Poles and wires: Consumers using less, paying more

Prime minister Malcolm Turnbull should be talking to Paula Conboy and John Pierce, the head of the AER and the AEMC respectively as well as network owners Singapore Power, China State Grid, CKI, an Aussie Super, as well as the gentailers he plans to talk to about energy prices this week.

In the end under-supply of generation leading to high prices in the wholesale markets will fix itself. The fix may be painful but if generation prices are high enough for long enough there will be new supply. Nothing is more certain. But network prices, the biggest driver of higher electricity bills, are unlikely to ever fix themselves.

The regulatory system guarantees the network business a return on capital for the next 100-200 years, no matter how much volume falls networks get a fixed amount of revenue. Like the NBN, this is not an easy problem to fix. Until it is fixed though expected to continue paying over the odds for electricity if it comes from the grid.

Electricity distribution in the NEM – one of the world’s best businesses

The electricity  network distribution business (excluding transmission) in the NEM have a regulated asset base [RAB] of around $70 billion and  their market value is around $100 billion. So long as the distribution businesses are valued by the market at 30-50% more than the regulator values, then it’s prima facie evidence that returns in the sector are too high

However, more than that, it’s very difficult to see consumption per point of presence going back to the levels of 10 years ago and at the same time the  RAB grows 1-4% a year. This means consumers pay “more for less”.  We  can see this with two examples.

One is in Victoria where we use United Energy.

The data we show is for the household sector, but the medium and even large volume consumer sectors in this distribution franchise show similar trends. Remember, this is just the distribution cost, and it doesn’t include transmission, let alone all the generation and retail factors.

But we are focusing on distribution because it’s our view that, as they are legally entitled to do, owners of these businesses are making risk-free profits paid for by consumers and we can’t see under the current rules how this will ever change.

Figure 1: United Energy household sector annual consumption and unit price. Source Company
Figure 1: United Energy household sector annual consumption and unit price.

That’s enough of a worry, but what happens when you get to Sydney – and the Augrid network – where there is more unit development (more energy efficient), much more solar PV, and much higher utility costs passed through to consumers>

Volumes per house are down a staggering 22 per cent, so a gold star for Sydney-siders: but, their reward has been a 127 per cent increase in the unit cost of electricity.

Figure 2 Ausgrid household sector volume and unit price. Source: RIN statements
Figure 2: Ausgrid household sector volume and unit price. Source: RIN statements

Note that we have adjusted 2016 to be flat with 2015. That’s our estimate following the recent Federal Court Case comprehensively won by the networks.

The regulatory system requires higher prices when volumes fall

Each regulated asset entity is allowed to recover a fixed amount of revenue each year. If volumes fall then prices rise. If volumes rise then prices fall.

This system works, more or less to everyone’s satisfaction when volumes are rising. The airline industry is an example, or the computer industry or many others. It brings nothing but misery when volumes fall.

Message 1:   Trends that have delivered the fall in consumption per household are likely to continue.

At least for a while. These trends are:

  • Higher prices. Consumers naturally react to higher prices by consuming less.
  • Energy efficiency. Fridges, swimming pool pumps, double glazing, insulation, LED lightbulbs. We are only part way through that process.
  • More behind the meter PV and batteries.
  • Household sizes are getting smaller.

Under the present system if consumption per house continues to fall prices have to go up.

Message 2. The system provides no explicit signal to the regulated companies to stop investing. It’s true that these businesses are run by extremely sensible people who will act appropriately, more or less.  There is clear evidence in numbers below of a decline in augmentation capex. Still, there is a financial incentive to invest.

Every $1 invested in RAB is valued by the market at $1.30-$1.60. Since $0.75 of that $1 invested is financed by debt the equity owner invests $0.25 and its valued $0.55 -$0.85

Figure 3 EV:RAB. Source: KPMG report to Duet
Figure 3 EV:RAB. Source: KPMG report to Duet

Note that EV = Enterprise value = Market value of equity (adjusted for minorities) + value of debt = The total cost of buying the business

The licenses these network businesses have are for 100 to 200 years.

Is it time for a fresh estimate of RAB?

The only way that prices can decline is if volumes pick up, returns on capital fall or the capital base is cut. Things like cutting opex just wont do it other than in NSW.

We simply note here that when the electricity sector was “created” following the Hilmer reforms of the early 1990s and the Initital Capital Base [ICB] of the wires and poles had to be considered. This was calculated using DORCs (surely invented for the engineering fraternity) = Depreciated Optimised Replacement Cost.

In other words an engineer went out and worked out what it would cost to replace the network in its depreciated state. This was of course a fairly wooly estimate and the various State Goverments then went around making arbitrary adjustments to suit their own purposes.  The ICBs were then written into law but nevertheless perhaps its time for a fresh look.

Where the money goes

We put together a small data base of the electricity distributions businesses across the NEM. The sources for the data were firstly the CSIRO/ENA technical modelling report

Brimsead CSIRO/ENA Network transformation economic benefits

And then the AER’s “final decision” reports not all of which are fully adjusted for appeals but are arguably still fit for this purpose.  For the NSW distributors we basically used their “revised final proposals” which involved skimming through 100s of pages to find a few tables.

As an aside we note that Paula Conboy recently said that in 2014 or 2015 there were about 100,000 pages of regulatory submissions!!! No wonder you need new glasses!

From this data we have complied a NEM wide aggregate annual revenue and capital expenditure view. That is we added up the revenues and capital expenditure forecasts/allowances of each regulated entity in the model.  Annual revenue is about $11 bn, of % is return on capital, 11% depreciation, 30% opex and 1% other.

Figure 4 NEM wide aggregate electricity revenue and capex. Source:AER and company docs, ITK estimates
Figure 4: NEM wide aggregate electricity revenue and capex. Source:AER and company docs, ITK estimates

What we see is that most of the revenue is going to the capital providers and most of the capex is going on maintenance. Maintenance way exceeds depreciation as calculated in this table.

Note that these numbers could have errors. The supporting tables are below but this sort of work benefits from a team approach.

Other data shows that non coincident max demand is basically flat. We expect the new connection capital expenditure to actually decline once the housing boom comes off, and it will. One day…

Still we think that there is arguably still room to cut the return on capital further. Alternatively the RAB needs to be rebased.

$67 billion of electricity distribution assets in the NEM

The following table shows the electricity distribution assets in the NEM.

Figure 5 CSIRO/ENA report
Figure 5: CSIRO/ENA report

These assets deliver about 78% of NEM wide  generation  to about 9.6 m customers of whom 8.5 m or 89% are residential.

Figure 6 Source: CSIRO/ENA technical report
Figure 6.  Source: CSIRO/ENA technical report

The residential customer base in the NEM consumes about 48 TWh or 34% of total consumption at a daily average (grid delivered) 15.5 KWh. The Jemena area households in Victoria are the thriftiest and Tasmanians consume the most on a per household basis.

Figure 7. Source: CSIRO ENA technical report
Figure 7. Source: CSIRO ENA technical report

Over the current regulatory period for this 14 networks they will recover $51 bn of revenue of which 51% is return on capital and 12% is depreciation. Note that our total numbers here have no real meaning as Tas Networks and SA Power don’t have five year time frames. Some years start in July and some in January.

However we have left the numbers in this form as they are those reported by the AER. The exception is the NSW distributors, Ausgrid, Endeavour and Essential where the AER is required to remake substantial parts of the decision and so we have used the network “final proposal” numbers.

Figure 8 Distribution revnue, excludes transmission. Source AER, Company reports ITKe
Figure 8: Distribution revnue, excludes transmission. Source AER, Company reports ITKe

We could ballpark  that of the 51% of revenue allocated to return on capital about 60% represents debt funding and about 40% equity funding.

Now let’s look at the capital expenditure.  What we find is that “only” about 20% of allowed capex is going on new customers and augmentation. By definition then about 80% of capex is going on maintenance and other purposes. But mostly maintenance.

In coming to that conclusion we have assumed that customer contributions are essentially for new connections and so they can be subtracted from the new connections capex as allowed by the AER.

Once again we use the NSW distributor final proposal for their capex.

Capital contributions are reported as revenue, profit and capex under Australian accounting standards but essentially paid up front by the connecting business at cost and don’t go into the RAB. In our view all new connections should be treated the same way but that’s a different part of the story.

Figure 9 Capital expenditure total over regulatory period. Source: AER, Company, ITKe
Figure 9: Capital expenditure total over regulatory period. Source: AER, Company, ITKe

Context

As any analysis will show there are a number of factors that are responsible for the high electricity prices paid by households and businesses in Australia. In no particular order these include:

  • High gas prices and gas is the price setter in some key markets;
  • Low consumption per $ of invested RAB.
  • Supply demand balance that favours supply;
  • Increasing costs of new generation relative to old generation
  • Poor market structure (too much vertical and horizontal integration)

Most of these factors are to an extent self-correcting. Other forms of generation will substitute for gas if the gas cost doesn’t come down and left to itself enough new supply will be brought. However even the market based solutions won’t work properly because of the market structure (the more the structure resembles a monopoly the more likely it is to be undersupplied).

More to the point of this article there is no real answer to the problem that consumption per $ of RAB is low. Essentially that means it costs plenty to get electricity delivered from a central generating unit to its point of consumption.

Many argue, and no doubt with some truth, that there has been gold plating of the networks, but it’s also worth bearing in mind that for maybe 20 years prior to the gold plating period there was significant under investment.

Again we can argue about consumer reliability required standards but there is no doubt that the public in general wants the electricity supply to be reliable. Who after all would want electricity to work like the NBN?

David Leitch is principal of ITK. He was formerly a Utility Analyst for leading investment banks over the past 30 years. The views expressed are his own. Please note our new section, Energy Markets, which will include analysis from Leitch on the energy markets and broader energy issues. And also note our live generation widget, and the APVI solar contribution. 

David Leitch is a regular contributor to Renew Economy and co-host of the weekly Energy Insiders Podcast. He is principal at ITK, specialising in analysis of electricity, gas and decarbonisation drawn from 33 years experience in stockbroking research & analysis for UBS, JPMorgan and predecessor firms.

Comments

10 responses to “Poles and wires: Consumers using less, paying more”

  1. David leitch Avatar
    David leitch

    I’d like to apologise for a few more than usual typos in this note. I’m going to blame a head cold…

    1. nakedChimp Avatar
      nakedChimp

      Still a nice piece of work.

  2. solarguy Avatar
    solarguy

    David, at the time of me writing this, Roof Top Solar in NSW is generating 321MW, so that would be considered volume, yes or no?

    1. David leitch Avatar
      David leitch

      Only that part exported to the grid

      1. solarguy Avatar
        solarguy

        I know your not feeling well and sorry about that, hope your better soon. It’s only what is exported back to the grid that is counted in the data, as behind the meter isn’t, or can’t be counted. Surely that’s correct?

        1. David leitch Avatar
          David leitch

          Yes. But these days the aemo report gross rooftop solar in their production data but not in demand

  3. Peter F Avatar
    Peter F

    Tremendous work as usual David
    However you should probably highlight the point that the 52% of revenue being return on capital or return on equity. If the capital base were high enough 52% of revenue might possibly represent only 5% ROI.

    However as you say the market is valuing every dollar of investment at $1.30 and as debt is debt that means that 30c of equity is valued at 60c.

    However the situation is actually worse than that, because the ultimate holding company itself may have 70% or more gearing and there is no requirement for the T&D operator to have 30% equity so you might find a holding company with 70% gearing investing in a distributor with 80% gearing, so the equity in A$1 of capital may only 30% of 20% i.e. 6%.
    Therefore for every dollar they can increase their asset base by, brings $1.30 rise in value less 94c in debt i.e 36c value for 6c increase in equity, a powerful incentive to over-invest

    1. nakedChimp Avatar
      nakedChimp

      Why would that help?
      In what timeframe does the consumer ‘pay’ for the asset base increase with ‘equity’?
      At first they have to lend the capital (gear up) and only later try to recoup that from the consumer.
      By goldplating they made the situation worse – only the capital providers look to walk away better in that scenario.
      I’m pretty sure the ones having profiteered the most from this aren’t the ones who will be found holding the bag in the end – though the only ones able to take over would be pension funds or the tax payer.

    2. James Hansen Avatar
      James Hansen

      One anomaly, which is hidden, is the way they include new office buildings as “essential” capital expenditure and subject to the guaranteed 10% return on capital. This is grossly excessive!

  4. Peter G Avatar
    Peter G

    Thanks for the Article David.

    The T&D companies interest on debt is also part of its regulated ‘cost’ base. But actually this is not the real ‘cost’, as regulations ‘generously’ deem the interest rate ‘cost’ higher than the interest actually paid!

    Given the yields, and regulated risk free returns, in this industry there is a good argument that interest ‘cost’ should be deemed at lower than market rate – or disallowed completely (to take a leaf from Warren Buffet’s approach to enterprise valuation).

Get up to 3 quotes from pre-vetted solar (and battery) installers.