Can you design a better time-of-use tariff than Chris Bowen’s Solar Sharer?

A couple of years ago I wrote an article about making renewables cheap. Maybe our energy minister, Chris Bowen was listening because this week he announced a Solar Sharer scheme offering three hours free solar to households.

The two key areas of pricing reform that I highlighted in my original article were:

– Cheap times when renewables are in surplus and our usage patterns need to change so we can soak them up before they are wasted, and;

– Expensive times when we deploy dispatchable generation and it is necessarily expensive because much of the rest of the year these generators spend their time unemployed.

Yes, I am a fan of time-of-use tariffs.

In this article I will explore cheap and expensive times in more detail. If you are more of a numbers person, skip to the end and download my tariff calculator to explore pricing options yourself.

Price signals

Some have suggested the Solar Sharer Scheme is a smoke and mirrors exercise where people save a few hundred per year during the middle of the day, only to find their bills rising by a few hundred somewhere else.

This is a valid concern, so I want to start this article with an assertion about what a good price signal can achieve.

If energy is priced well, it will inspire the user to consumer energy or not, and when they do, the seller will be no worse off. Perfect price signals are difficult to achieve but costs known as the short run marginal cost (SRMC) come close.

The argument used to go like this: “if you use a kilowatt-hour (kWh), I put more coal in my boiler and it costs me 4 cents, so I will charge you 4 cents. If you don’t use that kWh, I will save the coal for the next customer.”

You can instantly see the problem with transferring this method of pricing to surplus solar energy. If the user doesn’t use it when it is available, it won’t be there later. The short run marginal cost of unused renewable energy is zero, so “free” might be a sensible price signal to send to energy users.

In the energy market, we have short run marginal costs varying throughout every day and season. For much of the year, these costs are used to set the market dispatch price. 

But energy has other costs. It is capital intensive, leading to long run marginal costs (LRMC) plus there are also a host of business costs and expectation of profits that aren’t necessarily linked to the quantum of energy sold at all. The latter are known as residual costs.

Our electricity networks are capital intensive too. But the LRMC published for these businesses only account for about half their costs. A large proportion of network costs are considered residual costs.  

Our retail businesses invest in staff and systems that scale by customer rather than scaling by energy consumption. These can be passed onto customers as fixed, per-customer costs.

All of these considerations can be turned into prices but not always into efficient price signals.

Where possible, price signals should indicate to the customer the cost of producing a kWh in the short term, and the cost of building the generators and network in the long term.

It is worth repeating that the best price signals help both the customer and the supplier. The customer chooses whether to use electricity and the supplier chooses whether to produce it or constrain it, and neither are worse off.

Cheap generation

We already have a surfeit of renewable energy at some times and this is when it should be offered cheaply. In a state like South Australia, the output of scheduled wind and solar generators is curtailed when there is too much.

The output of rooftop solar is also constrained. Some rooftop solar has been installed with export limits. At some times, protection settings turn solar systems off when the voltages are too high.

But the level of surplus at the moment is tiny compared to the level we expect in future. David Osmond’s modelling hints at 30 per cent surplus with only a third likely to be soaked up by storage. My own experiments with models of 100 per cent renewable energy imagine 6,000 hours of the year when too much, rather than too little renewable energy is the challenge we face.

The cheapest use of surplus renewable energy is to use it directly. Hot water and electric vehicles will be two of the biggest energy users in the future home. Managing these will be about making the timing habitual for people and easy to override on occasions.

Some heating and cooling of our homes can occur when the sun is shining or the wind blowing, especially if we start building more energy efficient and comfortable homes with more thermal mass to hold our homes at a steady temperature.

Turning on appliances like dishwashers and slow cookers in the middle of the day could take off as a new social practice as our society at large learns to tune its energy use to the weather.

Peak load 

At the other end of the spectrum we need ways to signal the real cost of building new dispatchable generation for those times of peak load when the sun isn’t shining and the wind isn’t blowing. Increasingly, these times will occur on a few very hot evenings in summer and some very cold, still winter evenings in winter.

Unless we are successful at encouraging our businesses and industries to reduce demand at these times, the peak will be depend on the  energy use of homes.

That’s the case at the moment. Households respond to hot and cold evenings, using over four times their average load and the whole system experiences a peak.

As we electrify heating and cooking in the colder states, this feature of our system will only become more pronounced. Which is more economic: reducing peak load at the household level or the business level?

This question is best answered by consistent price signals, where either sector can respond.

Our tariffs do not offer a consistent price signal for peak load at the moment. Attempts to introduce demand and capacity charges to households have often failed.

Some industries work with demand charges, others do not. More importantly, we do not have a shared understanding about how the cost of peak load should be allocated to users across time and space.

And we only need a price signal for a short time period. Something like 30 per cent of our capacity is needed for less than 100 hours each year, making it an incredibly expensive investment that is rarely used.

In South Australia the average household contribution to peak load is 1.8 kilowatts (kW) and yet we build 5 kW of network capacity for new dwellings. Across the National Electricity Market (NEM) the residential sector spends $17 billion on electricity each year and over $3 billion can be considered the cost of building only for peak demand.

Allocating capital costs is a choice. Economists have some useful methods, but there are more ways than one to spread the burden of capital costs. At one extreme we can place all of those costs into a single moment. At the other, we can spread the cost across every kilowatt-hour (kWh) we use.

The tariff calculator, which you can download below, recognises the choice needs to be deliberate, and shows us the financial implications for different household types of making peak price signals sharper or blunter.

Residual costs

Residual costs, those that cannot easily be turned into a price signal, are typically spread across times of use when we can be guaranteed that everyone will contribute to these shared costs. Residual costs can be over 30% of the final household bill.

The allocation of these costs is a choice. Economists might argue, for example, that economic activities which are price sensitive to the cost of energy should not pay residual costs because it will lead to a less-than-ideal level of economic production.

By this measure, households who will choose to scrimp on cooking, heating or cooling because they cannot afford energy, should not pay residual costs because the most welfare is served by the use of the energy at a lower price.

Once again, the tariff calculator allows you to explore the implications of allocating the residual costs in different ways.

Reform

The Australian Energy Market Commission (AEMC) is in the midst of its pricing reforms. The NEM Review has struggled to extend its proposals through to incentives and behaviours for energy users. The Energy Charter’s co-design process has the networks attempting to design innovative tariffs for retailers.

The three hours of free solar announcement comes in the midst of all this reformist activity, and feels less timid. If the door is open for larger changes, (and you can comment on the Solar Sharer Offer until 21 November), now is the time to work out what YOU think tariffs should look like, and let the minister know.

Download the Tariff calculator here, and I invite you to share your proposals with me.

Heather Smith is an energy and climate change specialist, Churchill fellow and chair of Coalition for Community Energy.