1/10/2020 A comment on the nodal market design debate in Australia

A LinkedIn user posed some questions asking why the rule-maker for the Australian National Electricity Market is looking to implement a nodal electricity market design, asking, "Am I missing something?" Here is a brief note that attempts to raise a few points that might have been missed by folks uncertain of the rationale for change. Apologies for the avocado themed analogy, I wrote this up before breakfast!

https://www.linkedin.com/feed/update/urn:li:activity:6716586339128754176/

Reply: Perhaps to understand some problems that can arise with zonal market designs, it might help to think of whether an analogous uniform-price regulation would make sense for a different product.

Suppose a regulatory body imposes a rule that the price for avocados is fixed across Victoria at the price set in Melbourne. Although crude, to have it align a bit better with electricity, we’ll assume that avocados spoil in a day (can’t be stored), and there is a limited, centralised capacity to freely transport avocados throughout the state. Finally, we’ll assume that Mildura has better avocado growing conditions than Melbourne. What are the problems here?

- Mildura residents are forced to buy avocados at the Melbourne price. If there is a bumper avocado crop, but some avos can’t make it to Melbourne, rather than seeing the price drop in Mildura, they must still pay the Melbourne price and therefore the market won’t clear in Mildura and excess avocados must be dumped.

- Suppose that there is the occasional surplus of avocados in Mildura, which are currently being dumped and going to waste. An entrepreneurial guacamole maker sees this and says, “I’m going to set up shop in Mildura, use the avocados that are going to waste, and sell guac to the world.” After reviewing the market rules, they realise they can’t do this, because despite the avocados going to waste, they must pay Melbourne prices if they want to use them. Despite the aggregate economic benefits from having a guac plant set up in Mildura (converting waste into something productive), the private economic benefits to the guac maker don’t stack up because they must pay Melbourne prices, so they don’t invest.

- Instead, an avocado farm developer sees Mildura and thinks, “Wow, if I can start an avo farm here, I’ll get paid Melbourne prices for my output.” All the existing avo owners are saying, “Don’t do that, already we have to dump 10% of our yield, and if you set up here, were all going to have to dump 20% of our yield.” (remembering there is a limited capacity to freely transport avos to Melbourne). But the investor says, “Even if my new farm has to dump avocados, it is only 20% of my yield because all farmers up here proportionally share the dumping. I can still make money selling 80% of my yield because I get paid the high Melbourne price.” Despite the aggregate economic costs from having another avocado farm in Mildura, the private economic benefits to developer are positive because they receive Melbourne prices for their output, so they invest.

- In summary, a zonal market design incentivises the development of products with correlated output patterns in the same region, and disincentivises complements to those producers to be developed in that region.

- If instead there was a nodal market design, the price for avocados will rightly fall in Mildura when there is a bumper yield. Therefore, there is an efficient price signal for the guac producer and the avo developer to consider. Of course, if technology develops so that guac production gets cheaper, or demand for guac increases, we will see more guac producers which will in turn increase the ability for Mildura to have more avo farms developed.

I think this illustrates that the avocado zonal price regulation seems problematic. Obviously there are more intricacies with electricity, but this is the crux of the investment incentives argument – we need more development of storage or flexible demand such as hydrogen producers (the guac guys in the story) to mop up the excess production when variable renewable energy sources are generating in high volumes (the avo guys in the story) -- storage + flex demand would provide great economic benefit, but they privately won’t get to capture the economic benefit they create. Therefore, we won’t see the development of these technologies at efficient levels.

Another point that might be raised is why extra transportation capacity (transmission) between Melbourne and Mildura is not developed by the central authority. That is a fair point, and is a potential solution to bring the yields from Mildura to the demand centre in Melbourne. But to get a clearer picture of how much to invest in this new transportation capacity, it would be helpful if the waste that could be mopped up efficiently by guac plants was first incentivised to do so.

To address a couple of other points raised in the LinkedIn post:

- Regarding evidence: I am yet to see a study that provides evidence that zonal market designs provide better outcomes than nodal market designs, and theoretically the strongest argument seems centred on transaction costs – but I'd guess there would need to be large benefits from having everyone in a region facing a common spot price to overcome the economic costs arising from the inefficient price signals inherent to zonal markets.

- Regarding the comment that MLF price signals provide locational siting incentives: This point conflates losses with congestion. In the avo example it is like saying that it is ok to fix the price in Mildura at the Melbourne reference price, because in transportation a few avos bounce off the truck and the farmers aren’t paid for those avos. It still doesn’t price the impacts of curtailment, or set prices that reflect local demand and supply. By the way, dynamic loss factors would provide even better price incentives for the guac factory to produce more when conditions are such that when it does so it prevents more avos are falling off the truck, so to speak, further incentivising an efficient use of resources.

- I won't speak to why stakeholders might be against nodal market designs but there are two things worth considering:

o You would expect merchant storage and flexible demand investors to have their business cases vastly improved under a nodal market design. I’m not sure whether they have a voice right now, but it seems reasonable for policy makers to also consider the interests of future entrants when designing policy. Also, while I am not across stakeholder feedback, I would assume that consumer groups would be supportive of nodal market designs, and of course are less in number than other stakeholders, but perhaps are not outnumbered if you were to weight their views heavily given they represent the majority of stakeholders via proxy.

o Some markets around the world that have adopted zonal market designs have transitioned to nodal market designs. No nodal market has gone the other way. I am not aware of any nodal market considering a switch to a zonal market design.

Finally, although this not a definitive piece on the issue, this working paper of mine may be of interest. It contains a literature review that outlines some evidence surrounding the economic impacts from transitioning to a nodal market, provides an outline of the extent of the mispricing in the NEM, and ends with a discussion that expands on some of the points of contention in the nodal/zonal debate. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3501336. Although the piece has been presented at conferences and incorporated some feedback, it is not a peer-reviewed publication at this time and we are open to receiving more comments.