Delinking the New Zealand Emissions Trading Scheme from the Kyoto Protocol: comparing theory with practice

ABSTRACT

The New Zealand Emissions Trading Scheme (NZ ETS) presents an opportunity to compare the theory of linked emissions trading with practice. From 2009 until October 2012 New Zealand was linked to the market under the 1997 Kyoto Protocol and there was no indication that this link would be broken. A series of events starting in late 2012 cast doubt on the future eligibility of Kyoto units in the NZ ETS, made the future of linking in New Zealand uncertain, and may have contributed to price divergence between offshore and domestic units. Delinking was officially confirmed by the New Zealand government in December 2013 to take effect from 31 May 2015. This raises the question of what the effect of delinking and announcements had on the carbon market. To study this effect, we used daily price data in a difference-in-differences model over the period 1 January 2011 to 1 January 2016. From this model, we found evidence that, even with differing pre- and post-time specifications, the government announcements in 2012 caused prices in the two markets – for New Zealand units (NZUs) and Kyoto units – to decouple and NZUs traded at a premium based on their projected scarcity. This is further backed up by the behaviour of the raw price data and the change in the types of units surrendered over this time period. The NZU price reaction to the government announcements (rather than delinking itself) shows strong evidence of a well-functioning market.

Key policy insights

• The NZU price reaction to the delinking announcement shows strong evidence of the New Zealand carbon market reacting in the present to news about the future: a property of any well-functioning market.

• The long period of price divergence between the New Zealand and Kyoto markets has led to a large privately held bank that is a liability to the New Zealand Government and reduces the government's ability to auction NZUs.

• If the NZ ETS had delinked earlier from the Kyoto market it could have avoided a protracted period of arbitrage which significantly inflated the NZU bank.

KEYWORDS:

• Climate change
• emissions trading schemes
• Kyoto Protocol
• carbon markets
• emission trading
• New Zealand

Acknowledgements

We would like to thank Nigel Brunel at OMF and Lizzie Chambers of Carbon Match for supplying us with the price data used in this paper. We would also like to thank Catherine Leining, Matt Thirkettle, and Anna Robinson for their helpful comments on earlier drafts. Finally thank you to the participants at the 2014 annual NZAE conference and at a presentation at the Ministry for the Environment for their comments. Naturally all errors, omissions, and opinions expressed are the responsibility of the authors.

Datasets

The data from this study has been sourced from OMF and Carbon Match. It is not publicly available, and we do not have the rights to release the data.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 After legislative amendments in November 2009 (enacted before non-forestry sectors assumed unit obligations and received free allocation), all NZ ETS participants could buy and surrender eligible Kyoto units from overseas but only forestry participants could sell New Zealand Units (NZUs) overseas via exchange for New Zealand Assigned Amount Units (AAUs). The rationale for prohibiting offshore sales from other sectors was the addition of a $25 fixed-price option to the NZ ETS; allowing offshore sales would have created a potential arbitrage opportunity for industrial firms continuing to receive free allocation on an output basis. In practice only a small number of NZ AAUs were ever sold overseas because buyers were limited, and so, for simplicity, in this paper we speak of New Zealand as having only a one-way link in the linked period. Buyers of New Zealand AAUs from forestry included a Japanese company and the Norwegian and Danish governments. As of 30 December 2014, 1.2 million units were sold overseas out of a total more than 146 million NZUs and AAUs that entered private holdings. Since 18 November 2015, international transfers of units to or from the New Zealand Emissions Trading Register (NZETR) have no longer been available (Environmental Protection Authority, 2020).

2 A full one-for-one unit obligation was phased in by the start of 2019.

3 Legislative amendments passed in 2012 required a cap to be set before any NZUs were auctioned. Legislative amendments passed in 2019 provided for auctioning under a cap to begin in 2021.

4 Some qualitative restrictions applied to CERs and ERUs that were eligible in the NZ ETS. From inception, the NZ ETS did not accept ERUs and CERs associated with nuclear projects or CERs associated with forestry (tCERs and lCERs). From December 2011, CERs associated with industrial-gas destruction projects involving HFC-23 and N₂O were prohibited. From December 2012, ERUs associated with industrial-gas destruction projects and both ERUs and CERs associated with large hydro projects were prohibited.

5 AAUs were allocated to Annex I countries at the beginning of each commitment period in relation to their Kyoto target. Imported AAUs were never accepted in the NZ ETS; NZ AAUs have been accepted in the NZ ETS but few have been issued to private holders.

6 NZ ETS participants could still surrender NZ AAUs (e.g. that were issued to participants in the Permanent Forest Sink Initiative, Projects to Reduce Emissions or Negotiated Greenhouse Agreements).

7 See (Aatola et al., 2013; Alberola et al., 2008; Hintermann, 2010; Lutz et al., 2013; Mansanet-Bataller et al., 2007; Rickels et al., 2007).

8 Forestry participants, including those with liabilities for deforestation, were included in the NZ ETS since 1 January 2008; however, stationary energy and liquid fuel participants, who are the major source of demand, did not enter until 1 July 2010. Hence, while we have price data beginning January 2008, there were few trades and reporting was inconsistent.

9 For example, the average difference in NZU price between the two platforms was about 2 cents, and not statistically significant. Throughout our analysis we only use the CommTrade Carbon data, both for simplicity, and because we have access to the data for a more recent period.

10 See the New Zealand Emission Unit Register for annual surrender and other data.

11 NZUs can leave or enter private holdings in two more ways: they can be transferred back to the New Zealand Government for repayment or transferred to participants as reimbursement. This usually occurs when an entity is allocated units and the size of the allocation is subsequently revised. This number is not reported in the NZETR Holding & Transaction Summary, but can be found in the annual reports on the EPA website. Unfortunately, these numbers are not broken down by unit type, but net reimbursements are around half a million, and so for our purpose small enough to be ignored when calculating the bank.

12 The information for the number of NZUs surrendered and cancelled can be found on the NZETR Holding and Transaction page. Information on the NZUs converted for overseas sale can be found on the NZU conversions page.

13 A small number of international AAUs (5,281,818) have entered private holdings from overseas transfers. International AAUs are no longer allowed to be surrendered in the NZ ETS so we do not want to include them in our estimate of the bank, and hence we exclude them from our calculations. These units may be counted in the outflows, through units transferred overseas, in which case we should include them in the inflow. At any rate, how we deal with these units is not too important relative to the size of the bank, as can be seen from Figure 4.

14 This is close to the government estimate for 1 July 2015 of 140 million (Ministry for the Environment, 2015).

15 Changes in total surrenders partly reflect changes in gross emissions (and the increase from 2011 to 2012 reflects the mid-year entry date of the stationary energy and liquid fuel sectors) but are mostly driven by surrender of units to cover harvesting or deregistration of forestry ETS participants. Forestry participants were able to take advantage of low Kyoto prices to reduce their liabilities and to deforest (Kerr et al., 2016).

16 The theoretical prediction is subject to the caveat that New Zealanders banked all their NZUs for future use, which as we will see in Figure 2, happened bar a trivial number of surrenders.

17 See the New Zealand Emissions Trading Register for annual surrender and other data.

18 We imagine that the small amount of NZUs surrendered reflect small operators facing relatively large transaction costs for trading. Since many viewed the cheap Kyoto units as non-additional, it is also possible that some non-profit-maximising participants may have surrendered NZUs in an attempt to be more ‘green’.

19 Environmental Protection Agency (2015, p. 13).

20 Ministry for the Environment (2013b, p. 100).

Additional information

Funding

This work was supported by a grant from the Aotearoa Foundation (New York City, USA).