What does the future look like for energy?

After you read the next sentence, you might call me a lunatic or think that I'm smoking some illicit substance. In this article, I will assess how cost declines in solar generation and batteries will cause rooftop solar generation and batteries to disrupt utility-scale generation, creating a new energy architecture where every rooftop and parking lot will have solar and batteries for purely economic reasons.

Watch this presentation by Tony Seba of RethinkX on "Clean energy disruption and the collapse of coal, oil, and gas". This link will jump straight to where Mr Seba discusses GOD parity, but I encourage you to watch the whole video.

Here are screenshots of three slides from the presentation that discuss GOD parity:

Tony Seba defines GOD Parity as when the cost of rooftop solar, or solar and 4 hours of energy storage in batteries (or any other amount of energy storage), falls below the cost of transmission (plus the cost of distribution). It stands for Grid On Demand parity. He adds that in some places, the cost of transmission is really high, like 12 c/kWh in Australia. He goes on: "So when you can generate for less than the cost of transmission, that means that even if the utility can generate at less than $0/kWh… it's never going to be able to compete with your own rooftop generation… So essentially, utility-scale electricity will never be able to compete with your own [rooftop solar] generation. That is GOD parity."

To repeat, Tony Seba is stating that any and all utility-scale will never be able to compete with your own generation. Now, you could argue that there is a future for growth with utility-scale clean energy generation because installing solar on high-rise buildings won't supply enough generation to meet on-site consumption, thus requiring off-site generation. However, it is very common for buildings with solar to generate more than their energy usage, so energy demand that can't be met on the same building's roof can be met by local buildings.

It seems like utility-scale solar could be supported by and support the growth of other technologies, such as—to quote just a few:

• water desalination and treatment
• waste processing and recycling
• metal smelting and refining
• chemical processing and manufacturing
• cryptocurrency mining
• distributed computing and communications,
• fuel production
• carbon removal
• manufacturing of solar panels, wind turbines and batteries

On the other hand, utility-scale solar saw less growth in 2023 than in 2022, and the number of new projects is declining in 2024 (Sunwiz).

You could also argue that utility-scale batteries are very profitable, even lucrative at the moment, and have a lower levelised cost of electricity than residential batteries. Secondly, the continued growth of utility-scale batteries will, according to Wright's Law, reduce the cost of all battery applications and vice versa. Thirdly, any energy storage, utility-scale or otherwise, is helping to firm the capacity of intermittent renewable generation and disrupt gas-peaking plants and inflexible 'baseload' fossil-fuel generation.

Additionally, currently consumers cannot avoid paying for the transmission and distribution costs of electricity unless they go off-grid entirely, otherwise there would have to be changes at a regulatory level, e.g. reduction in costs for the transmission cost if in a microgrid, embedded network, or with local peer-to-peer energy trading. At the moment, going off-grid is not an attractive option unless you're building a new property that is not near a local grid supply. On average, the cost to connect to the grid for a new-build property can range from $10,000 to $30,000, but in more remote areas, this cost can escalate to over $50,000 or more.

If the cost of an off-grid system were able to decline by 70% to $9000, then that would be the same order of magnitude level as an on-grid solar-only system today, and so at such cost declines, it does not seem unrealistic to expect off-grid solar and battery uptake to see its own exponential growth. In Australia, over a third of homes have solar, meaning that we are well into the early majority stage of solar adoption. Off-grid installations are growing year on year. (Off Grid World). This growth, if accelerated, would first start to reduce the growth of utlity-

Mr Seba goes on to say that many places around the world (such as Australia and California) are already below GOD parity. Compare also with the following studies:

• this study indicates that the LCOE of solar in Australia in 2021 was 3–9.9c/kWh: https://lnkd.in/dbBXc56H
• see figure 2.1 (see comment below) of the AEMO 2021 Residential Electricity Price Trends report, which shows that the 2021/2022 nationally averaged transmission cost component of a bill is 2.28 c/kWh for transmission and 9.70 c/kWh for distribution. Here's a link to the report: https://lnkd.in/dyrbEqAp)

The conclusion from comparing these figures (9.9 at the top-end is less than 9.7 + 2.28 = 11.98 c/kWh, approx. 12 c/kWh) is that the levelized cost of rooftop solar only (without batteries) is already cheaper than the combined cost of distribution and transmission.

What about the levelized cost of solar and 4 hours of storage, unsubsidized?

By comparison, a recent study in Kenya found that: "The systems with the smallest levelized cost of electricity (i.e., 0.11?USD/kWh = 0.16 AUD/kWh as of 13/7/2023) use either 7.5?kW or 10?kW of solar with 20?kWh of storage" using second-life batteries (SLBs) e.g. from electric vehicles. Note that this is the total LCOE of the solar and battery system. While this value is in a developing country, this is above the GOD-parity cost in Australia of 12 c/kWh. I also did an analysis here of "first-life" stationary energy batteries, but the levelised cost was 33 c/kWh, which is twice as much as the above figure. This shows the value of second-life batteries, suggesting that startups in this sector, such as Relectrify, are poised for strong growth.

4 c/kWh (16-12) is a small difference between a solar and second-life battery system and transmission + distribution costs, and the gap is closing quickly. Costs are projected to continue to decline in line with Wright's Law, also known as Swanson's Law, when applied specifically to solar photovoltaics (PV) modules, as elaborated e.g. here, here, here, and shown below.

Interestingly, in a footnote here states: "Plausibly, it isn’t just the passing of time that drives the progress in computer chips, but there too it is the learning that comes with continuously expanding the production of these chips. Lafond et al (2018) explain that the two laws produce the same forecasts when cumulative production grows exponentially, which is the case when production grows exponentially. More precisely, if production grows exponentially with some noise/fluctuations, then cumulative production grows exponentially with very little noise/fluctuations. As a result, the log of cumulative production is a linear trend and therefore predicting cost by the linear trend of time or the linear trend of log cumulative production give the same results."

From these remarks, one may also posit that the same logic can apply to solar, wind and batteries, i.e. that predicting cost by the linear trend of time or the linear trend of log cumulative production give the same results. Which leads us to the next chart:

The price of solar PV electricity declined by 90% between 2009 and 2019. Therefore, it seems that it could well decline by another 90% between 2019 and 2029, and further extended from there. During this time, log cumulative capacity may be expected to increase from 1 TW to 10 TW. As of the time of editing in May 2024, the total cumulative capacity of solar at the end of 2022 is reported to be 1.053 GW, up from 861 GW at the end of 2021.

Switching to the logarithmic view of the same chart, we can see that cumulative capacity went from 100 GW in 2012 to 1 TW by 2022:

Of course, if we are more conservative due to supply chain shocks from COVID, wars in Ukraine and Gaza, etc., it may only decline by 80% or even 70%. Regardless, a cost decline of 70%, 80%, or 90% will drive more demand and new applications, further driving down the cost "like gravity", as Tony Seba puts it. In other words, a cost decline of another 90% is not unrealistic.

To digress with envisaging an accelerated solar application, one may envisage a tenant paying to get solar and get a return on investment in a matter of months rather than years. Or a landlord just paying for it because it would increase the property value. Or both contributing to the cost since both would benefit. This would accelerate another market for solar, driving faster cumulative growth and a decline in cost as S-curves from multiple growing markets compound on each other. Ark Invest has similarly described the same phenomenon with Tesla launching new vehicles in different car types: luxury sedans, luxury SUVs, a mass-market sedan and crossover, etc., with a new S-curve of growth for EVs in that vehicle body type driving faster S-curve growth for EVs as a whole.

The cost of solar and batteries is falling (conservatively assumed by RethinkX to be 12% p.a. annually on average for solar, 15% p.a. for batteries, and 5.5% p.a. for wind—see here). Additionally, repurposing EV batteries as second-life stationary energy batteries (a la Relectrify) could play a big role in future as (autonomous) EV uptake disrupts human-driven ICE vehicles (a double-disruption of labour—paid or unpaid—and fossil fuels), and this uptake itself is a major driving force (no pun intended) in the declining cost of batteries with the cumulative growth in battery production, used for both transportation and stationary energy, causing positive feedback loops that drive further disruption.

So it isn't difficult to see that at these rates of cost decline, it won't take much longer for GOD parity to be reached for solar and battery systems, while solar-only systems are far below GOD Parity. I crunched the above numbers further here to work out that it may take about 3 years for a solar and 2nd-life EV system to reach GOD Parity and about 7–8 years for a solar and 1st-life battery system to reach GOD Parity. Note that my analysis only accounts for the capital cost and the energy generated/stored in calculating LCOE; the drawback of LCOE calculations is that they don't account for the value of the energy saved by not drawing from the grid or credited from exporting electricity or providing grid services (like demand response and FCAS, e.g. through a VPP). I may calculate that in the future.

Also, note that electric vehicles are quickly disrupting ICE vehicles, and with vehicle-to-grid charging and second-life EV batteries repurposed as stationary batteries, they will help to drive energy storage adoption at the demand side, lowering the need for big batteries. Considering that solar rooftop systems without batteries have already reached GOD parity, investing in new utility-scale generation doesn't seem to make much sense, although it can be used to meet the energy needs of consumers who can't generate enough energy for themselves. Nevertheless, these energy needs can also be met through peer-to-peer trading with more local generation (a la Localvolts). On the other hand, RethinkX's Energy Report posits overbuilding generation to get Super Power, which may be harder to achieve with rooftop solar alone.

So should we just about end all investment in utility-scale generation (except for wind farms, big batteries, and solar farms, which I'll elaborate on) and invest it into demand-side clean energy tech instead? Well, I have already addressed this above. The answer is that utility-scale solar, wind and batteries can still play a role in quickly ramping up energy generation and overbuilding it to get Super Power. However, it seems that there will be disruptive pressure on all utility-scale generation (especially fossil fuels) as GOD Parity is achieved for solar and battery systems. Nevertheless, overbuilding generation, including utility-scale, can more quickly get to a more economical 100% renewable energy electricity mix without the need for any fossil fuel generation, as well as having the additional benefit of Super Power, which can create a larger market and new applications for abundant, low-cost, clean energy.

Examples of demand-side clean energy tech include rooftop solar, batteries, virtual power plants (e.g. with Tesla and several other VPP providers), peer-to-peer energy trading (a la IOEN, Localvolts Pty Ltd), microgrids, embedded networks, demand response management tied with access to spot prices (a la Amber Electric, Flow Power—which has particularly smart contract plans to let customers find a balance between minimising exposure to volatile spot prices and incentivising demand response), Local Volts), use of distributed solar and batteries for participation in FCAS markets and the spot market (a la Tesla Powerwall and Reposit Power), bidirectional electric vehicle charging, electric vehicles, robotaxis, bitcoin mining, etc. Maybe there will still be some demand for utility-scale generation for high-density buildings, but this could also be met by overgeneration locally—some people/businesses generating much more than they need and selling it to others with P2P energy trading or on the spot market.

Utility-scale wind generation should be necessary for sufficient generation at night and when the sun isn't shining. Otherwise, the less wind generation, the more batteries need to be installed. Small-scale wind energy is yet to be able to compete, although there are some technologies that may have potential, e.g. Vortex Bladeless. Big batteries are growing quickly, and this is because they can offer additional benefits, such as a lower LCOE due to economies of scale, generating revenue from providing stability to the grid (grid services such as FCAS) to keep lights on and prevent load-shedding blackouts, and load management / demand-side management to store energy when prices are low and sell it when prices are high, flexible ramping (which helps to flatten the duck-curve), reducing utility-scale renewable energy curtailment, and renewable capacity firming.

Many companies globally are investing money, time and resources into renewable utility-scale generation, and there is a push for major transmission and interconnector projects. Some reasons for continuing to invest in big batteries, and big solar and wind, are listed above. Distributed generation appears to be growing faster than utility-scale. Additionally, manufacturers can produce and sell batteries without having to own the system for their lifetime, and developers can also develop and sell rather than own a system. System owners may not see a risk of disruption, for reasons outlined above. Institutional investors may be drawn to big solar and batteries because they don't see a near-term risk of disruption by distributed generation, and because they see big solar and batteries as a way of meeting sustainable development goals and net-zero targets.

There are many players in the clean energy utility-scale sector, and it's hard to get a big ship (or an armada) to change course quickly, as history has shown with disruption. And yet, rooftop solar and batteries cannot but disrupt all utility-scale generation: this will be driven purely by economics, where it will make more financial sense to buy your own rooftop solar and battery generation, rather than buy it from utilities or a retailer. Yet again, one may argue that utility-scale solar, wind and batteries will still have a role to play with demand from apartments, rental properties, high energy users that can't generate and store enough on-site to meet all of their energy needs, all of the time, and for Super Power applications: creating new or expanded markets due to low-cost electricity.

Next question: can utility-scale generation still make a net profit if it sells energy at a loss (to be able to compete with demand-side energy generation—rooftop solar and batteries) but makes money by providing grid services or through more innovative forms of revenue, such as co-located bitcoin mining? It seems possible. On the other hand, utility-scale generators shouldn't have to sell energy at a loss, because:

• there should be demand from apartment buildings, rental properties, etc. Even if rooftop solar generation penetration greatly increases, there also needs to be energy storage to avoid curtailed generation. On the other hand, that demand may be met more competivitively by local generation, provided that regulation enables local peer to peer energy trading while paying less for network component costs. Conversely, a rule change to the Australian Electricity Rules to enable such local peer-to-peer trading was blocked. That doesn't mean that it will be permanently blocked, however.
• demand can increase and again be co-located with big renewable energy e.g. bitcoin mines, smelting, factories, etc., as Tony Seba describes as Super Power in the RethinkX Energy report)

There are concrete examples of co-located bitcoin mining, e.g. Australia's first solar-powered crypto-mining centre in Whyalla, which "provides the developers of the Tregalana Solar Project with a guaranteed baseload, ensuring that the project is financially viable." Research has also been done in this area, e.g.:

• by ARK Invest, as detailed here. They show that bitcoin mining can help to increase the share of renewable energy in the electricity mix to 100%, and this may be more economical than overbuilding solar, wind and batteries, a la RethinkX's Energy paper.
• Mining Bitcoin With Tesla Solar & Powerwall: this article demonstrates that Bitcoin mining can be done on a small-scale, as can grid services.
• Batteries or Bitcoin: Is Mining the Best Use for ‘Excess’ Green Energy?

However, note that there is also a possibility that Bitcoin could be disrupted, e.g. by Ethereum, or the Holochain ecosytem (including Holo and IOEN).

Going off-grid and maxing out on solar and batteries is not economical, unless building a remote new-build property that would be very expensive to connect to the grid. Even with cost declines, in most it will make more sense economically to continue using the grid and growing utility-scale generation of solar, wind and batteries, and disrupting coal, oil and gas. If the cost of solar and energy storage falls below the cost of distribution, it may then make sense for many properties to disconnect from the grid, although the incentives to stay connected to the grid may still outweigh the drawbacks of disconnecting. For example, not being able to sell excess energy back to the spot market, not having a backup source of power during days of cloudy weather, etc.

The outlook for utility-scale renewable energy is not clear—predicting the future is not an easy task! However, it seems that utility-scale renewable energy will continue to grow, since properties may not get solar and batteries or not get enough to supply all their energy needs—for instance, rental properties, apartments, high-rise buildings, and high-energy consumers—unless disruptive innovation helps here. Home and business owners will be strongly incentivised to max out on solar, get at least 4 h of battery storage, bitcoin mining, and be net exporters. Smart tech is helping to accelerate the clean energy disruption, like peer-to-peer energy trading, virtual power plants, automation, demand response, and access to wholesale electricity pricing with smart contracts.

You can help to accelerate the technology disruption to a new energy architecture by sharing with your networks, and investing in disruptively innovative clean energy tech, including on your own property. Leave your thoughts, comments, suggestions, critiques in the comments! Feel free to challenge my assessment or build on it!

Further reading:

• Browse the RethinkX website
• Read this thread here for insights from chatGPT on the predicted cost reductions in solar, wind, and batteries by 2030. The insights include the cumulative capacity of these technologies in 2030, based on Wright's Law, and how robotaxis and bidirectional charging would also play a dominant role in providing energy storage.