The energy giants are looking down a barrel as households build their own rooftop generation. Research shows their prospects hinge on where they set feed-in tariffs.

The clean energy crowd loves to look into the future. When they do, they see fossil fuels giving way to renewables as the grid is redesigned to support generation by a vast array of wind and solar assets, supported by all manner of storage.

What do the energy incumbents see when they look into the future?

For certain, they see turmoil. Yes, they are welcome to be involved in the clean energy transition and many of them are – but the slow switch over brings plenty of danger. What is the best way for them to deal with it?

If the great energy baton change from polluting to clean, consumer gain may translate as incumbent pain. To what extent can a balance be found, then, where the majors protect their businesses while not being seen to be standing in the way of change?

Kelvin Say, a PhD researcher at Curtin University in Perth, looked at this problem while studying the effect of customer-scale renewable energy technologies on the socio-economics of electricity markets. His angle is to consider the economics of renewable energy on households as they adopt solar with or without storage, and also on retailers.

“What I find really curious is [household solar owners’] ability to undermine the utility business model,” Say tells EcoGeneration. This tension between the demand side and supply side of the electricity market never existed before the advance of solar PV, he says, but now it has fired up an academically interesting power dynamic between customer and provider.

Kelvin Say has looked at the trade off energy companies will make when setting feed-in tariffs for solar exports.

Feed-in tariffs in focus

In his research Say has set himself the task of isolating the tipping point up to which providers are willing to compromise. If energy companies understand how an increasingly more dynamic market can expose weaknesses they never suspected they had, they will be willing to change their services to ward off a dire scenario – even if it means forfeiting some long-held advantages.

In a world where more and more households can be expected to invest in solar, to what extent is an energy retailer’s business reliant on where the feed-in tariff is set? Simply, that’s the question Say’s research aims to answer. “Would you raise it, would you lower it, and if you do, what are the repercussions on the households investing in their own systems into the future?”

If a retailer raises the feed-in tariff it pays customers for their surplus solar, it not only dampens revenue but also incentivises system owners to increase the amount of PV on their roofs. However, higher feed-in tariff levels also act as a disincentive for system owners to invest in storage. For them, the greater the differential between the price of electricity and the feed-in tariff, the greater the incentive to own a battery (see charts).

As the difference between what energy can be bought and sold for narrows, the longer households will delay the purchase of a battery. “[Batteries] will have to become more cheap in order to justify that return,” he says.

A simple tactic to use if retailers wanted to delay the take-up of batteries would be to raise the feed-in tariff. “But because they pay for the feed-in tariff the uptake of solar PV ends up costing them more,” he says. “They may delay batteries, but they also encourage more PV users, which then hits their bottom line further – so they are constrained in how far they can raise the feed-in tariff.”

Lowering feed-in tariffs would have the effect of incentivising customers to size systems to maximise self-consumption and minimise exports – or to trigger investment in storage (see chart at bottom). That’s worrying for retailers because owners of moderate-sized systems with batteries will be able to rely on their generation for about 75-80% of all the electricity they need across the year, he says.

Damned if you do…

Put simply, depending which way a retailer shifts its feed-in tariff dial, it might inspire earlier adoption of storage if tariffs are lowered or trigger a loss of revenue if tariffs are raised and customers super-size their systems.

“There’s this tension from the retailer’s perspective where if they go up they are damned and if they go down they are damned,” Say says. Retailers should be looking at this dilemma while residential storage is still a long way from mass adoption.

Say’s paper Power to the people: Evolutionary market pressures from residential PV battery investments in Australia is co-authored with Michele John of Curtin’s Sustainable Engineering Group and Roger Dargaville of Monash University. It relies on load data from 261 homes and predicts uptake of solar and/or storage over 20 years. Forecasting is based on a 5kW inverter limit on a feed-in tariff, but households are free to add extra PV generation.

Who owns storage?

Reading Say’s paper it dawns in EcoGeneration that an obvious solution for retailers would be to join in the fray and invest in storage themselves, with assets deployed at optimal spots around networks. Say concedes he likes the idea of “distribution-scale batteries” as a cost-effective solution for customers, network providers and generators. “It would make financial sense,” he says. “It gives the customer a faster way to adopt batteries and it gives the system operator and network provider control over how it would operate. That is one viable way of getting out of this mess.”

Another option is for networks or retailers to assume some control of residential batteries in return for a subsidy or financing, as happens in the various trials of virtual power plants around the country.

Because batteries are so expensive, it’s unlikely buyers would size them to soak up all surplus solar generated on a searing summer’s day. Exports during summer will still be high, he says, unless storage prices fall a long way. Until then, the duck’s stomach will be low to the ground and the ramp to the evening peak will be uncomfortably steep.

Export bonanza

Before batteries, households with low feed-in tariffs available to them would install small PV systems. When they do finally invest in storage, they would also need to upgrade to a larger PV system for the decision to be judged as rational by a kitchen table panel of judges. “The net effect is, during summer you’re just going to get bigger and bigger peaks … and more and more exports.”

As batteries are installed and additional PV capacity is added, Say’s paper estimates by the end of the 20-year period with feed-in tariffs between zero and 50% that residential customers have the potential to export greater than 7.7 times their annual imports.

“You can quite easily go down to 75-90% reduction in all your consumption and at the same time you’re still exporting,” he says. “Even if you don’t pay customers for that energy, they’re still happy because they’ve still got significant bill savings.”

The challenge in that alternative future will be to know what to do with all this enormous amount of surplus energy. An extreme option is curtailment, “to remove generation from the network when the network can’t support it”, which seems a shame. A better option would be some form of central authority that would control PV and inverter usage, he says. “It could be a virtual power plant, aggregator or regulated authority that sits on the network and sends a signal to the feeder that says, ‘Everyone back off a little bit.’”

Horizon Power’s already playing this role on some of its microgrids in Western Australia, he says. How this melee of communication among networks, inverters and household appliances will be coordinated in a world of five-minute settlements, and among millions of customers, is yet to be addressed.

“I can imagine there is going to be a lot of necessary investment to come to terms with that,” he says. “There is no off-the-shelf solution right now.”

A calm and calculated clean energy future is still some way off. Expect plenty more hard work from Australia’s energy brains in the meantime.