MCY Mercury NZ ( MRP Mighty River Power)

[Snoopy]

To me the most important metric is cash flow after subtracting interest,operating costs and depreciation .

CEO Fraser Whineray has made a big thing on how operating costs have greatly reduced since MCY was an SOE. But what is the actual pattern of operating costs? Strangely for a statistic so well promoted, it takes a bit of digging to find the operating costs of the business going back. The only graph I could find was the 'Operating Expenditure', listed in the financial commentary section (p12) of the Mighty River Power Annual Review for FY2014. In latter years the figure was listed in the financial commentary sections.

Year	Normalised Operating Costs
FY2017	$214m
FY2016	$217m
FY2015	$217m
FY2014	$221m
FY2013	$243m

The big saving was in the FY2014 year and this is what the company said about that at the time:

"Following the completion of the IPO in May 2013 and with no large scale development projects currently planned due to the current demand conditions, the company has focussed on lowering the cost base and achieved $20m of permanent savings relating to optimization of the life cycle maintenance programme, insurance changes, professional service fees and international geothermal development costs. <snip> Year on year operating expenses fell $99m with $69m one off expenses in FY2013 relating to international geothermal restructuring and IPO costs."

Of course in my comparison table I would never have included IPO costs in operating costs, and I didn't. Likewise I would miss out the financial operating resources formerly dedicated to the international geothermal business which was quickly abandoned when Fraser Whineray took over as CEO.

The trumpeted $20m in ongoing savings is useful, even in light of the large number of MCY shares on issue:

$20m/ 1,400m = 1.4cps

On an after tax basis this means profits have been permanently increased by 0.72 x 1.4c = 1cps

To put this in context, normalised profit (NPAT) was 12cps in FY2017, and 1cps represents 8.3% of that total. Significant and worth having.

Nevertheless, it does look like the 'game changing' reduction in operating costs is over. I would imagine that supervising all of those overseas geothermal projects would have been quite expensive for little immediate return. So good savings would have been made cutting out those costs. The easiest savings in operating costs is to redefine what normal operations are. This seems to be what has happened here: game changing in the past, but not likely to be game changing into the future.

SNOOPY

[Snoopy]

There is only one problem and that is that the cost of new generation is DECREASING. There should be an impairment against the asset values. The gravy train is over.

As horus has identified, if power can be brought to the home in the future at much reduced price, then the value of existing power stations will decrease. But, as tautological as this sounds, the cost of power is not just determined by the cost of producing power. The factors that Mercury look at when valuing their assets can be found in the footnotes of the property plant and equipment pages, specifically p17 of AR2017 (for example), I have tabulated these for the last few years, where available, so that investors can see how these valuation assumptions have changed over time

Financial Year	Average Operational Expenditure	Wholesale Energy Price	Net Average Production Volumes	Post Tax Discount Rate	Net Revaluation Movement
FY2017	$158m /p.a.	$70 to $104 /MWh	6567 GWh/year	7.5% to 7.9%	$52m - $4m = $49m
FY2016	$174m /p.a.	$66 to $102 /MWh	6556 GWh/year	7.4% to 7.9%	$137m - $1m = $136m
FY2015	$168m /p.a.	$63 to $97 /MWh	7131 GWh/year	7.5% to 7.9%	$497m - $76m = $421m
FY2014	$188m /p.a.	$70 to $95 /MWh	7107 GWh/year	Unknown	$40m - $0m = $40m
FY2013	Unknown.	Unknown	Unknown	Unknown	$80m - $5m = $75m

So what does all the above mean?

1/ The first thing to recognize is that the above table is looking at long run average prices and costs. In FY2017, the actual power produced by Mercury was 7533 GWh. That was 14.7% above long term projections. Mercury are not valuing their assets based on 'one good year'.

2/ The post tax discount rate looks surprisingly stable, and in absolute terms quite high. It is a pity we shareholders are not privy to the discount rate used in earlier years. But I feel that modest interest rises from current near term lows may not affect the value of generation assets on the books much, if at all.

3/ The modelled wholesale energy price is expressed as a range. This suggests to me that a range of possible future scenarios have been used for valuation purposes and probabilities applied to the respective scenarios evaluated. The higher priced scenarios have a rising maximum price over time, yet the minimum price scenarios look flat. I am not surprised by this. In times of plenty, the price of power generated should trend towards the 'backbone price' of hydro and geothermal generation. In times of shortage, the on market price is likely to be not only higher but more volatile, particularly as 'surplus' thermal power generation stations have been closing.

4/ After a difficult three years (low Waikato inflow over FY2013, FY2014 and FY2015) , the long term projected electricity to be generated per year has dropped by around 8%. Yet the value of the power generating assets has not dropped (they are modestly up in value) over this time.

5/ Average projected Operational Expenditure has declined by 16% over the four years disclosed. This will have an after tax profit effect (based on a 28% tax rate) of 0,72x the 'Operational Expenditure' on an after tax profit basis (for FY2017 $158m x 0.72 = $114m). Net profit for the year FY2017 was $184m. So cutting projected expenditure looks to be having a large effect on the net profit and hence the underlying value of the generation assets that Mercury owns. I note that the largest increase in 'asset generation value' occurred after projected operational expenditure was cut the largest from $188m p.a. to $168m p.a. (co-incidence or not?)

6/ It is a pity that I can't fill the unknown gaps in my table, as this would provide a much better medium term overview of how the revaluation process works in practice.

Summary

Provided:

1/ the projected operational expenditure cuts are sustainable, AND
2/ interest rates do not rise that much AND
3/ 'maybe' generation increases back towards the average projected up to FY2015

I do not see the slashing in value of generation assets that horus is predicting will come to fruition. I feel comfortable as an MCY shareholder that things will be able to continue 'as normal' for a few years yet. Beyond that I should add that Mercury are quite capable of putting up their own solar panels and compete toe to toe with any new generation start ups (including the horus solar co-operative).

SNOOPY

[Joshuatree]

Apologies if you've already covered it Snoopy. How old are the dams, they are getting on arent they?. At some point major works and expense to modernise/strengthen, replace,bring up to today stds etc?

[Jantar]

Hydro dams are built to virtually last forever. There are cement type dams built in the Roman days in Europe that are still working today with no problems.
As for Mercury's dams, the oldest is Arapuni, commissioned in 1929. There were early problems when a crack developed between the dam and the spillway, but this was soon corrected. A new spillway structure was completed around 1990. Karapiro was next, then the main Waikato stations were built during the mid 1950s and early 1960s.

The plant within the dams does require regular refubishment and upgrading. As a general rule the generators have a design life of at least 30 years before refurbisment, and the the turbines are usually at least 50 years before refurbisment. I worked in the Central Waikato power stations from 1979 to 1992, and the first major upgardes were just starting as I left that area to move to the South Island. With 8 stations on the river, and around 2 years to complete each, it will be a continuous programme to work through them

[Joshuatree]

We are fortunate to have your in depth knowledge and understanding on the threads Jantar. Thanks again.

[Snoopy]

Apologies if you've already covered it Snoopy. How old are the dams, they are getting on arent they?. At some point major works and expense to modernise/strengthen, replace,bring up to today stds etc?

Joshuatree, there is a distinction made between CAPEX for operating expenses and CAPEX for generation development. I refer you to page 16 of the November 2017 investor presentation.

http://issuu.com/mercurynz/docs/inve...54184/55357570

You will see that 'operating expenses' (stay in business expenses) are projected to be flat at about $200m for FY2014 and all subsequent years. The operational expenses are what count towards the book asset valuations. One off upgrades, the kind of thing that you are talking about, do have an effect but are depreciated over many years. What is more this depreciation can be undone with subsequent asset revaluations. So there is an argument to be made that such assets do not depreciate at all.

Mercury has an extensive plan to modernize their dams. This program started in earnest in FY2017. But reading between the lines, any modernization is to the electrical systems and the turbine generator mechanics of the dam. The civil structure, the concrete dam that you see, is not down to be upgraded. Technically the dams (concrete structures) have to be depreciated to fall in line with accounting standards. In practice their life may extend to hundreds of years. So I don't think there is much to fear in hidden very high expenses to come from rebuilding the dams. Some things really are built to last!

If you look at the reference I gave you above, the generation Capex is now much reduced from the FY2008 to FY2013 period. It looks to be of the order of $100m in FY2017 and FY2018. The biggest 'threat' om the horizon is if Mercury decide to construct a whole new power station (probably geothermal or wind). But although it would be expensive, such a project will increase the earnings capacity of the company. That's as I see things anyway.

SNOOPY

[Joshuatree]

Many thanks too to you Snoopy your analysis is without peer (even if i dont get it all

[Dassets]

IFRS requires assets to be assessed for value in use essentially. Directors are required to make a decision and will typically assess this by measuring cashflow from a cash generating unit. Historical cost only provides a start point to either write up or down. That move is reflected in PPE. It is actually quite simple. The main drivers for value move are revenue, cash costs and discount rates. Other factors include time the assets can generate revenue, capital items like refurbishment etc. Depreciation doesnt come into it btw from memory as it is not cash and you are valuing the cash generating capacity.

[Snoopy]

IFRS requires assets to be assessed for value in use essentially. Directors are required to make a decision and will typically assess this by measuring cashflow from a cash generating unit. Historical cost only provides a start point to either write up or down.

I understand that historical cost has no direct effect on future cashflows. But neither does historical cost reflect current balance sheet values. I would say it is actually current balance sheet values that provide the start point to either write up or down, In a similar vane, current balance sheet values, because they were determined up to a year ago, do not necessarily reflect future cashflows. And that is the reason that Mercury assesses these asset values annually.

That move is reflected in PPE. It is actually quite simple. The main drivers for value move are:

1/ revenue,
2/ cash costs and
3/ discount rates. Other factors include
4/time the assets can generate revenue,
5/ capital items like refurbishment etc.

'Quite simple' ? Yet you have listed five factors (for a start) that go to making up the picture in your 'simple' summary! I accept that this kind of calculation is a 'straightforward' exercise if you are an accountant that deals with these things all the time. But simple? Your brain must be working on overclock Dassets.

Depreciation doesn't come into it btw from memory as it is not cash and you are valuing the cash generating capacity.

Thanks, that makes sense. I do find it hard with the concept of separating the 'operating cashflow' from the 'investing cashflow' from an overall investment perspective though. I am saying this because often the 'investments' become key operational assets in the not so distant future. To give an example from the Mercury universe.:

1/ Suppose Mercury update an existing dam. In the course of upgrading to the latest technology as part of a planned maintenance, the power output is able to be raised by 5% for the same input say 20MW.
2/ Yet if Mercury were to commission a brand new 20MW hydro power station that would be classed as a new investment, and not part of operational cashflow.

The net result of either 1 or 2 is the same from a power making perspective, yet accounting standards demand each scenario is treated differently. I don't get it.

It is a similar dilemma neglecting depreciation in a cashflow analysis. Granted, I understand that depreciation is not a cash item. But before an asset came to be depreciated, it was bought with real cash once. Leaving out depreciation from a cashflow analysis I see as a time trick, that leaves out historical cashflow by selecting an analysis start point to deliberately leave out historical cash flows.

SNOOPY

[Dassets]

You just make an assumption on what the price curve looks like. Not too hard. Apply nodal discounts etc if any. Costs can be forecast. Let me cover the historical cost point. It only provides the start position. Then the value assessment takes over. Eg I build a $100m plant in 2017. In 2018 the price curve falls. I them reassess my expectation of future prices. Costs dont change. I them record a loss as a result of assessment and mark the plant down to $50m. A loss of 50m. Assessment is not done continously. In reality the forward curve doesnt change much either.