy]

Your figures give total generation of 760 GWh, which is correct. But look at the line of the report that says Contracted Electricity Sales of 751 GWh. That only leaves 9 MWh exposed to the wholesale market, a long way from the whole 760 GWh.

Let's take one step back just to be clear what that table I posted using 2022 information (my post 2290) is meant to show. It is what I term a 'scenario analysis'. It answers the question:

------------------------

1/ If the core generating capacity at Contact Energy remained 'as is' in the future, -AND-
2/ If the climate and demand picture in NZ produced the same energy from Contact owned generation assets as was delivered over FY2022 (but with thermal assets removed, and replaced by baseload geothermal assets) -AND-
3/ Onslow was built and operating, and selling into the power market energy at an average price of $100/MWh whenever the Onslow start up delivery price signal would be triggered by a third party lowest offer market price at $111/MWh - or more -

THEN what would be the effect on Contact Energy's profits?

---------------------

I see that figure of 760GWh total generated over June 2022 in FY2022 Jantar (from the monthly June report), which is exactly the same figure that would be generated in my scenario analysis. But that second figure you mention of 751GWh in the reporting information, that 751GWh figure was the energy contracted for sale in June 2022. That figure is not relevant to this discussion, because I am talking about some future June of a year that will come when Onslow is operating. Whatever happened back in June 2022, in terms of contracted sales demand, will not be the same figure in the future June I am looking at. And it is the future we are talking about here.

Now fast forward to this 'future June'. Will there be a large proportion of these wholesale sales signed off under contract? We don't know for sure, but probably. And what price will these contracts be signed off at? I would imagine it will be based around a forecast future spot price series, averaged over the length of the contract. And Contact will probably throw in a discount, to account for the fact that their wholesale customer(s) have committed to Contact Energy as a supplier for a fixed term. Come the day, the actual contracted power price for wholesale customers may not equal the spot market price. I would go so far as to say it almost certainly won't. But without knowing the future, the best tool for forecasting the future is the seasonally adjusted projected spot price curve. So for this reason, I do not think it is unreasonable to forecast future customer demand and sales prices - whether contracted or not - that are based on the expected spot price. The way I look at things, those future 'wholesale contract prices' are still 'at risk'. It is just that the risk has been taken out at a different (earlier) stage of the power sales process, rather than 'pay on the day' at spot rates. Given this, I don't think it is unreasonable to suggest that future demand pricing - including that in sales contracts yet to be inked - is best estimated by 'spot pricing'. IOW the projected EBITDAF loss that I am forecasting for Contact Energy is the relevant figure.

There is absolutely no reason to say that all generation above $111 per MWh should be capped at that price. Onslow would place a soft cap, not a hard cap on prices. Overall wholesale prices above the start price of Onslow generation would be lower, .

You are right of course. If you had a 'soft cap', then you would get a distribution of offered prices. So what I have done is to make a simplifying assumption. I am saying is that the 'volume weighted average of trigger prices' that would see Onslow 'start generation' would be $111/MWh. 'On average' over all the times Onslow was started, the time weighted average of power sold would flow onto the market at $100/MWh. But these figures are averages that do not necessarily reflect the trigger price, nor the selling price, of any individual Onslow generation event.

Nor have you included the savings in gas purchases due to there being no thermal generation in the market.

Good point. I see from the 'operational data' listed information that 1.3PJ of gas was used for 'internal generation.'

But where do I find the dollar figure associated with that? I can't see it declared separately in the monthly operational report.

SNOOPY

[Snoopy]

At the other end of the scale, wholesale prices lower than the start price of Onslow pumping would be higher, but again would not have a fixed limit. You have not included that aspect in your calculations.

I wasn't going to go into detail on the 'buy side' upside of Onslow for Contact Energy. The reason for that will become apparent. But since Jantar specifically mentioned it, I have decided to put the bonus 'buy side' table on this thread after all.

I am assuming that Onslow will start buying electricity for pumping once the trigger power price signal drops below $35/MWh. I am assuming Onslow on average, will pay $50/MWh for pumping purposes. This will provide a net benefit to Contact Energy of:

$50/MWh - $35/MWh = $15/MWh for each MWh purchased from Contact.

So what dollar value benefit will Contact receive if Onslow pumps according to this plan?

Wholesale Power Price Effect of potential Onslow on Contact Energy (FY2022 re-modelled retrospectively)

Month	Percentage of Wholesale Power Priced < $35MW/h {A}	Total Energy Generated (GWh) {B}	Energy Generated < $35MW/h (GWh) {A}x{B}	Onslow Gain in Wholesale Value for Contact Energy @ $15MW/h
June 2022	6.5%	760	49	$0.74m
May 2022	0%	692	0	$0m
April 2022	0%	597	0	$0m
March 2022	0%	637	0	$0m
February 2022	23%	566	130	$1.95m
January 2022	2.5%	606	15	$0.23m
December 2021	37.5%	664	249	$3.74m
November 2021	10%	666	67	$1.01m
October 2021	30%	710	213	$3.20m
September 2021	18%	696	125	$1.88m
August 2021	22%	763	168	$2.52m
July 2021	2.5%	911	23	$0.35m
Total				$15.62m

Calculation Notes.

1/ The "Percentage of Wholesale Power Priced < $35MW/h" (table column 1) is arrived at by looking at the monthly "Distrbubution of wholesale market price by trading periods" (from each of twelve Monthly Operating Reports) 'bar graph' in the bottom category < $35MW/h, for the Otahuhu electricity market power pricing node (the three 'bright red bar', representing the current month. I write down this figure to an accuracy of about 5%. Next, I do the same exercise for the 'Benmore power pricing node'. I average the two results to get a final 'representative percentage'.

---------------

The gain for Contact Energy is assuming the weighted average pumping price paid to Contact Energy is $50/MWh, providing a $15MW/h incremental margin over the price that without Onslow, Contact might have otherwise expected to get for that same energy. In practice, and if Jantar's story about some run of the river hydro energy being sold for as low as a few cents per MWh is true, the actual return for Contact could be double what I am modelling here, say $30m.

But such an amount, although substantial, is still within the error bound of the downside of Onslow for Contact. I have estimated the downside figure, that must be offset against this upside to be $208m. But it could be much higher, approaching $300m. IOW the upside is more than covered by the uncertainty of the error bar around the downside. That means I have wasted my time working out the upside because it has no significance in the big picture. I could see this was going to happen before I started the upside calculation, which is why I didn't do it. However, I appreciate that not everyone can see this in advance, which is why I have done the upside calculation for all to see - so they can appreciate it wasn't worth doing.

SNOOPY

[Jantar]

Thanks Snoopy.

I see you are starting to see some of the pricing issues, but you have somehow put them the wrong way round.

Lets start with "3/ Onslow was built and operating, and selling into the power market energy at an average price of $100/MWh whenever the Onslow start up delivery price signal would be triggered by a third party lowest offer market price at $111/MWh - or more -"

The startup point of any generator is when they are dispatched by transpower, and that dispatch is set according to the offer price. Here is a Bid/Offer scenario I set up for a 960 MW Onlsow. I can show this one as it not one of options under consideration.As you can see from this stack there is no possible way that the average generation price can be lower than the initial generation price, nor can the pumping price ever be higher than the initial pumping price.


Bid-Offer.JPG

Under this Bid/Offer stack, Onslow would first be dispatched when the nodal price reached $104, and the quantity dispatched at that price would increase until it reached 108 MW. At that point Transpower would dispatch the next highest offer, which may be Onslow, or may be another hydro station belonging to another party. As the nodal price increases to $130 then the next machine at Onslow would be dispatched in increasing quantities until the dispatched quantity reached 207 MW. Again, at that point Transpower would dispatch the next highest offer, which may be Onslow, or may be another hydro station belonging to another party. This dispatch process is repeated until at a nodal price of above $619.90 all 960 MW of Onslow is dispatched.

Similarly, pumping would begin with the first unit as the nodal price slips below $45.50, and there is at least 120 MW of demand between the actual wholesale price and the bid price of $45.50. This is to try and prevent the price increasing above the Bid price when that 120 MW of pumping is dispatched, although in practice it may slip above as not all market participants watch closely to what is happening. As demand drops off, and/or more wind comes online, the price would drop further until the nodal price slips below $39.00, and there is at least 120 MW of demand between the actual wholesale price and the bid price of $39.50. This process continues until the entire 960 MW of pumping is dispatched when the price is below $6.50.

Here is the result of a single day under this scenario. The price column is the actual price on that trading period, the MW Mod is the MW that would have been dispatched, and the Adj Price is the final price if Onslow was dispatched according to the above Bid/Offer list. This shows that the prices are not affected as much as you have assumed.

Result.JPG

[Snoopy]

Thanks Snoopy.

I see you are starting to see some of the pricing issues, but you have somehow put them the wrong way round.

Lets start with "3/ Onslow was built and operating, and selling into the power market energy at an average price of $100/MWh whenever the Onslow start up delivery price signal would be triggered by a third party lowest offer market price at $111/MWh - or more -"

The startup point of any generator is when they are dispatched by Transpower, and that dispatch is set according to the offer price. Here is a Bid/Offer scenario I set up for a 960 MW Onlsow. I can show this one as it not one of options under consideration. As you can see from this stack there is no possible way that the average generation price can be lower than the initial generation price.

Bid-Offer.JPG

Under this Bid/Offer stack, Onslow would first be dispatched when the nodal price reached $104, and the quantity dispatched at that price would increase until it reached 108 MW. At that point Transpower would dispatch the next highest offer, which may be Onslow, or may be another hydro station belonging to another party. As the nodal price increases to $130 then the next machine at Onslow would be dispatched in increasing quantities until the dispatched quantity reached 207 MW. Again, at that point Transpower would dispatch the next highest offer, which may be Onslow, or may be another hydro station belonging to another party. This dispatch process is repeated until at a nodal price of above $619.90 all 960 MW of Onslow is dispatched.

I need to explain a bit more about my modelling that has the dispatch price of 'greater than $111/MWh' verses the 'signal price' (startup point) which determines when the turbine starts up of $100/MWh in my power pricing scenario.

I chose that $111/MWh figure because the "Distribution of wholesale market price by trading periods" is listed under grouped power price bands in the Contact monthly Operating Report as follows:

<$35, $36-$60, $61-$85, $86-$110, $111-$135, $136-$160, >$160

Taking the $100/MWh as a figure that you Jantar, suggested might be the starting point for dispatch from Onslow, I noticed that figure lay in the middle of the $86-$110 band. So the question I asked myself was, if the power generation market was operating in that band, would there be generation from Onslow or not? The answer being 'yes' if the matched price was above $100, but 'no' if the matched price was below $100. It was a function of the way that the power price grouping was made, between $86 and $110, that meant it was impossible to know whether power would be dispatched or not with dispatch market pricing in that price band. So I then moved up to the next price band where there would not be such a yes/no dilemma.

If the start up price was between $111 and $135 (or one of the two higher bands), then in those I cases I could be sure that Onslow would be started up (if the minimum targeted dispatch price from Onslow was $100).

Jantar you say
"there is no possible way that the average generation price ($100/MWh) can be lower than the initial generation price ($111/MWh)."

Yes I get this. So what I needed was the percentage of generation for the month with a wholesale market price of >$100/MWh. But unfortunately that figure is not available from the Contact monthly reports. So I had to move up to the next price band, starting at $111MWh+ to make sure that Onslow would be generating.

This means that if we stick to your >$100/MWh Onslow generation start point, my figures will be underestimating the generation from Onslow. That is because I will not be counting the generation that occurs between $100/MWh and $110/MWh part of the lower category $86-$110 power band.

To be clearer, what I should have done is said that my estimate of power generation from Onslow that would be dispatched when the dispatch market price was greater than $100/MWh was equal to the percentage of power generated across the three higher power price bands ($111-$135, $136-$160, >$160) added up together, while at the same time noticing that this was an 'under-reporting total', (because it did not include power generated in the $100-$110 band that was not separately disclosed). If I had said that, then the average generation price assumed of $100/MWh would not have been less than the initial generation price -that it looked like I was claiming to be $111/MWh-, (but in fact was $100/MWh in my own mind at least).

This also means that in my post 2290, the $208.68m of 'lost value' for Contact would be an underestimate. Because I did not account for the energy that Onslow was putting back into the Clutha River when the wholesale power price was between $100/MWh and $110/MWh.

SNOOPY

[Jantar]

I need to explain a bit more about my modelling that has the dispatch price of 'greater than $111/MWh' verses the 'signal price' (startup point) which determines when the turbine starts up of $100/MWh in my power pricing scenario.

I chose that $111/MWh figure because the"Distribution of wholesale market price by trading periods" is listed under grouped power price bands as follows:

<$35, $36-$60, $61-$85, $86-$110, $111-$135, $136-$160, >$160

Taking the $100/MWh as a figure that you Jantar, suggested might be the starting point for dispatch, I noticed that figure lay in the middle of the $86-$110 band. So the question I asked myself was, if the power generation market was operating in that band, would there be generation from Onslow or not? The answer being 'yes' if the matched price was above $100, but 'no' if the matched price was below $100. It was a function of the way that the power price grouping was made, between $86 and $110, that meant it was impossible to know whether power would be dispatched or not with pricing in that band. So I then moved up to the next price band.

If the start up price was between $111 and $135 (or one of the two higher bands), then in those I cases I could be sure that Onslow would be started up if the targeted dispatch price was $100.

Correct, but that only relates to first dispatch of the first machine. So that cannot be the average price received, nor the capping price. If the offer price was $100 for the first tranche then a maximum of 9% of Onslow's capacity would be dispatched at that price. About equivalent to a single Stratford peaker, so that cannot set a maximum wholesale price.