Building transmission in a market

The release of the Integrated System Plan this week, and last week the draft report of the Riverlink Regulatory Test, has put transmission planning back in the spotlight. Energy Insider will provide some insights on these key documents in the coming weeks. In the meantime, we take this opportunity to reflect at a high level on the key issue of the interface between transmission planning and the electricity market.

Networks and the Market

In electricity, the “market” only applies to half the industry: the business of generating and storing electricity, and engaging with customers. Meanwhile the grid is a natural monopoly and economically regulated. Once built, returns on network assets are effectively guaranteed, and recovered from ongoing levies on customers. Once connected, generators don’t pay to use the grid, no matter how close or far they are from their customers. It’s called “open access” and is analogous to the way (non-toll) roads are built, used and paid for. It contrasts with “contract-carriage” infrastructure, such as gas pipelines, which are built by entrepreneurs who charge haulage transportation fees on those using their pipes.

As a monopoly, the electricity grid must be planned by a non-market actor. This is a very challenging role, as they have to predict the market well in advance – they don’t themselves determine how many customers will connect, or what generators get built and where. On the other hand, it is also an easy role, as the planner is not financially exposed to the impact of their decisions.

One can immediately see the ups and downs of this approach:

• Pros:
  • Having built the grid, using it becomes as simple as driving down the road – thankfully we don’t have to sign a haulage contract to drive our local streets!
  • Along with this simplicity comes equal access, which assists competition between the market-facing actors.
• Cons:
  • As costs are socialised, incentives for market-facing actors to consider their own impact on the grid are lost. For example, a generator who locates far from load does not face additional costs from using more infrastructure than one that is closer – so we lose a valuable incentive to construct in a more efficient location.
  • Because there are no access rights, users face an uncertain risk of losing output to congestion if there is less network than suppliers want to use – again, analogous to a traffic jam. Contrast this to a gas pipeline, where a firm haulage contract will buy you access to the express lane.
  • Planning has to be performed outside the market, with the only motivation our social good. Again, this is just like roads. From an ivory tower, a road-planner must decide where to spend scarce public funds to try to manage, but not eliminate, congestion. Some more socially minded readers might even see this as a pro, but most economists prefer planners who are directly exposed to their own decisions.

Since the NEM’s inception there have been several proposals attempting to move from the current fully open-access regime to one with at least a few contract-carriage elements. The most recent of these was the AEMC’s Optional Firm Access proposal. All have failed to move beyond the drawing board.

The Regulatory Investment Test

Given we have accepted the open-access regime, we need to set a framework such that the planners develop an efficient grid. Thus, every large network development is subject to a detailed cost-benefit test overseen by the Australian Energy Regulator (AER) known as the Regulatory Investment Test for Transmission/Distribution (RIT-T/RIT-D). This is a crucial assessment, and fulfils various functions:

• Firstly, it seeks cost effectiveness for the customer. It strives to eliminate the planning of unnecessary or inefficient assets by getting the planner to transparently justify their decisions and defend them against appeal.
• Secondly, it seeks to encourage the planner to consider as wide as possible a range of solutions. For example, if some demand-side management can cheaply lop off a peak, this needs to be pursued ahead of building traditional assets.
• Thirdly, it creates a predictable framework around which the market-facing parts of the industry can invest. Every network investment will inevitably “move the deckchairs” in the market – creating winners and losers. The RITs give market investors some confidence about which deckchairs are likely to move when.

Like everything in electricity, RITs are hideously complex, but at their core they try to achieve the key goal of all classical economics – maximising total social welfare.

Let’s simplify to the extreme by imagining two unconnected market regions, A and B. A has a surplus of $100 generators, whilst B has only $150 generators. Every MW of interconnection between them can therefore save $50 of generation cost. If each MW of interconnection can be built for less than $50, the line passes the test.

The line will also affect prices in both regions, raising the price in A but lowering it in B. Price equalisation is not the objective of the test nor can it directly be included in the assessment. In the very first years of the NEM the test was actually based on total customer costs, but the flaws in that approach became immediately obvious. In one famous case, an economically efficient line was knocked back[i]. As expected, the assessment forecast a price drop in the importing region B and a smaller increase in the exporting region A. However because A had more customers than B, as a total, NEM-wide customers actually paid more, so it failed. And to follow this logic further, it implied that in some cases existing interconnectors should actually be cut!

The AER promulgates the test with guidelines[ii] that help the planner know what assumptions they must make. As it’s all about predicting the future, of course complicated models are involved, along with the industry’s usual band of consultant soothsayers.

The treatment of future generation is a key issue. If a generator already exists but is losing output due to congestion, then the value from removing this congestion is a benefit that can be included in the test. If however the generator is not yet built, a planner can incorporate a build cost relativity: if the line allows future generators to be built in region A that are cheaper to build than future generators in region B, then the difference in this build cost can be included.

This is key to the concept of Renewable Energy Zones as discussed in the Finkel Report. Engineers have always fantasised about exploiting the resources in Australia’s heartland. However wind and sun can be found everywhere, albeit at varying intensity. The key question is whether the savings in harvesting it at the point of greatest intensity exceed the cost of transporting it over great distances to the consumers. The RIT has always been able to assess this equation and usually concludes the obvious – in the negative. As a result, the RIT itself is frequently criticised. But the fantasy’s problem lies in its own economics, not the test.

Another key feature of the RIT guidelines relates to the project scope. There is a clear incentive for network companies to incorporate additional components into a favourable project in order to get as many assets through the one test as possible. With transmission – where a project’s boundary is always vague, it is a great challenge for the AER to stop this practice so that every separable component can be judged on its individual merits. AEMO’s Integrated System Plan[iii] has lumped together the benefits of all of its many recommended augmentations over 20 years. It will be very important for a much more granular approach to be adopted before customer funds are committed.

Conclusion

Whether or not you think the NEM’s open access regime is the right approach, it’s the approach used in the NEM and likely to stay that way, so we need a good economic framework for non-market planners to use to develop the grid. The RITs are based on the sound economic concept of maximising net social welfare. Like everything in monopoly regulation, the task will be highly complex, suffer information asymmetries and be subject to ongoing debates that frequently revisit its original basis.

However the Australian Energy Council sees a good RIT as a key part of making the whole industry work, both by protecting customers through efficient use of their money, and by protecting the market through providing a predictable framework that market investors can understand.

We should be skeptical of the RIT’s critics, who may not hold these objectives as dearly.

[i] Academia Edu, S. Littlechild, Australia: SNI and Murraylink Revisited

[ii] AER, Regulatory investment test for distribution application guidelines, September 2017

[iii] AEMO, Integrated System Plan, July 2018