South Australia, the second cheapest electricity in Australia … or the most expensive?

Previous post detailed electricity prices of four states in Australia, coming to the conclusion that the electricity prices of South Australia were without any doubt the highest of those four states. In the meanwhile, I came across several other sources that also claimed that South Australia has the highest electricity price for consumers in Australia.

Now imagine my surprise reading the last paragraph of the Wikipedia article on “Energy in Australia” (my emphasis):

It was claimed in 2017 that South Australia had the most expensive electricity in the world [50] Another analysis claimed that South Australia has the second cheapest electricity in Australia.[51]

South Australia, the second cheapest electricity in Australia?!?!

In my limited dataset with only four states, South Australia had -by far- the highest average electricity price for consumers. This means that South Australia is at best the fourth cheapest in Australia (if all other not listed states were more expensive). How does this “second cheapest electricity” claim square with being at best the fourth most expensive, possibly even the most expensive?

The link goes to an article on SolarQuotes, titled South Australia Is Actually Second Cheapest State For Household Electricity and written in 2016 by Ronald Brakels. The article is a rebuttal of the claim that “South Australia has the most expensive electricity” made by Malcolm Turnbull back then. Brakels says that this is not true and therefor also disagrees with the assertion that South Australia has the highest electricity price because of its high share of renewables. He even claims the contrary: according to his calculations, it is Victoria (a state with a large share of coal power) that had by far the most expensive electricity and South Australia (with a large share of renewables) is the second cheapest.

Pivotal to his argument is the distinction between the average price of electricity and the price that the typical consumer pays for electricity. I can somehow understand what he is trying to do here. Averages are not necessarily relevant. For example, a body of water that is on average a few inches deep, can at places be deep enough to drown. If typical consumers get their electricity cheap, then generally consumers in that state have cheap energy, no matter what the average is.

But then, is this playing here too? When a state has for example a high average electricity price and the “typical” consumer gets electricity at a low price, then the rest of the consumers need to pay an insane amount for their electricity in order for that state to have a high average price. Unless there are extreme differences in electricity prices between the typical users and the rest, I expect the price that the typical consumer pays for electricity will stay in the neighborhood of the average price of that state. In other words, I don’t expect the typical price to be that far off the average price.

Let’s now see how he arrives at South Australia being the second cheapest and Victoria by far the most expensive. He starts from the “representative consumers”. The profile of these representative consumers is defined in the report Residential Electricity Price Trends and in that report the representative consumers of South Australia have a lower electricity consumption than the representative consumers of most other states/territories. South Australia (SA) has the second lowest demand of all states and territories, just behind Victoria (VIC):

Figure 1. Annual grid electricity consumption for typical households graph from the article

Brakels than looked up what the annual bill would be for those representative consumers. He did this by comparing what consumers with such annual consumption would have to pay under the Maximiser plan from Origin Energy in those states and territories. This is the result of his comparison:

Figure 2. Typical annual electricity bill graph from the article

Finally, he takes the total price of the typical annual electricity bill (AU$) and divides it by the consumption (kWh). This gives him the price of electricity per kilowatt-hour for these representative consumers:

Figure 3. Cost of electricity per kWh graph from the article

The first thing that caught my attention was that South Australia is in the fourth position, after ACT (Australian Capital Territory), NSW (New South Wales) and NT (Northern Territory). So, how could South Australia then be the second cheapest when his own calculations puts it in the fourth position?

I think it is just semantics. On a Governmental level, Northern Territory and Australian Capital Territory are “territories”, not “states”. Therefor by arguing that South Australia is actually the “second cheapest state” allows him to exclude the two territories in the top 4 and advance two places in the final result… If he had just followed the data from the datasets he used, he would have to say that South Australia is the “fourth cheapest” and that definitely isn’t as catchy as being the “second cheapest”.

From that graph, it is indeed clear that Victoria had the most expensive electricity back then. Not only that, it rises well above the rest. This seems odd, in all the comparisons of the average electricity prices that I found earlier, Victoria ranked somewhere on the higher end of the middle, while South Australia always stood head and shoulder above the rest. Yet, in Brakels calculations, Victoria has an incredibly high electricity price for its representative consumers and South Australia is somewhere on the lower end of the middle…

How is it possible that the representative consumers in Victoria paid such a high electricity price? Looking at the data in the article, Victoria had the lowest electricity consumption (first graph) and the lowest annual bill (second graph). It is of course possible that this low annual bill is nevertheless relatively high compared to the annual consumption. This is the calculation that Brakels made: the representative consumer in Victoria consumes 4,026 kWh per year and has an electricity bill of 1,426 AU$ per year, therefor it has a calculated price of 1,426 / 4,026 = 0.354 AU$/kWh or 35.4 AU$ cent/kWh.

That is mathematically correct, but … 35.4 cent per kWh? That seems extremely high.

I was puzzled by the discrepancy between the average and typical prices of consumers in Victoria and South Australia … until I started to read the “Residential Electricity Price Trends” report (the report that he used as the source of the consumption data for the representative consumers). I indeed found a graph with the trend of the residential electricity price in that report:

Figure 4. Residential electricity price trends graph from the report

Weirdly enough, those prices were also for the same “representative consumers” having the consumption that was used in the article, but those prices looked nothing like the prices that were used in the article… Also, it is clear that the residential electricity price of South Australia is not only the highest, it is also well above the rest. Exactly where I expected it to be. The Victoria electricity price is on the higher end of the middle, also exactly where I expected it to be.

This was the point where I started to realize that Brakels rather selectively used data from this report. This report did not only provide the annual electricity consumption of those representative consumers (these values were used in the article to create the first graph), but also their corresponding annual electricity bill (these values were not used, instead were replaced by the Maximiser plan data). Even the price per kWh for those representative consumers (based on the the annual consumption and annual bill data of the representative consumers) was provided in that report, yet for some reason he didn’t use it in his article…

So, why did Brakels go to such great lengths to gather that annual bill data from Origin Energy? He could have saved the effort and just could have taken the price per kWh for those representative consumers straight from the report! No calculations needed whatsoever and only one graph to create.

If he would have created that graph, this would be the result:

Figure 5. Cost of electricity per kWh graph with the data from the report

That is quite a difference with the graph that was shown in the article. Victoria lost its position of the most expensive state/territory and it is replaced by South Australia that is now head and shoulder above the rest. Which is in line with the average electricity price data that shows the same pattern.

Now everything becomes clear. The conclusion that South Australia has the second cheapest electricity in Australia is in fact an artifact of the replacement of the annual bill data from the report, combined with the use of semantics to eliminate two inconvenient data points.

Apply both tricks and South Australia will have the second cheapest electricity of all states (excluding the territories).

Don’t apply those tricks and South Australia will have the most expensive electricity of all states and territories.

8 thoughts on “South Australia, the second cheapest electricity in Australia … or the most expensive?

  1. vuurklip

    Brings to mind Mark Twain: “There are three kinds of lies: lies, damned lies, and statistics.” (Popularised by Mark Twain if not originally by him).

    I also remember how the university I worked for used some dubious number juggling to justify new (expensive) accounting software – fiddling with assumptions showing a favourable cost/benefit analysis.

    Liked by 1 person

  2. oiltranslator

    I am in the southern hemisphere about even with South Australia. Where I used to live is about 150 miles north of Brownville, Texas, but suppose Brownsville, where the average is 61ºF. Spring, summer and fall you broil and it takes $300 a month to air-condition the house down to 80ºF. No idea how they got that average temp. In Curitiba, 1 km above sea level, the average is also 61ºF and it is about that all year. Sometimes it gets up to 80 for a day or three, or down to 40ºF. Nobody uses air conditioning in home or car, and the average matches the evidence of the senses.


    1. trustyetverify Post author

      You are right that averages can sometimes be deceiving, but this was not at issue here. It was the discrepancy between the average prices and the representative prices of the states South Australia and Victoria that led me to look closer into the methodology used in the article. Figuring out in the process that the conclusion of the author was fully dependent on the combination of 1) the use of semantics and 2) a selective use of data from that report…


  3. Chris Morris

    One of the hidden factors that may be lowering the “average” SA electricity consumption is the high level of domestic solar cells, particularly in Adelaide, the only large centre of population. It is supposed to be around 40%. That is how in can generate these self-congratulatory headlines.
    No mention of what happens in the other 8759 hours. So all the back up generation (and its costs) have to paid for by the average consumer making their power more expensive. And they don’t get rebates, like those in NSW and Vic do.


  4. Pingback: Pumped hydro more expensive than batteries: the intro – Climate-

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s