Turitea Windfarm Potential Capacity Factor (Utilisation)
Quote:
Originally Posted by
Snoopy
We can also subtract the $464m (AR2020 p10) required to complete the new Turitea wind farm to the full extent of the 60 consented turbines.
The new Turitea windfarm is scheduled to be built in two stages. There is the 33 turbine northern section timetables to be completed by the end of Q1 FY2021. The second 27 turbine southern section is set to be completed by the end of Q2 in FY2022. When finished this windfarm will consist of 60 turbines, and is budgeted to cost $465m including capitalised interest (AR2020 p40).
The Turitea wind farm will be New Zealand's wind farm with a total capacity of 222MW. Once operating the average projected energy to be produced each year is projected at 840GWh/year. If the wind farm were to operate 24/7 then the total energy generated would be:
222MW x 24 hr/day x 365 day/year = 1944720 MWh/year = 1945 GWh/year
This would suggest the windfarm will operate at: 840/1945 = 43.2%
It is interesting to compare this with Trustpower's Tararua wind farm.
http://www.windenergy.org.nz/tararua-wind-farm
This has a combined capacity of 161MW with average annual generation of 620GWh. The maximum energy that can be generated in any year is:
161MW x 24 hr/day x 365 day/year = 1410360 MWh/year = 1410 GWh/year
This would suggest the Trustpower Tararua windfarm will operate at: 620/1410 = 44.0%
Both of those operating yields seem high for a wind farm. Nevertheless they are in close agreement. That would suggest the wind farm Capacity Factors that I have calculated are correct.
SNOOPY
Mercury's Hidden Power Stations (FY2020 Perspective)
Quote:
Originally Posted by
Snoopy
A couple of interesting statistics from FY2012, a year without an unusual river inflow. Mighty River Power had a 51.4% capacity utilisation from their hydro stations and a very impressive 94% capacity utilisation from their geothermal stations. This gives an idea of the relative importance of the two kinds of generation in relation to total energy generated by MRP. Since the commissioning of the latest geothermal station (Ngatamariki) in 2013/2014, MRP have enough revalued capital on the books to build yet another 'free' geothermal power station if they so choose. Let's say this potential new station could deliver 100MW. By how much would that increase the base generating capacity of MRPs portfolio?
1044MW Hydro (existing) x 0.514 = 537MW (effective)
463MW Geothermal (existing) x 0.940 = 435MW (effective)
100MW Geothermal (new) x 0.940 = 94MW (effective)
Hence the effective new capacity increase is:
94 / (537+435) = 10%
OK that new power station is not yet built, or even hinted that it will be started. But I would argue that MRP already has this new hidden value built into the company. The company is effectively 10% bigger than its current production capacity, and could up size by 10% seemlessly if management so chose to do it.
Now we are getting to the nub of why I believe Mercury is priced so highly by the market.
The above quote is from six years ago. That new 100MW Geothermal power station that I mooted was never built. But the capital that could have built that power station is still within Mercury. And now Mercury is moving in a new direction with the new Turitea Wind Farm near Palmerston North already under construction. And they have consent for another wind farm, Pukeito, to the east as well.
Quote:
Originally Posted by
Snoopy
The remaining thin air capital on the balance sheet is therefore: $897m - $215m = $682m
The net total of this 'thin air capital' that has been accumulated could theoretically support extra debt 'd' according to the company's optimised gearing ratio.
'd' / $682m = 45% => d=$307m
We thus have a total available for investment amount of: $682m (equity) + $307m (debt) = $989m dollars, while still staying within Mercury's own optimised balance sheet guidelines.
From this 'available for investment total', we can subtract the $144m paid to acquire a 19.9% stake in Tilt Renewables in FY2018. We can also subtract the $464m (AR2020 p10) required to complete the new Turitea wind farm to the full extent of the 60 consented turbines. By my calculations that still leaves:
$989m - $464m - $144m = $381m
We can add to this the $272m proceeds from the sale of the Metrix metering business in FY2019.
That makes total capital of $381m + $272m = $653m still available for investment.
With Turitea already budgeted for, the balance sheet suggests there is also enough investment capital available to build Pukeito as well. And Pukeito could be 40% larger than Turitea with the amount of investment capital available. So how much would these two wind farms boost Mercury's operational generating capacity?
1063MW Hydro (existing) x 0.514 = 546MW (effective)
463MW Geothermal (existing) x 0.940 = 435MW (effective)
222MW Turitea Wind (new) x 0.432 = 96MW (effective)
310MW Pukeito Wind (new) x 0.432 = 134MW (effective)
Hence the effective new capacity increase from the fully funded but as yet not switched on power stations is:
(96+134) / (546+435) = 23.4%
I realise this is a strange concept for some. That is, giving value to a couple of Mercury consented power stations that are not yet in service. But as you will see it does appear the market is doing just that. Because the current MCY share price is very difficult to justify on Mercury's earnings today.
SNOOPY
Capitalised Dividend Valuation (FY2020 view) Part 1: Data
Quote:
Originally Posted by
Snoopy
Financial Year |
eps |
Dividend Paid (per share) |
Ordinary Dividend Paid (per share) |
2014 |
13.3c |
7.2c + 5.2c = 12.4c |
7.2c + 5.2c = 12.4c |
2015 |
9.7c |
8.3c + 5.0c + 5.6c = 18.9c |
8.3c + 5.6c = 13.9c |
2016 |
10.1c |
8.4c + 2.5c + 5.7c = 16.6c |
8.4c + 5.7c = 14.1c |
2017 |
12.0c |
8.6c + 4.0c(NI) + 5.8c = 14.4c +4.0c (NI) |
8.6c + 5.8c = 14.4c |
2018 |
|
8.8c + 5.0c(NI) + ?c = ? |
8.8c + ?c = ? |
Total FY2014 to FY2017 |
|
62.3c + 4.0c (NI) |
55.1c |
Notes:
1/ The above 'per share' table is based on the 1,400m MCY shares on issue.
2/ (NI) means 'Not Imputed'
For valuation purposes I intend to work with the ordinary dividend. I do this because doing so incorporates a 'management judgement' of what management thinks normal earnings are. The special dividends that I haven't counted I consider 'capital adjustments' to restore the company to management's desired equity ratio. I do not think that these special; dividends will be indicative of future payments.
From the table, the average annual normalised dividend payment has been: 55.1c/4 = 13.8c. However, I prefer to use the latest available data. In this instance this means I drop the final dividend payment for FY2013 (which was paid in FY2014) of 7.2c, and replace it with the final dividend paid for FY2017 (paid in FY2018) of 8.8c. So my four year tax paid normalized dividend average becomes:
( 5.2c +(8.3c + 5.6c) + (8.4c + 5.7c) + (8.6c + 5.8c) + 8.8c) /4 = 14.1c
Financial Year |
Normalised 'eps' |
Net Dividend Paid (per share) |
Gross Dividend Paid (per share) |
2016 |
10.3c |
8.4c + 2.5c(S) + 5.7c |
11.67c + 3.47c + 7.92c = 23.06c |
2017 |
12.1c |
8.6c + 2.88c(NI,S) + 5.8c |
11.94c + 4.0c + 8.06c = 24.00c |
2018 |
14.0c |
8.8c + 5.0c(S) + 6.0c |
12.22c + 6.94c +8.33c = 27.49c |
2019 |
11.7c |
9.1c + 6.2c |
12.64c + 8.61c = 21.25c |
2020 |
11.7c |
9.3c + 6.4c |
12.92c + 8.89c = 21.81c |
2021 |
?c |
9.4c + ?c |
13.06c + ?c = ?c |
Total FY2016 to FY2020 |
59.8c |
84.68c |
117.61c |
Notes:
1/ The above 'per share' table is based on the 1,400m MCY shares on issue.
2/ (NI) means 'Not Imputed', (S) means 'Special Dividend'.
In a change of policy I have decided to work with all dividends. Despite special dividends not being repeatable, they are paid frequently. And, using this particular valuation method, I have not recognised these special dividends as a benefit for shareholders in any other way.
From the table, the average annual normalised gross dividend payment over the last five years has been:
(23.06c+24.00c+27.49c+21.25c+21.81c)/5 = 23.52c
Nevertheless time has moved on and I have decided to include the first dividend for FY2021 (13.06c gross) -that has already been paid- and remove the equivalent dividend from five years previously (11.67c + 3.47c gross).
(7.92c+(24.00c+27.49c+21.25c+21.81c)+13.06c)/5 = 23.11c
SNOOPY
Capitalised Dividend Valuation (FY2020 view) Part 2: Calculation
Quote:
Originally Posted by
Snoopy
Nevertheless time has moved on and I have decided to include the first dividend for FY2021 (13.06c gross) -that has already been paid- and remove the equivalent dividend from five years previously (11.67c + 3.47c gross).
(7.92c+(24.00c+27.49c+21.25c+21.81c)+13.06c)/5 = 23.11c
Based on a 4.5% gross yield that I now used for utilities in this ultra low interest rate environment, I can calculate a capitalised earnings valuations for MCY.
23.11c / (0.045) = $5.14 (based on averaged, dps)
However, this valuation does not take into account the two hidden not yet operational wind farms for which the capital is already built into today's balance sheet. Adjusting for that (my post 1323) my fair value for MCY today is:
$5.14 x 1.234 = $6.34
As I write this MCY is trading at $6.81. That means I see it as overvalued, although not by as much as some might think ( +7.4% ).
SNOOPY
NPS 'Net Promotor Score' (FY2020 view)
Quote:
Originally Posted by
Biscuit
Negative attitudes probably don't drive consumer electricity choices.
I don't think Mercury would agree with you Biscuit as they seem very concerned at getting their 'Net Promotor Score' up. Why have I brought up this topic? Because Mercury thinks it is important!
So what is NPS?
According to:
https://exerciseindustryawards.co.nz...lityentry/nps/
it will involve one simple question:
------------------------------
”On a scale of 1-10 how likely are to recommend your (power retailer) to a friend or colleague?”.
Depending on the score that is given the respondents can be categorised into one of three groups:
Promoters – giving a score of 9 or 10, Passives – giving a score of 7 or 8, Detractors – giving a score of 0-6.
The Net Promoter Score is then calculated as the difference between the percentage of Promoters and Detractors.
For example: If 100 responses come back – 45 of them giving a score of 9 or 10, 30 of them a 7 or 8 and 25 of them 0-6, the calculation would be 45-25 (45% Promoters, 25% Detractors) leaving a score of 20.
-------------------
According to AR2020 p7, the NPS for Mercury Energy is 13.7
For comparison:
1/ Trustpower Net Promoter Score is 40 (Infratil March 2017 Presentation)
2/ The Contact Energy 'Net Promoter Score' is 36 ( CEN AR2020 p12)
3/ The Meridian Energy 'Net Promoter Score' is 6 ( MEL PR2019 p11)
4/ The Genesis Energy Net Promoter Score is -4 ( https://customer.guru/net-promoter-score/genesis-energy)
SNOOPY
Growth Modelling for Mercury
Quote:
Originally Posted by
Ferg
In a growth environment the valuation formula is D/(r-g) where D is the annual dividend, r is the required rate of return and g is the rate of growth in perpetuity of the dividend stream.
Your own calculations of growth imply quite a high growth rate. If we assume annual growth in dividends at a lower rate of say 1% p.a. then the valuation is:
23.11c / (0.045 -.01) = $6.60.
This is not too far off the current price so I suspect some players are pricing growth into their dividend valuation as per your earlier post on generation capacity.
There are various ways of modelling growth when evaluating investment prospects Ferg. Many of my investments are for income. For these I generally model 'no growth'. Then if my valuation comes in below the market valuation, the difference between the two numbers becomes the 'growth premium'. I then have to assess whether the growth premium is fair or whether the market is being a little too optimistic on growth prospects.
There are a couple of reasons why I tend to model 'no growth'.
1/ It is easier. I don't have to assess what I think the growth rate should be.
2/ It is conservative. If I model 'no growth' yet get growth, that growth is a bonus. If an investment makes sense under 'no growth' assumptions it also makes sense if there does end up being growth.
The 'Capitalised Dividend' model assumes 'no growth'. By implication that means historical earnings assume extra importance. That is because if there is indeed 'no growth' those historical results are good predictors of future results. That in turn means my forecasting is based on real historical results over a business cycle. I am much more comfortable with this than the alternative: Appointing myself as a 'pointy head' and trying to arbitrarily assess future prospects in an industry where I really have no expertise.
If you look at MCY dividend income that I have summarised in my post 1326, the sad truth is that MCY is a 'no growth' company. I think this is a fair assessment. There is the spurious claim that ordinary dividends have been increasing for a decade or more that carefully leaves out the fact that special dividends of recent years that have been dropped. There are the excuses of unusual water inflows (both less and greater than average) that has disrupted earnings patterns. Recent history shows that most years have had unusual water inflows. That means the 'hypothetical average' is a by product of feast and famine years. The chance of any given year being average is actually very low.
Going back to my calculated valuation figures this means the growth premium for MCY is very high. At yesterday's market closing price of $6.81 the 'growth premium' is:
$6.81 - $5.14 = $1.67 ( +32%)
At first glance that seems crazy for 'no growth' share.
However, what this doesn't take into account is that real growth for Mercury comes in lumps. Mercury currently have the capacity to build two giant new wind farms from existing resources (one Turitea is already under construction) , with no capital raising required. The fuel costs for these wind farms is zero. So what we have here is the capacity to earn a significant increment in revenue with all construction costs already built into the balance sheet. Furthermore power is a 'must have' in the modern world. So demand is not cyclical, and that in turn makes the cashflow from those future power plants more valuable.
My calculations show that even accounting for this 'hidden power station effect' (my post 1327) the share price for MCY is too high. But you can say the same for most NZX shares right now. So by selling MCY and investing in another overvalued NZX share you are no better off. And you are certainly better off holding MCY for income purposes than putting your money in the bank! This is why I am not selling MCY. And no doubt many MCY investors feel the same way.
Because the growth of MCY is by construct very lumpy, my own view is that trying to model a long term growth rate for MCY doesn't make sense. It is true that modelling a 1% growth rate compounding for 10 years.
(1+0.01)^10 = 1.105
is equivalent to a zero growth model with a one off growth increase of 10.5%. But since the actual growth rate is not 1% per year compounding, I don't see any point in modelling the growth this way.
SNOOPY