The Paris Agreement and its impact on cattle and food sectors of New Zealand

ABSTRACT

The Paris Agreement asserts that greenhouse gas emission pathways should be consistent with holding the increase in global temperature below 1.5 °C, or 2 °C above pre-industrial levels. The purpose of this paper is to assess the economic impact of this agreement on the cattle and food product sectors of New Zealand. We used a general equilibrium approach to evaluate the economic impacts, and the Global Timber Model to estimate forestry carbon sequestration. We simulated eight scenarios where we allow accounting/not accounting for sequestration, pricing/not pricing agricultural emissions, and linking/not linking the New Zealand Emissions Trading Scheme (ETS) with the European Union ETS. We found that significant negative impacts occur if sequestration is not accounted, the ETS remains unlinked and agriculture is priced. Competitiveness, in turn, is not significantly affected if sequestration is accounted, regardless of the linking scheme of the ETS.

KEYWORDS:

• Agriculture
• emissions trading scheme
• forestry carbon sequestration
• general equilibrium
• global timber model
• linking
• New Zealand
• pricing

JEL CODES:

• C68
• Q51
• Q54
• Q56

Introduction

Negotiations towards a new international climate change agreement under the United Nations Framework Convention on Climate Change concluded in Paris in December 2015. The Paris Agreement (PA) asserts that future greenhouse gas (GHG) emission pathways should be consistent with holding the increase in the global average temperature below 1.5 °C, or 2 °C above pre-industrial levels. The PA is due to enter into force by 2020 and seeks for global emissions to peak as soon as possible and then to undertake rapid reductions thereafter. It also requires each country to adopt its own intended nationally determined contributions (INDCs), which will reflect each country’s ambition for reducing emissions, taking into account domestic circumstances and capabilities (UNFCC 2015). Although INDCs are not yet enough to keep global warming below 2 °C, the PA traces the way to achieving this target (European Commission 2016).

New Zealand committed to reduce GHG emissions to 30% below 2005 levels by 2030 (59.2 million tonnes of carbon dioxide equivalent [Mt CO₂e]) and also announced a target of reducing emissions to 50% of 1990 levels by 2050 (33.4 Mt CO₂e) (Ministry for the Environment 2015b). Thus, the purpose of this paper is to assess the impact of the PA on two of the agricultural sectors of New Zealand, namely, cattle (beef, sheep, goats and horses) and food products (meat products, dairy, oils, rice, sugar, beverages and tobacco). We used a general equilibrium approach to evaluate the economic impacts, and we used the global timber model (GTM) to estimate forestry carbon sequestration (FCS). We simulated eight scenarios where we allow accounting/not accounting for FCS to calculate net emissions, pricing/not pricing agricultural emissions, and linking/not linking the ETS with the EU ETS.

The paper is structured as follows: the modelling approach, the simulation results, and the conclusion.

Modelling approach

In the following, we present the quantitative framework of our analysis. We first introduce the modelling approach and then the scenarios for simulation.

Climate and trade dynamic general equilibrium model

To estimate the economic impacts we used the climate and trade dynamic general equilibrium (CliMAT-DGE) model developed by Landcare Research. CliMAT-DGE is a multi-regional, multi-sectoral, forward-looking dynamic general equilibrium model with a relatively long time horizon of 100 years or more. This model is suited to studying the efficient (re)allocation of resources within the economy and the response over time to resource or productivity shocks. CliMAT-DGE primarily uses the Global Trade Analysis Project (GTAP) version 8 data set. The base year of the benchmark projection is 2007. The model then develops a benchmark projection of the economic variables and GHG emissions, and simulates scenarios to evaluate the impacts of mitigation policies. Based on long-run conditions and constraints on physical resources, which restrict the opportunity set of agents, the model predicts the behaviour of the economy, energy use, and emissions by region and sector (Fæhn et al. 2013).

CliMAT-DGE covers 18 aggregated production sectors; this paper focuses on the cattle and food sectors. All production sectors are modelled using nested Constant Elasticity of Substitution production functions. These functions represent the potential substitution between production technologies and inputs. The nesting structure in CliMAT-DGE partly follows Paltsev et al. (2005). Model dynamics follow a forward-looking behaviour where decisions made today about production, consumption and investment are based on future expectations. The economic agents have perfect foresight and know exactly what will happen in all future periods of the time horizon. Thus, households are able to smooth their consumption over time so that the savings rate varies endogenously, that is, savings are determined within the model (Dellink 2005; Babiker et al. 2008). In common with other general equilibrium models, international asset positions and the financial sector are not explicitly modelled in CliMAT-DGE. Financial stocks and flows of financial assets (debt, equity, currency) are not modelled either. Thus, while a current account deficit is financed by a capital account surplus, we cannot say anything about the composition of the capital account. Foreign trade allows countries to temporarily run foreign account imbalances in response to environmental policies, as long as that imbalance is made up for in later years (Babiker et al. 2008). In addition, labour market closure depends on fixed labour supply where all policy shocks are accommodated through changes in wages. For a thorough description of CliMAT-DGE, see Fernandez & Daigneault (2015).

Global timber model

GTM is an economic model capable of examining global forestry land-use, management, and trade responses to policies. In responding to a policy, the model captures afforestation and forest management, and avoids deforestation behaviour. The model estimates harvests in industrial forests and inaccessible (virgin) forests, timberland management intensity, and plantation establishment, which are all important components of both future timber supply and carbon flux. The model also captures global market interactions, global timber supply, and the associated carbon accounting, including carbon stored in harvested wood products (Fernandez & Daigneault 2015).

GTM tracks more than 200 forest types across 17 timber regions. The New Zealand region includes 12 regional Pinus radiata and other exotic forest plantation areas as well as native forest. It solves in 10-year increments to 2150, taking into account the long-run dynamics of forest growth and harvest schedules. The model has been used in a variety of forest and climate change policy assessments internationally (Daigneault et al. 2012). More details on GTM can be found in Sohngen & Mendelsohn (2003).

For this analysis, we feed the regional carbon prices estimated with CliMAT-DGE into GTM. GTM can then estimate the change in regional forest stock and FCS as a result of the proposed emission reduction targets. The endogeneity effect of the carbon price affects FCS, which then reduces required emission reductions from other sectors of the economy, which would then lower the carbon price coming out of CliMAT-DGE that we ideally feed back into GTM until we converge to a steady state GHG price to achieve the reduction target pathway using both models. This approach we refer to as a soft-link.

Policy scenarios

CliMAT-DGE develops a baseline scenario where the global economy is projected from the base year of 2007 to 2082, in 5 year periods, in the absence of mitigation policies for climate change. The impacts from the PA are analysed in terms of deviations (or percentage changes) of the variables of interest relative to the baseline. We imposed caps on the baseline emission pathways so that emissions from the rest of the world followed trajectories consistent with temperature increases of 1.5 °C by 2100, whereas New Zealand follows trajectories consistent to the submitted INDCs.

We define eight policy scenarios that analyse the effects, under the PA, of accounting/not accounting the FCS to calculate national GHG net emissions; pricing/not pricing agricultural emissions under the ETS, that is, agriculture does not participate in the ETS; and, linking/not linking the ETS to the EU ETS. Below we provide some background on the definition of the scenarios.

FCS is one of the most cost-effective domestic abatement options in New Zealand (Qi et al. 2013; Ministry for the Environment 2015a). In 2013, almost a quarter of New Zealand’s gross emissions (21.1 Mt CO₂e) were offset by FCS (Ministry for the Environment 2015c). Unlike previous environmental agreements, FCS is acknowledged in the PA as a potential source of GHG abatement. Thus, we subtract FCS from gross GHG emissions where the resultant net emissions have to be mitigated through other alternatives (e.g. the ETS). As foresters’ decisions on forest management are influenced by prices in the ETS, then accounting FCS may lower the required abatement from the economy and the corresponding impacts.

New Zealand is in a unique position as a developed country because of its unusual emissions profile: agricultural non-carbon dioxide emissions (e.g. methane and nitrous dioxide) make up about half of the country’s gross emissions (Ministry for the Environment 2015c). Although since January 2012 the agricultural sector has to report their emissions under the ETS, currently there is no legislated date for when agricultural emissions will be priced (Climate Change Information 2012). Thus, meeting the PA may become costly if GHG abatement has to rely on only a few sectors, i.e. manufacturing and energy.

Finally, linking the ETS with the EU ETS means that firms participating in the ETS can import relatively cheaper permits from the EU and use them to comply with GHG abatement policies in New Zealand. Thus, at least in theory, linking the ETS would lower the overall cost of meeting the PA targets. However, linking may lead to changes in the competitiveness positions of sectors participating in the NZ ETS, which may not necessarily be welfare enhancing for the economy as a whole (Flachsland et al. 2009a, 2009b).

Results and discussion

Baseline output levels for cattle and raw milk (2012 figures) are NZ$8.9 billion and NZ$18 billion, respectively. Imports of cattle reach $0.17 billion and food products $6.9 billion. Exports of cattle reach NZ$0.62 billion and exports of food products reach NZ$33.25 billion. Table 1 shows the impacts on GDP.¹

Table 1. GDP impacts of alternative policy scenarios in 2030 (% change with respect to baseline).

	No FCS		FCS
	No linking	EU linking	No linking	EU linking
Agriculture priced	−4.94	−5.01	−0.96	−0.96
Agriculture not priced	−7.12	−5.41	−0.58	−0.58

If FCS is not accounted and agriculture is priced, GDP decreases by 5% relative to the baseline where the slightly greater impact from EU linking is because of competitiveness effects. If agriculture is not priced, impact on GDP is higher because abatement must rely on fewer sectors and the pool for non-GHG emissions from agriculture is not available. On the other hand, if FCS is accounted, the GDP impacts are less than 1% below the baseline regardless of whether agriculture is priced or not; moreover, linking with the EU ETS does represent any difference on impacts.

Table 2 shows that if FCS is not accounted for and agricultural emissions are not priced, negative impacts occur for both cattle and raw milk production, while linking with the EU ETS does not help to mitigate the losses. However, accounting for FCS alleviates the stringency of the reduction target and linking with the EU ETS does not make a significant difference on the impacts on output. On the other hand, if agricultural emissions are not priced, small decreases in output occur if the ETS remains unlinked, whereas linking with the EU ETS leads to increases of 3% above the baseline for both sectors. However, if FCS is accounted for, deviations with respect to the baseline output are less than 0.2% in absolute value.

Table 2. Impacts of output of alternative policy scenarios in 2030 (% change with respect to baseline).

	No FCS		FCS
	No linking	EU linking	No linking	EU linking
Agriculture priced
Cattle	−31.8	−32.7	−11.1	−11.1
Raw milk	−26.4	−26.6	−10.0	−10.0
Agriculture not priced
Cattle	−3.1	3.1	−0.01	−0.1
Raw milk	−2.3	3.8	0.2	0.1

Table 3 shows impacts on domestic prices, which are consistent with standard economic theory, that is, in the case where agriculture is priced and FCS is not accounted, prices of cattle and raw milk increase significantly, relative to the baseline, because of the lower supply shown in Table 1. Moreover, even if FCS is accounted there are still significant price increases in those sectors. Thus, the negative effects of pricing agriculture are only partially mitigated by FCS accounting. Linking with the EU ETS does not produce differences on the impacts on prices. On the other hand, if agriculture is not priced and FCS is not accounted, linking makes a difference on the direction of price changes; whereas if FCS is accounted, price changes are less than 0.2%, in absolute value, relative to the baseline.

Table 3. Impacts on domestic prices of alternative policy scenarios in 2030 (% change with respect to baseline).

	No FCS		FCS
	No linking	EU linking	No linking	EU linking
Agriculture priced
Cattle	119.1	118.0	44.3	44.3
Raw milk	33.8	33.5	13.1	13.1
Agriculture not priced
Cattle	2.0	−6.3	0.3	0.3
Raw milk	−0.7	−7.1	0.2	0.2

Table 4 shows that if FCS is not accounted for and agricultural emissions are priced, imports of cattle decrease almost 40% below the baseline if the ETS remains unlinked. That is, we find that there is no substitution between imported and domestic cattle production. Linking with the EU ETS mitigated decreases in the imports of cattle. In turn, imports of food products increase regardless of whether the ETS is linked with the EU. Exports of cattle decrease by 4% below the baseline if the ETS remains unlinked; this decrease is larger if the ETS is linked with the EU ETS. Similarly, exports of food products decrease significantly regardless of whether the ETS is linked with the EU ETS. On the other hand, if FCS is accounted for, cattle imports decrease by an average 48% below the baseline regardless of whether the ETS links with the EU ETS. Cattle exports increase significantly by 31% above the baseline whereas exports of food products decrease by 10%.

Table 4. Trade impacts of alternative policy scenarios in 2030 (% change with respect to baseline).

		No forestry		Forestry
		No linking	EU	No linking	EU linking
Agriculture priced
Imports	Cattle	−39.3	−18.4	−47.9	−47.9
	Food	47.4	47.7	14.5	14.5
Exports	Cattle	−4.2	−10.9	31.2	31.2
	Food	−26.8	−27.2	−9.8	−9.8
Agriculture not priced
Imports	Cattle	−2.4	−18.5	−0.7	−0.1
	Food	−6.0	−24.4	−1.1	−0.7
Exports	Cattle	−4.6	6.5	−0.8	−0.5
	Food	−3.2	9.0	0.1	−0.1

If agricultural emissions are not priced and FCS is not accounted for, imports of cattle and food products decrease further when the ETS is linked to the EU ETS, whereas exports increase. On the other hand, if FCS is accounted for, different scenarios have little effect, with variations of imports and exports about 1% in absolute value with respect to the baseline.

Table 5 shows sectoral impacts through the Revealed Comparative Advantage (RCA) indicator. The RCA examines the export specialisation pattern and compares the trade performance of an economic sector with the performance of all sectors within the region (Balassa 1965; Malmberg & Maskell 1997). The RCA indicator relates the ratio of a region’s exports in a specific sector over the world’s exports in this sector to the ratio of a region’s exports in all sectors over the world’s total exports (Alexeeva & Anger 2015). Baseline RCA for cattle is 16.96 and for food products is 17.17.

Table 5. Competitiveness impacts of alternative policy scenarios in 2030 (% change with respect to baseline).

	Revealed comparative advantage (%) vs business as usual
	No forestry		Forestry
	No linking	EU linking	No linking	EU linking
Agriculture priced
Cattle	25.1	9.4	42.3	42.3
Food	−17.0	−17.4	−5.0	−5.0
Agriculture not priced
Cattle	10.9	−0.2	2.5	1.6
Food	12.8	1.1	3.1	2.3

If FCS is not accounted for and agricultural emissions are priced, cattle gain competitiveness if the ETS remains unlinked. This gain weakens if the ETS is linked with the EU ETS. Food products lose competitiveness regardless of the linking scheme. In turn, if FCS is accounted for, cattle gain competitiveness whereas food competitiveness drops by 5% below the baseline. Linking with the EU ETS makes no difference. On the other hand, if agricultural emissions are not priced and FCS is not accounted for, cattle and food gain competitiveness if the ETS remains unlinked, but those gains almost vanish if linking with the EU ETS occurs. Finally, if FCS is accounted for, there are gains in competitiveness for both sectors, and linking with the EU ETS slightly weakens those gains.

Conclusions

In evaluating the impacts of the PA on cattle and food products, lowest negative impacts on GDP and output occur if agricultural emissions are not priced and FCS is accounted for when calculating New Zealand’s net emissions. If agricultural emissions are priced, accounting for FCS partially mitigates output losses. In agreement with economic theory, price increases occur if output of cattle and raw milk are negatively affected by the policy scenarios. Trade effects are complex and can only be evaluated through a computable general equilibrium setting. We could not find any scenario where exports of cattle and food are simultaneously increased. However, if agricultural emissions are not priced and FCS is accounted for, differences in exports with respect to the baseline are less than 1% in absolute value, and both sectors gain in competitiveness. Hence, although agriculture is not part of the ETS, accounting for FCS alleviates the stringency of the PA on the priced sectors (e.g. energy and manufacture).

Acknowledgements

The authors wish to thank Tommy Robertson, Suzie Greenhalgh, Pike Brown and three anonymous referees for helpful comments.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Results on GDP come from Fernandez & Daigneault (2016a). For further details see Fernandez & Daigneault (2016b).