Levelised Cost of Electricity: What is this?

Levelised Cost of Electricity, or LCOE is a means of comparing generation technologies, by considering the cost of the electricity that comes out over its lifetime. Simply put it is the lifetime sum of all the costs; construction, planning, maintenance, land purchase, waste disposal, pollution charges, mining, divided by the amount of electricity produced during it’s lifetime. Choosing some non-indicative numbers, if you spend $2000 installing solar panels, and they generate 4000kWh over their life time, your LCOE is 50c/kWh.

This graphic from BREE (Bureau of Resource and Energy Economics) paints the very complicated picture of LCOE comparisons in Australia.

It is a pretty coarse metric, but probably the best one for comparing costs of generating technologies. But, as I tell anyone who will listen, the key to performance metrics, KPIs, benchmarking activities, is knowing the assumptions on which the metric is based. In small samples this means knowing the ways in which the situations you are comparing differ, so you can gain an understanding of how those differences drive change in the metric. It is not a stand-alone decision making tool, but a guide. Note also that there is no consideration for when the electricity is produced; there is no modifier for a gigantic baseload plant that makes 1000MW whether you want it or not, nor for solar output that doesn’t match demand. It is a huge, slightly dumb aggregate. But not without uses.

To illustrate with an example that is based on made-up numbers; LCOE calculations show that Hazelwood (Victorian brown coal) can make electricity at $10/MWh and Bayswater (NSW black coal) can make it at $20/MWh. Does this mean electricity is cheaper in Victoria? No. Does this mean that electricity from brown coal is cheaper? Not really. A bit of research finds that brown coal is usually closer to the surface and that Hazelwood has a massive deposit right next door. Some of the NSW plants rail their coal in, but it’s such high quality it is worth it. The main thing to take from this is that if there are differences, go looking for the reasons. Don’t assume that means X is cheaper than Y.

Most of the time when LCOE is used in public discussion it is as a stick to beat renewables with, when compared to our current technology costs. Comments like “sure we could power the country on wind and solar but it would be expensive. Look at the predicted cost and compare it to what we’re paying now” have been common on the bits of the internet that I frequent.

Is this a valid comparison? Consider the assumptions bundled into the costs of both technologies. Start with our current technology.

Most of our coal plants were built by state governments, who also owned the network and approved the network connection, granted land and resource access and determined pollution requirements. There are even examples of state governments encouraging large energy users to their state, and then negotiating the power purchase agreement between this large user and the new plant; aluminium smelters are a typical example. All of these items would fall into the “cost” side of the equation, but don’t count as they were never paid by the power-station. Should they count as costs? To some degree I think yes as there is a decent argument that forgoing profits to one arm of government (the network business) is the same as a cost to taxpayers. But you get loads of cheap MWh so where’s the harm?

Consider a renewables project instead, say a big wind farm. Before the project will secure finance they most likely need a Power Purchase Agreement, an agreement that someone wants their electricity. In the aluminium smelter/coal plant scenario above, government will help negotiate that. The wind farm owners have probably done loads of modelling and have a site organised; they then need to buy the land. I have heard that up to 30% of a wind project cost is in land purchase, but this link, from the EU wind Association puts it closer to 5%. I suspect it varies with location. How much did the coal plant pay for their land? Are they paying a royalty to mine and burn the coal?

In such a simple aggregate (costs/amount of electricity) the way the costs are calculated drives the final metric significantly. For the coal plant, are the construction costs now or at construction dollars? Does it include any of the government subsidies? Does it include the cost of the carbon tax now? Does it include a fee for the sum of emissions to now? And of course the same can be asked of wind; does that include renewable energy certificates? What about storage? Do the emissions of the gas turbines that provide network support for the wind turbines count?

To engage in this conversation you need to understand the shorthand, but also the assumptions implicit in the shorthand. The current hold the fossil fuel burners have on our electricity network is more than anything the benefit of incumbency and the legacy of economic development in the 70s and 80s. Before we knew about Global Warming, these policies of encouraging investment to use our abundant energy made sense. Not any more though, and the conversation needs to move to how to unravel them, not how to work around them.

As a renewables advocate, I actually favour a flatter playing field in the electricity market, not more subsidies. While renewables are benefiting now from the economic stimulus of RECs, I suspect that summed and normalised for the value of money over time, fossil fuel burners in Australia have received considerably more subsidies and allowances. This view is supported by research done in the US, explained here which found nuclear and fossil burners received substantial subsidies when compared to the federal budget, but mostly in the past. The American example is not necessarily a great pointer for Australia, as they have domestic oil production, but also the inextricable link between military and nuclear spending, which may inflate those figures. The international picture is broadly similar, as discussed in the International Energy Agency’s “World Energy Outlook”, summarised here. I have seen research on Australian fossil fuel subsidies, but it’s a bit out of date now, and it didn’t have a comparison with renewables.

I am confident that it is technically feasible to meet Australia’s stationary energy needs with renewables, and a good deal of transport energy could be met with renewables too. Currently though, the barriers to implementation aren’t so much in the technology or the cost, but the structure of the grid and the market they are trying to compete with. What would the network look like if the fossil burners started paying for their pollution? Renewables were given the same connection advantages as the fossils? If people could buy as many solar panels as they wanted and the network can compete to provide them electricity in the down times, rather than policies that encourage energy use from baseload plants to try and keep them cheap.

I think I might try and find out. I am strongly considering a research project that attempts to answer this question; What has the total cost of fossil fuel been to Australian taxpayers, and what would renewables cost if they were given the same subsidies? It’s a big question, and I might not like the answer and it probably involves a lot of fairly courageous assumptions. I’ll continue pondering. Any ideas on how one would complete such a project would be appreciated in the comments.



About evcricket

Extreme gardener, engineer and bird nerd. View all posts by evcricket

3 responses to “Levelised Cost of Electricity: What is this?

  • Sean

    I wonder how challenging getting all the subsidy in kind costs would be? There must be accounting / valuation principles that can be applied. The externalities are a bugger, car plants & Aluminium smelters being obvious co-location opportunities/costs. Maybe you can convince an economist to provide some modelling on realistic alternative use cases.

    Bloody good idea though

  • Scott

    Sounds like a great project! I would be very interested in reading it

  • Emma

    Sounds like a great project. You could maybe start by looking at the methodology used in the US study you linked to. That looks to calculate the proportion of GDP spent on coal, gas, nuclear in their early years. In some ways that would be neater than looking at total subsidy, at least for purposes of comparison with subsidies to renewables, which just haven’t been in place that long.

Leave a Reply

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

WordPress.com Logo

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

Google+ photo

You are commenting using your Google+ 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

%d bloggers like this: