Uncertainty abounds in the NZ ETS
Exploring the Government’s consultation on auction volume and price control settings for 2025-2029
Summary
Many of my readers will have discovered my Substack through my articles earlier this year on the New Zealand Emissions Trading Scheme (NZ ETS). In those pieces (here and here), I explored how foresters' behaviour can affect surplus drawdown rates and resulting prices in the scheme.
Given that I've already covered key variables affecting price outcomes in depth, I'll keep this analysis more brief. This article focuses on key proposals from the Government's consultation on auction volumes and price control settings (MfE, 2024a) that closes Friday. Specifically, I examine:
Surplus reduction targets and consequent auction volumes: Using MfE’s updated NZ ETS Market model, I explore the implications for the market over the next 2-6 years.
Auction price floor: I analyse the proposals in the context of current market dynamics and discuss related public commentary.
Surplus reduction & auction volumes
The Government’s proposals
First, let's recap the Government’s proposals for surplus reduction and consequent auction volumes. Three options were put forward (MfE, 2024a):
Status quo (option 1): No change to surplus estimates and consequent auction volumes.1
Option 2: Update surplus reduction quantum and consequent auction volumes for 2027-2029.
Option 3: Commission’s advice with updated surplus reductions and lower auction volumes for 2025-2029.
The consultation document from the Ministry for the Environment (MfE) frames each option in terms of its methodology for NZ ETS unit settings, where options are viewed as differences in surplus reduction. For example, option 1 is described as “continuing with the surplus estimate of 49 million units, and not updating surplus reductions for 2025–28” (2024a, p. 21).
However, for the purposes of this article, we’ll avoid delving into the intricate logic of MfE and the Commission’s methodology, and discuss the options more directly as proposals for auction volumes.
Release of NZ ETS Market model
A few weeks after the consultation began, MfE released an updated version of their NZ ETS Market model, which was previously used as part of the evidence base for the 2023 ETS Review (MfE, 2023a). The release of the model offers an opportunity to examine the impacts on market conditions of:
Auction volume settings.
Uncertainties in the liquidity of post-1989 forest units held for harvest.
Afforestation rates for the remainder of the decade.
Results for each of these analyses are presented using two baseline scenario assumptions. The first baseline sees the model calibrated to the Commission’s ‘demonstration pathway’ used as part of their 2023 advice for the second emissions reduction plan (Commission, 2023a, 2023b), but with a $0 carbon price. The second baseline calibrates the model to the Commission’s (2023b) ‘current policy reference’ scenario, but again with a $0 carbon price starting point.
Two baselines are used in light of the Government’s recent cuts to several complementary measures outside the NZ ETS (e.g., the clean car discount, recycling of auction revenues to climate mitigation). This presents some ambiguity as to which trajectory emissions would take in the absence of the NZ ETS.
I suspect the ‘demonstration pathway’ baseline may be overly optimistic about reductions in gross emissions without the NZ ETS in light of the Government’s recent actions. However, this is based largely on intuition. Hence, the divergence in results between the ‘current policy reference’ and ‘demonstration pathway’ baselines should be viewed more as indicative of uncertainty in potential market outcomes.
Figure 1 below shows gross emissions covered by the NZ ETS for each baseline scenario (with a $0 carbon price applied), highlighting the differing effort required to balance supply and demand under each scenario.
As a final note before results are presented, I advise caution in interpreting specific NZU price values in analysis set out below. While MfE’s model can be useful for understanding the direction and, to some extent, the magnitude of price level changes, it should not be seen as a predictor of price.
Current auction volumes versus the Commission’s volumes
First, let's examine the difference in modelled NZU price and the use of stockpile volumes held in private accounts (including both surplus and non-surplus units) for the current legislated auction volumes (option 1) and the Commission’s proposed volumes (option 3). This is shown in figure 2 below.
In this figure, switching from the current volumes to the Commission’s lower volumes (that see 26 million fewer units available below the cost containment reserve from 2024 to 2030 in the model) leads to a negligible increase in NZU prices over the remainder of the decade.
Instead, the main adjustment to the reduced auction volumes in the model comes through:
Demonstration pathway baseline: Increased afforestation (from 45k hectares [ha] to 54k ha per annum) along with increased trading and use of units held in private accounts (from 125 million NZUs to 135 million).
Current policy reference baseline: Increased trading and use of units held in private accounts (from 113 million NZUs to 138 million).
These relatively responsive shifts in afforestation and the use of units in private accounts (e.g., forest units held for harvest) generally reflect a rational market hypothesis, implying that the market can and will adjust efficiently to changes in auction volumes.
However, there is reason to be sceptical about the flexibility of the NZ ETS to adapt swiftly to changes in market settings in a real-world context, particularly in light of the muddled signals seen in the market in recent months.
So let’s proceed to stress-test the model’s assumptions for the liquidity of stockpile units in private accounts and afforestation response in turn
Sensitivity of NZU price to assumptions for trading of non-surplus units
The NZ ETS Market model includes default limits for the proportion of non-surplus units that can be traded in any given year. The consultation version of the model has its default parameter set at a limit of 20% of non-surplus units held in private accounts (within a default range of 10-30%).
The model then uses trading of these non-surplus units to meet shortfalls in supply. This assumption is based on a simple rational market view from economics. As in theory, a forester with units held for harvest could trade such units when demand is high relative to supply, benefiting from the short-term rise in NZU price, before then repurchasing units at a future point when prices are lower.2
The default parameters in the model also assume these units will be repaid in 25 years. While I am particularly sceptical of the payback timeframe assumption, let's keep things simple here and focus on comparing MfE’s default midpoint limit on trading of non-surplus units (20%) to a scenario where only 5% of non-surplus units can be traded and used by other participants. This is shown in figure 3 below, using the current legislated auction volumes as the assumed available supply from the Government.3
As seen in figure 3, reducing the extent to which the market can draw on non-surplus units in the model results in a significant increase in NZU price for the ‘current policy reference’ baseline. The increase in NZU price for the demonstration pathway baseline is somewhat less pronounced, with increased afforestation helping to keep price rises in check. Additionally, the reduced gap between gross emissions demand and available supply in the demonstration pathway baseline mitigates the need for a spike in prices to induce further gross emissions reductions.
Though not shown in the figure, switching to the Commission’s proposals for auction volumes led to significant NZU price increases when use of non-surplus units is limited. For instance, even under the demonstration pathway baseline, the average price from 2024-2030 rose to $150-200 for 10% and 5% limits on trading of non-surplus units.
So, what can we learn from this? Firstly, as mentioned in my earlier articles, assumptions regarding the extent to which non-surplus units are likely to be willingly traded at different NZU price levels remain critical to forming a view on the 2-6 year outlook of the NZ ETS.
Secondly, to my mind, these results suggest the need for a more nuanced assumption about the extent to which non-surplus units can be drawn upon, potentially differentiating between ordinary market fluctuations (e.g., swings of $20-30 in the last couple of years) and higher sustained upward movements.
Assumptions for afforestation
MPI recently published its annual Afforestation and Deforestation Intentions survey (ADS) (MPI, 2024). This year’s ADS hasn’t garnered as much attention as past years' surveys, which were conducted at the height of afforestation over the last couple of decades.
This year’s ADS shows a downward shift in future afforestation intentions, as well as a drop in afforestation achieved in 2023 (68,500 ha) from what was intended in the 2022 survey (88,000 ha) (MPI, 2023). The range of reasons for this decline in afforestation is vast, so those interested are encouraged to read the ADS report itself (link). However, two factors relevant to the current context are worth highlighting here.
First, recent drops in spot-market prices following the March auction, combined with signals in the consultation document that the Government is considering dropping the auction price floor, are likely to contribute to a more reserved investment environment for forestry. While some forest sites may still make a return at NZU prices just below $50, the low prevailing prices (which may persist in the near-term) are likely to dampen enthusiasm for afforestation.
Second, afforestation projects require at least a year or two of lead time. Land purchases, resource consents, financing, orders for seedlings, and securing labour all need to be arranged ahead of planting. As details are not yet known on how the Government will proceed with the National Party’s pre-election commitments suggesting land use restrictions, foresters and their investors have valid reasons to remain hesitant about near-term investments. This was well covered in a recent podcast interview with Ollie Belton (of Carbon Forest Services and Permanent Forests NZ), here.
Some carbon forest operators have arguably already pivoted to safer investments (such as securing contracts with farmers to integrate woodlots on their farms). However, for the purposes of this article, it would be brave to expect a large turnaround and influx of new forests over the next 2-3 years.4
Returning to MfE’s model, we can examine how different approaches to incorporating forestry within the model affect afforestation and NZU prices. The model allows for two approaches to incorporating forestry:
Using an endogenous response function derived from prior work by Bruce Manley (2018, 2019), which links afforestation rates to carbon and log prices.
Using fixed exogenous projections of afforestation (with the Government’s official projections released in late 2023 (MfE, 2023b) included in the model).
When both of these are used, high afforestation rates are seen relative to the surveyed intentions in the ADS.
Figure 4 below illustrates this by comparing afforestation rates in the 2023 ADS with calculated afforestation using the Manley equation. Afforestation rates from the 2023 ADS are broken down into ‘intended afforestation’ (i.e., what foresters responding to the survey stated they intend to plant), a 100% continuation rate (where, when foresters do not provide intentions for years further out, it is assumed they continue at 100% of the last year of their provided intentions), and a 50% continuation rate (as before, but at 50%) (MPI, 2024).
As highlighted in figure 4, afforestation in the NZ ETS model increases over the next few years relative to current surveyed intentions when afforestation rates are calculated endogenously within the model. This has the effect of reducing the gap between available supply and demand from gross emitters.
When switching to more conservative assumptions for forward afforestation, the model needs to turn to other levers (e.g., price rises, triggering the CCR volumes, and non-surplus units) to ensure that supply and demand are balanced each year.
Figure 5 below illustrates this by comparing NZU prices and stockpile use when afforestation is endogenous in the model to the 50% continuation rate from the 2023 ADS survey.5
In figure 5, higher NZU prices are observed for both the current policy reference and demonstration pathway baselines when afforestation is reduced. The effect is more pronounced for the current policy reference baseline due to the greater gap between latent gross emissions demand and supply.
I remain quite sceptical about the application of the Manley (2018; 2019) equation as an endogenous part of long-term optimisation models for a range of reasons.6 However, rather than delve into that topic here, the main takeaway point is that apparent low rates of afforestation over the next 2-3 years are quite likely to see price trends diverge from the default set-up of the model, and similarly for other optimisation models that draw direct causal links between carbon price and afforestation.
Do findings from the model align with your previous work?
I haven’t included any analysis using my own spreadsheet tool relied on for my previous two articles, as the latest forestry forecasts incorporated in this year’s March baseline update are not publicly available. Generally, a key difference between MfE’s NZ ETS Market model in its default set-up and my prior work relates to assumptions regarding the liquidity of post-1989 forest units held for harvest.
In MfE’s model, the use of non-surplus units—particularly when high limits on trading these units are allowed—enables the market to smooth out NZU price changes over time and avert scarcity of supply. When tighter limits are assumed for the ability to draw on non-surplus units (in either my prior work or MfE’s model), NZU prices increase.
When I attempted to replicate parameters in the NZ ETS Market model similar to those covered in my prior articles examining the Commission’s auction volumes,7 the model showed the surplus being drawn down after the 2027-2028 surrender year, compared with after the 2026-2027 surrender year in my earlier analysis (figure 2 here).
Summing it up
By covering the impacts of different (i) auction volume options; (ii) assumptions for the liquidity of non-surplus units; and (iii) afforestation rate assumptions, I hope I’ve highlighted for readers that analysis of market outcomes in the next ~2-6 years are as much a product of assumptions made around these second two variables as they are of the auction volumes themselves.
Regarding afforestation rates, it seems fairly clear that the near-term is likely to see lower rates of afforestation than those projected in the Government's or Commission’s recent work. This deserves close attention when considering both market trends over the next ~2-6 years and whether New Zealand meets its second emissions budget.
For the liquidity of non-surplus units, I won’t make a case here in favour of one assumption over another. However, the message should be clear that getting this assumption wrong can cause significant issues for market actors and regulators alike.
As for a policy assessment of the options, I don’t have any particular advice on which of the auction volume proposals being consulted on makes the most sense. The options provided in the consultation are challenging to comment on rigorously given that the Government’s released material (MfE, 2024a, 2024b) offers ambiguous details regarding its workings and assessment of each of the three options. Particularly for their consistency with meeting New Zealand’s emissions budgets. Nor are actual auction volumes provided for options 1 and 2 in the discussion document from what I can see.
There is certainly a higher likelihood of scarcity and rising prices towards the end of the decade if the Commission’s revised lower auction volumes are adopted. However, in the absence of a clear assessment of different options against New Zealand’s emissions budgets, it is hard to evaluate how to weigh up the options.
Auction floor proposals
The Government’s consultation document presented two options for the auction price floor:
Option 1 - status quo ($68 in 2025, rising to $83 in 2029).
Option 2 - lower price control settings (no levels stated).
I’ve seen speculation in the public domain that this second option could have been inserted at the last minute to appease a Cabinet colleague or coalition partner (Carbon News, 2024), or prompted by a desire to ensure revenue generation from auctions (Kivi, 2024). Both of these conjectures seem credible.
What it means for the remaining auctions this year and whether the remaining units available under the cost containment reserve (CCR) are cleared is a little harder to figure.
In theory, a lower auction floor price should align the auctions more closely with spot-market prices observed over the last few months and help mitigate risks of cash drag losses for volumes bought at auction that aren’t tagged to fixed contracts with emitters.
However, recent volatility in spot-market prices following the auction is partly a response to the Government’s consultation itself, and the market has consistently demonstrated its volatility in response to political signals. Therefore, with the ongoing surplus overhang in the NZ ETS, a lower auction price floor does not guarantee that spot markets won’t plumb lower depths.8
Considering that both the Commission and the Government have further annual decision points that could be used to inject additional liquidity back into the market to offset any failed auctions this year (if necessary), my personal view is that maintaining the status quo auction price would lead to a healthier outcome for the market. With the obvious benefit of providing a clearer signal to participants to get on with reducing emissions.
Chasing after spot-market prices influenced by the current surplus glut seems to be a fool's errand to me. Though I suspect I’d get short shrift putting such arguments to those preoccupied with short-term revenues.
Final parting comments
I don’t have any further NZ ETS posts planned in the near-term, as I’ll be focusing on other article ideas and pieces of work.
As for those other pieces of work, one update is worth mentioning, I’ve set up a consulting business, Kapiti Climate Insights Limited!
Feel free to reach out if there is work on carbon markets, climate change policy, consultation submissions, or technical analysis along the lines of any of my recent articles I can help your organisation out with. You can get in touch with me via LinkedIn or Substack.
References
Carbon News (2024) ‘ETS consultation suggests lower prices for NZUs’, 15 May. Available at: https://www.carbonnews.co.nz/story.asp?storyID=31530 (Accessed: 9 June 2024).
Climate Change Commission (2023a) Advice on the direction of policy for the Government’s second emissions reduction plan. Wellington, New Zealand: Climate Change Commission, p. 193. Available at: https://www.climatecommission.govt.nz/our-work/advice-to-government-topic/advice-for-preparation-of-emissions-reduction-plans/2023-advice-to-inform-the-strategic-direction-of-the-governments-second-emissions-reduction-plan-april-2023/ (Accessed: 14 March 2024).
Climate Change Commission (2023b) ‘Advice on the direction of policy for the Government’s second emissions reduction plan: Updated demonstration path and current policy reference scenarios (2022)’. Available at: https://www.climatecommission.govt.nz/our-work/advice-to-government-topic/advice-for-preparation-of-emissions-reduction-plans/2023-advice-to-inform-the-strategic-direction-of-the-governments-second-emissions-reduction-plan-april-2023/ (Accessed: 14 March 2024).
George, D. and Belton, O. (2024) ‘Carbon Forest Services Ltd founder Ollie Belton’. (REX Podcast feed). Available at: https://omny.fm/shows/rural-exchange/carbon-forest-services-ltd-founder-ollie-belton (Accessed: 8 June 2024).
Green, K. (2024a) ‘Is the Climate Change Commission setting the NZ ETS up for a bumpy ride in the 2020s?’, Kapiti Climate Insights (Substack), 18 March.
Green, K. (2024b) ‘When will the NZ ETS surplus reach zero?’, Kapiti Climate Insights (Substack), 1 April.
Kivi, L. (2024) ‘Govt blamed for low carbon price’, Carbon News, 22 May. Available at: https://www.carbonnews.co.nz/story.asp?storyid=31591 (Accessed: 9 June 2024).
Manley, B. (2018) ‘Forecasting the effect of carbon price and log price on the afforestation rate in New Zealand’, Journal of Forest Economics, 33(1), pp. 112–120. Available at: https://doi.org/10.1016/j.jfe.2017.11.002.
Manley, B. (2019) Impacts of carbon prices on forest management. MPI Technical Paper 2019/13. Wellington, New Zealand: Ministry for Primary Industries, p. 37. Available at: https://www.mpi.govt.nz/dmsdocument/37113-Impacts-of-carbon-prices-on-forest-management (Accessed: 1 June 2024).
Ministry for Primary Industries (MPI) (2023) Afforestation and deforestation intentions survey 2022. MPI Technical Paper 2023/09. Wellington, New Zealand: Ministry for Primary Industries, p. 26. Available at: https://www.mpi.govt.nz/dmsdocument/57130-Afforestation-and-Deforestation-Intentions-Survey-2022 (Accessed: 9 June 2024).
Ministry for Primary Industries (MPI) (2024) Afforestation and deforestation intentions survey 2023. MPI Technical Paper 2024/14. Wellington, New Zealand: Ministry for Primary Industries, p. 28. Available at: https://www.mpi.govt.nz/dmsdocument/62313-Afforestation-and-Deforestation-Intentions-Survey-2023 (Accessed: 9 June 2024).
Ministry for the Environment (MfE) (2023a) Review of the New Zealand Emissions Trading Scheme: summary of modelling. ME 1785. Wellington, New Zealand: Ministry for the Environment, p. 46. Available at: https://environment.govt.nz/publications/review-of-the-new-zealand-emissions-trading-scheme-summary-of-modelling/. (Accessed: 1 March 2024).
Ministry for the Environment (MfE) (2023b) ‘Updated emissions projections to 2050 released’. Available at: https://environment.govt.nz/assets/what-government-is-doing/climate-change/2050-historical-and-projected-sectoral-emissions-data-November_2023-for-publishing-v01.xlsx (Accessed: 29 March 2024).
Ministry for the Environment (MfE) (2024a) Annual updates to New Zealand Emissions Trading Scheme limits and price control settings for units 2024. Consultation document ME 1828. Wellington, New Zealand: Ministry for the Environment, p. 39. Available at: https://environment.govt.nz/publications/annual-updates-to-new-zealand-emissions-trading-scheme-limits-and-price-control-settings-for-units-2024-consultation-document/ (Accessed: 8 June 2024).
Ministry for the Environment (MfE) (2024b) Updates to the New Zealand Emissions Trading Scheme Market Model. ME 1832. Wellington, New Zealand: Ministry for the Environment, p. 7. Available at: https://environment.govt.nz/assets/publications/climate-change/ETS-Settings-Updates-to-the-ETS-Market-Model-May-2024.pdf (Accessed: 6 June 2024).
There are other matters consulted on in the consultation document that could affect auction volumes, even if option 1 - status quo is adopted. But we’ll omit discussing these for brevity.
Similar logic can be applied for other non-surplus units too (e.g., hedged volumes).
I’ve chosen to use the current legislated auction volumes with each variable to ensure a common status quo / baseline comparison.
With the slow early growth of forests in their first few years, an uptick in afforestation towards the end of the decade isn’t particularly consequential for the market or New Zealand’s emissions budget for 2025-2030.
I’ve chosen the 50% continuation rate as it serves as a general midpoint in the 2023 ADS survey range and provides an effective low rate of afforestation to stress-test outcomes from the model with.
Though this should not be read as a knock on the underlying equation itself, which was not created with the intention of its use in such models.
Using the Commission’s proposed auction volumes, the ‘current policy reference’ baseline, a 5% limit on trading of non-surplus units’ and MPI’s central projections for afforestation.
Nor are any decisions likely to be seen until after the next auction on 19 June.