Where did all the energy demand peaks go?

It has been estimated that around one quarter of the revenue earned by Australia’s electricity generators – and nearly all their profits – come from a short period – between 36 and 100 hours – of “super peaks” of demand that push wholesale electricity prices beyond $5,000 per megawatt hour or more. But what if those peaks never came?

We’re about to find out, because this graph below published by the Australian Energy Regulator reveals that there were virtually no peaks above $5,000/MWh in 2011/12. The graph tells the story for previous years.

The returns for Australia’s generators for this period have not yet been announced, but as we saw in the December half with Loy Yang A’s plunge into the red (before its purchase and refinancing by AGL Energy) those generators with high debt levels will be suffering badly. Those generators owned by vertically integrated companies can take it on the chin and offset it with higher earnings from their retailing operations.

Sadly, the record low wholesale costs being recorded this year in the NEM are not passed on to consumers. At least not yet.

And what is the reason for the disappearing peaks? For that we can go to the Australian Energy Market Operator report and find that milder weather (fewer extreme hot days when everyone flicks on their air conditioners at the same time), reduced demand from manufacturers, more efficient and cautious use by commercial and residential users in the face of higher retail prices, and the growing impact of rooftop solar (its deployment trebled in the past year). Welcome to the merit order effect.

Comments

13 responses to “Where did all the energy demand peaks go?”

  1. Paul McArdle Avatar

    Giles and others,

    For a longer list of reasons why demand is declining, see here:
    http://www.wattclarity.com.au/2011/10/some-possible-reasons-why-demand-is-declining/

    In the AEMO 2012 Forecasting Report, they attribute only 1,702 GWh (or 0.9% of annual energy – or about 200MW average load over the year) to rooftop solar PV in 2011-12. See p3-1 in this report:
    http://www.aemo.com.au/en/Electricity/Forecasting/~/media/Files/Other/forecasting/2012_National_Electricity_Forecasting_Report%20pdf.ashx

    See this article for how the growth rate in average demand has changed over the past 12 years:
    http://www.wattclarity.com.au/2011/09/how-is-demand-changing-in-the-nem/
    Extrapolating from the growth rate for 2006 or so, we see that it could have been as high as 3000-4000MW more, but has not been the case.

    In this light 200MW (from solar PV) is not inconsequential, but is still small.

    The more significant factors appear to be such things as:
    1) Industrial consumption changes (as a result of GFC and higher prices); and
    2) Energy efficiency in homes and commercial enterprises – plus response to higher prices.

    Hence, whilst solar PV is playing a role, it’s important to not overstate the case.

    1. Giles Parkinson Avatar
      Giles Parkinson

      Agreed. I think AEMO said solar PV accounted for 0.7% of NEM demand reduction in last year. The BIG question is what happens with 10 times that amount of solar, and with battery or other storage. Check out our site tomorrow for a ground breaking story on solar/storage hybrid in Australia.

      1. Paul McArdle Avatar

        Hi Giles

        Curious to know where your forecast of “10 times that amount of solar” comes from.

        Most of the forecasts I have been reading (e.g. the AEMO report I linked above) have suggested it might grow 3-4 times the amount in 2011-12 by 2020-21.

        e.g. p3-1 in the report suggests “By 2021–22, this is forecast to increase to 7,558 GWh or 3.4% of annual energy.”

        If it was going to be 10 times, this would indicate that the LRET target might be wound back even more significantly under current moves to keep a lid on cost increases across the board.

        Paul

        1. Giles Parkinson Avatar
          Giles Parkinson

          I think you will find that your number is their “moderate” demand number. I suspect it will be higher than that, as most official forecasts have grossly underestimated solar PV deployment in recent years -bot here and overseas. So i’ve gone with their high growth forecasts, which are still below private consultant estimates (18GW solar PV by 2022 vs 1.4GW now), and neither they nor the AEMO numbers include utility scale PV (add in another 5GW or so).

          1. Paul McArdle Avatar

            Thanks Giles

            So with reference to the AEMO “Rooftop PV Information paper”
            http://www.aemo.com.au/en/Electricity/Forecasting/~/media/Files/Other/forecasting/Rooftop_PV_Information_Paper_20_June_2012.ashx
            it looks like the 10x assumption is based on both:
            1) the rapid uptake scenario; and
            2) projecting out to 2030 (further than my assumption of 2020).

            This explains why your multiple was so much more than what I was working with.

            What private consultant reports do you refer to, and are they publicly accessible?

            Paul

          2. Giles Parkinson Avatar
            Giles Parkinson

            Sunwiz’ is the most prominent, and worked closely with AEMO, although AEMO took an average of all the different reports. We wrote on Sunwiz and SBS here https://reneweconomy.wpengine.com/2012/18gw-of-solar-by-2022-that-depends-on-whos-connected-45308

          3. Paul McArdle Avatar

            Thanks Giles,

            Just so I am clear – the SunWiz and SBS report you refer to is this one, right?
            http://www.sunwiz.com.au/index.php/solar-market-intelligence/market-forecast.html

            Paul

  2. Rob Avatar
    Rob

    That has got to hurt!

    We’ll see a few peaks again when El Nino comes back. The first year of El Nino will just dry everything out and not cause too many peaks. The second year we (in the East) will bake when a westerly wind comes over the desert. Then the aircon will all run at once. Worse all the fire fuel build up during our extreme wet will be tinder try and make things dangerous.

    Apparently a return to El Nino is forecast this year. It doesn’t always happen on cue, but I think we should expect it will come soon enough.

    1. Tim Avatar
      Tim

      Ironically, I think extreme weather shows up as evidence of climate change in peoples’ minds (whether it’s justified or not).

      I have no stats to back up my hunch though.

  3. Warwick Avatar
    Warwick

    Giles,

    Just a few points need to be clarified in this article.

    1) There looks to be confusion between “intervals” and hours. The generators’ dispatch interval is 5 minutes. I suspect your chart refers to 5 minute intervals not hourly prices.

    2) If you define superpeaks as hours >$5,000/MWh, then there would be very few active periods in every year. The market cap (VOLL) is over $12,500/MWh but was only $5,000/MWh until July 1, 2004. This would mean that your charts in the first few years would be zero as the market price was never above $5,000/MWh.

    3) Again, if you define it “normal” to have 36 to 100 hours per year of prices above $5,000, then the minimum annual price for 36 hours at $5,000/MWh would add $189/MWh to the annual price which is many times what it has averaged for any full year for any state in the national electricity market. It is not possible for the 100 hours at or above $5,000/MWh in a year as it would violate the market’s cumulative price threshold (CPT) (refer to AEMO market operations for more detail on this).

    4) Contract prices are not at record lows. As an example, annual contract prices were as low as $29/MWh in Queensland with the introduction of new coal fired plant in 2001/2. QLD annual contract prices have more than doubled since then.

    5) Spot prices have trebled since July 1 with many regions above $90/MWh as at this evening. What happened to the “Merit-Order Effect”?

  4. Russell Harris Avatar
    Russell Harris

    I think it’s also worth noting that these are spot market prices. Most (about 80% I understand) of total electricity sales are contained in long-term hedging agreements. While I’m not disputing the impact of MOE on the spot market, generators and retailers will simply realign their positions to have more or less exposure to the spot market.

  5. Ronald Watts Avatar

    All very interesting stuff. Another factor in the decline of peaks is network enhancements: when network capacity grows, generators remote from demand can elbow out peaking plants. I don’t know how significant this could be, but as demand falls even further behind capacity, it must be happen.

    On rooftop solar: with sub $1/W panel prices now advertised here and finance available, there should be some strong take-up. Batteries are getting better (US Energy Secretary Steven Chu claims “we are on track to reduce the battery cost for a hybrid car from $12,000 to $3,600 by 2015”). If more than, say, 20% of people in a given area go off-grid, the local distributor must raise prices for the remainder. This in turn drives more off-grid, …etc. A virtuous circle, but not if you’re an investor in networks. Perhaps O’Farrell should think about selling off those networks sooner rather than later?

  6. Matt Avatar
    Matt

    As you mention in the article, another very important factor in the disappearance of price spikes has been two La Ninas in a row during which there has been vastly reduced use of air-conditioning. The demand spikes generally occur on blistering summer days, and most of the eastern seaboard has had few of those over the past two summers.

    I’m not suggesting that efficiency, declining manufacturing and so on are not contributing, but surely they’d contribute more to an overall decline than an impact on extreme price spikes (although of course the former would slightly affect the latter).

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