The electricity supply sector, for most, has to be one of the most boring sectors around. Electricity production is a simple process but it is operated on a huge scale. The sector has high capital costs but the product never changes so the risks are low. For all the differences in the geography, political systems and wealth of countries, the basic physical structure of their electricity supply systems look the same.

You have large generators, preferably at a distance from large population centres, fed by fossil fuels (mostly coal) that feed electrical power at high voltage into a transmission grid which connects to the first substation and onto the next level of customer, heavy industry and the distribution grids of cities and towns which then step down the voltage progressively which, in turn, supplies retail customers. Along the way there are losses: in transmission and in the step-down of voltage.

There are obvious differences between countries and regions. Some countries don’t have enough fossil fuels and have to import them. Others use less fossil fuels and instead use other sources like nuclear (France) or, if naturally endowed with water and aspect, hydroelectricity (Norway, Brazil). But the basic schema described is the same.

Customers expect and largely get sufficient electricity in a usable format (220 volts/30 amperes) whenever they flick a switch. All customers pay for their electricity consumption largely based on the power (measured in kilowatt-hours) – that they consume. The difficult part of all this is to balance supply and demand. For the most part, electricity generated has to go somewhere. It has to be consumed at the same rate that it is produced.

As electricity demand is uneven through a day, month or year, utilities have to tweak the system in real time. One way to address this is to have baseload generators that produce below peak demand on a 24/7 basis and have additional generators, known as ‘peakers’ that can be switched on and off quickly to fill in the peaks. The other method is to have some storage, mostly by pumping water up a hill/mountain during off-peak times and then releasing the water through turbines that generate additional electricity during peak times. Most electricity utilities use a combination of both. Once again, for the utility set-up as described, providing baseload power is inexpensive but meeting the peak is expensive. For customers paying a flat-rate tariff, the charge is a blend of the baseload rate and the peak rate. Over time, the peak, standard and off-peak demand times are very predictable. In general terms, provided that the plant is properly maintained and replaced when required, supply is also very predicable.

The basic electricity sector model has been with us for so long that most of us can’t imagine anything different. But the future will be different. Germany with its Energiewende policy has already made significant strides and is already able to provide the rest of us with some lessons. Bloomberg New Energy Finance founder Michael Liebreich argues that we are moving “from a centralised, fossil-based, analogue, geopolitically risky system to one which will be cleaner, more decentralised, local, smart and less exposed”.

Far too many of our discussions are framed by the current ‘energyscape’. Just a decade ago, renewable energy sources seemed impossibly expensive but now they are, in some respects, cheaper than conventional sources and they are getting cheaper all the time. Wildly wishful projections by environmental activists such as Greenpeace on installed capacity of renewables from just a few years ago are now the present day reality.

As renewables become mainstream, perhaps policy needs to adjust to how they are incentivised. If intermittency of supply is a problem, what does that mean for the grid that has to maintain a reliable supply? An excellent explanation by Jean Carsten shows the complexities of integrating renewables on a large scale into the grid and the interaction between generation and storage (even into the future).

The key take-away are that not all units of energy (measured in kilowatt-hours) are the same – they never were. Megawatt-hours (MWh) produced closer to where they are consumed are better – less energy is wasted, especially at peak times, being able to instantly ramp up and curtail a MWh instantly is more valuable than a constant flow; and adding an MWh that also helps the synchronisation of voltage and current (system power factor) is more valuable than an MWh that doesn’t do the same thing. You should therefore pay differently for the different types of MWh supplied. What the analysis also exposes is just how inefficient and wasteful our legacy electricity systems are.

Conventional energy sources such as coal or nuclear do provide baseload power but can only do so inefficiently. They are located far from the load so suffer from transmission losses and also from power losses and they can’t ramp up or down quickly. Procuring more nuclear energy is front of mind right now. While nuclear power is hugely expensive upfront and requires a huge amount to be procured to make economic sense (leaving aside being over-budget and over time, over and over), advocates say that if you take a long view, say 30 years, nuclear investments pay off. One need not only dwell on the immediate opportunity costs of that type of argument (ie are there other more pressing needs needing a similar investment), the real cost in a transitioning energy system is losing optionality. Exploiting optionality is exploiting large upside opportunity while taking on a limited downside risk. In circumstances of uncertainty, allocating scarce capital resources that commits us to a large nuclear programme for a 60-year or more period (the lifespan of a nuclear reactor) does not make sense. We lose optionality.

What might this upside optionality be? Outlines of what a different energy future may look like are emerging. If the growth of renewables have exceeded the wildest expectations over the previous 10 years what of 10 years hence? Twenty years? More?

What else might change? Energy storage. Most are familiar with Elon Musk’s gigafactory which will bring the price of lithium-ion batteries right down. Lithium-ion is just one option, and there are other competing technologies. Aluminum batteries may become viable and last only a month, a battery using salt water won a major prize. Author and futurist Ramez Naam sets out, in fascinating detail, the technology and economic transformations in storage that are already under way, explaining why energy storage is about to get very big and cheap. He describes an inflection point that will boost renewables and disrupt electricity business models. He describes a virtuous cycle of new markets, declining costs leading to more new markets. Grid-based storage costs are also falling for utilities and given that they know that not all kilowatt-hours have the same value, they would also know that storage does not need to compete with baseload power. Instead, storage offers flexibility and competes instead with the “price of peak demand power, the price of outages, and the price of building new distribution and transmission lines”.

The changes could happen quite quickly. Last year Citibank released a report entitled Energy Darwinism II. Its forecasts are based on the assumption that battery storage costs will halve over the next five to seven years, becoming cheap enough for early adopters which then push prices down further, making home solar generation plus storage more viable and, increasingly, also attractive at a local grid level. Solar power and storage hit grid parity in large parts of the world (the sunny parts) by 2030.

Surely anyone would want to have the option to exploit the changes ahead? If you have already committed to 20^th century technology, you lose the upside option.

In a widely publicised article, Liebreich, says energy systems becoming more local and cleaner is not even controversial anymore. While he juxtaposes the enormous co-benefits of clean energy, such as improved air quality, local jobs, technological breakthroughs and economic resilience, with the colossal externality costs of fossil fuels, in terms of effects on public health, defence costs, exposure to commodity prices and so on. He makes an additional point that the type of energy system we want must also be understood, and that is in terms of the social structures in which we want to live. Fossil fuel-based energy lends itself to scale and centralisation, both the physical centralisation that defines our present as much as the “inevitable fellow-travellers of political and economic centralisation”. Clean energy, he explains, is inherently more local, more distributed and more accountable. As the energy system goes clean, so it will become more local, more human in scale and so more open, more accountable, more answerable to the community, and more socially inclusive.

Inclusiveness means that we all become active participants in our energy system. But this inclusiveness extends to social inclusiveness. Liebreich notes that the global energy sector has an overwhelmingly white, male and macho orientation.

All of this has profound implications for existing utility companies. The Citibank paper says most utilities are staring at the future like rabbits in the headlights. A former US Energy secretary says his discussions with US-based utilities show a varied response from “tell us what to do” to “deer in the headlights” and, of course, “we will fight this”.

Our own utility, Eskom, and our country are uniquely unprepared for the future. Our only energy export, coal, is going to become increasingly uneconomic as demand for it falls. Eskom is one of the most centralised large electricity utilities anywhere in the world and it relies on coal to generate over 90% of the power that it produces. Right now it is simply malpractice to close down upside optionality through ill-considered investment that assumes the continued viability of a centralised electricity utility model.

If Eskom is poorly placed then our municipalities are arguably in a worse position. They distribute over half of all electricity consumed in the country. A recent Statistics South Africa publication shows how dependent municipal finances are on the current system. Large metros derive about 30% of their total revenues from selling electricity. Their net margins (profits) are anywhere between 15%-20% even after all cross-subsidies for low-income users are taken into account. This can’t continue as a recent study published by the Institute of Security Studies into the long-term future of the South African electricity sector shows. It will increasingly cause grid defection by higher income customers, who fund the cross-subsidies, to happen more quickly.

If one wants to improve energy efficiency, it is not good policy to subsidise electricity consumption. Instead, we should figure out the direct costs of cross-subsidisation and pay the cash equivalent to beneficiaries directly. There is now significant evidence that cash transfers are more effective for the poor than buying stuff for them.

The municipal revenue model will have to change. Perhaps this means charging an access fee for the local grid. Some utilities elsewhere are experimenting with this. Certainly, time-of-use charging is an essential first step. Other changes will be necessary. Procuring electricity generated locally makes sense (there has to be a spinning turbine in the mix) but the Municipal Management Finance Act makes it very difficult for a local municipality to enter into any long-term agreements. Power purchase agreements for significant generation capacity typically have a 20-year duration.

Embracing the inevitable, if yet uncertain, changes is hard to do. As strategy advisor Shawn Hagedorn points out, South Africa, based as it is on an extractive economy, has been slow to adapt to a globalising world in which we are but bit players. Perhaps the challenge should be put this way: it is far more risky to hold onto the legacy energy system than to adapt and change the electricity system.

Next year, the Department of Energy is due to publish its next installment of the Integrated Development Plan, mapping out our electricity system for the next 30 years. Let’s see what it says. DM

Photo: A handout image provided by NASA on 23 December 2014 shows observations from by NASA’s Nuclear Spectroscopic Telescope Array (NuSTAR) overlaid on a picture taken by NASA’s Solar Dynamics Observatory (SDO) of X-rays streaming off the sun. This is the first picture of the sun taken by NuSTAR. The field of view covers the west limb of the sun. The NuSTAR data, seen in green and blue, reveal solar high-energy emission (green shows energies between 2 and 3 kiloelectron volts, and blue shows energies between 3 and 5 kiloelectron volts). The high-energy X-rays come from gas heated to above 3 million degrees. This image shows that some of the hotter emission tracked by NuSTAR is coming from different locations in the active regions and the coronal loops than the cooler emission shown in the SDO image. EPA/NASA/JPL-Caltech/GSFC