## 1nzlea2023002

## Source details

**Canonical URL:** [1nzlea2023002](https://www.imf.org/-/media/files/publications/cr/2023/english/1nzlea2023002.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2023/english/1nzlea2023002.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2023/english/1nzlea2023002.pdf.json)

---

### Historical and projected greenhouse gas (GHG) emissions and targets
- Gross GHG emissions in 2021: about 76.8 million metric tons of CO2-equivalent (MtCO2e), about 4 percent below the 2019 level.
- Net emissions in 2021: about 0.6 percent lower than in 2019 and 3.1 percent higher than in 2020, as carbon captured by LULUCF declined.
- Projection path:
  - Net emissions projected to rise until around 2030, then stabilize and decline rapidly as recently planted forests mature and LULUCF absorption increases.
  - Updated projections: New Zealand is on track to meet its 2050 commitments and is almost matching the ambitious scenario for biogenic methane abatement.
  - Gap to 2030 NDC narrowed from 24 MtCO2e (estimated in 2022) to 17 MtCO2e, but additional abatement remains required to meet the 2030 NDC.
- Sector profile:
  - Agriculture (methane) is the largest source: around 62 percent of net emissions (Net emissions in 2020, MtCO2e).
  - Energy and transport are the next most significant sources.
  - Electricity generation largely zero carbon (hydropower), but energy use remains fossil-fuel dependent, particularly transport.
  - Vulnerability: reliance on hydropower increases exposure to droughts, which could force fossil-fuel use unless renewables scale up.

### Emissions pricing, ETS developments, and market risks
- Secondary market and auction price path:
  - Secondary market peak: NZ$88.50 per NZ unit (1 tCO2e) in mid-November 2022; fell to NZ$76 per unit by end-December 2022.
  - Official auction clearing price: NZ$79 in December 2022.
  - Continued fall in 2023: end-May 2023 market price NZ$55 per unit; after ETS review announcement secondary market price fell to NZ$41.
- Auction dynamics and failures:
  - March 2023 ETS auction failed to clear (no new units released); unsold units rolled into June auction, which also did not clear; unsold units rolled into September. If September and December auctions also fail to clear, unsold units will expire.
  - Concerns for auction failures and price weakness: policy uncertainty, possible over-provision of units for auction, CCC concerns about an overhang of allocated NZUs.
- ETS review (government options presented during public comment period ending early August 2023) includes:
  - Decrease units offered at auction.
  - Increase demand by allowing government and foreign investor purchases.
  - Introduce restrictions/conditions on removal activities.
  - Bar emitters from purchasing units directly from foresters.
- CCC recommendations (July 2022) versus government decisions (December):
  - CCC recommended reserve price increase to NZ$60 (from NZ$30 in 2022); cost containment reserve trigger prices of NZ$171-214 (vs NZ$70 in 2022); make 16.3 million units available for sale (2022: 26.3 million NZUs, including cost containment reserve).
  - Government set reserve price NZ$33.06, cost containment reserve trigger price NZ$80.64, and make 17.9 million units available for auction.
  - Government’s more modest 2023 settings were below CCC recommendations and below secondary market prices at announcement; market and auction prices converged and declined rapidly thereafter.
- Key risks and financing implications:
  - Continued emission price declines risk undermining ERP and NDC achievement.
  - Auction failures reduce proceeds intended for climate investments, risking mitigation and adaptation plans.
  - Need for a robust and rising price floor to meet targets; ETS adjustments may be required to align with emissions budgets.
  - Mechanisms under development to fund public mitigation/adaptation and crowd in private investment: further boost to CERF, increase capital of NZGIF, issuance of green bonds.

### Emissions Reduction Plan (ERP), emissions budgets, and role of the ETS
- ERP and budgets:
  - ERP (released June 2022) sets emissions budgets up to 2035.
  - First budget: 2022–25 total emissions restricted to 290 MTCO2e, a reduction of around 11.5 MtCO2e relative to 2017–21.
  - Subsequent budgets: 2026–30 and 2031–35.
- Role of ETS and mix of measures:
  - ETS envisaged to contribute around one-third of emissions reduction in the first emissions budget period using an average price of NZ$75 per NZ unit; other cross-sectoral and sectoral initiatives expected to deliver remaining two-thirds.
  - Energy and industry, the ETS, transport, and in some scenarios agriculture are the biggest contributors to emissions reduction.
- ERP’s five ETS focus areas:
  - Ensure ETS unit limit and price controls consistent with emissions budgets and the NDC.
  - Adjust ETS to balance net and gross emissions reductions.
  - Develop market governance and regulatory framework to protect participants and ensure reliable price signals.
  - Address emissions leakage: domestic overallocation to industry and external competitiveness concerns.
  - Support voluntary carbon market and develop complementary market policy framework.

### Policy scenarios, CPAT analysis, and price-impact assessment
- IMF CPAT scenario: raise real emissions price from 2023 levels to US$100 (NZ$165.40) by 2030.
- Impact estimate:
  - Around two-thirds of the gap between New Zealand’s projected emissions and its NDC for 2030 could be closed by this higher price scenario.
- Caveats:
  - Scenario is ambitious: requires real emissions price to more than double in 7 years and may be politically difficult; nonetheless confirms price-centered approach can deliver substantial gains.

### Ensuring ETS drives gross (not just net) emissions reductions
- Concern: ETS may be biased toward afforestation/net-emissions abatement rather than driving gross emissions reductions.
  - CCC modelling: afforestation becomes less costly under net-emissions focus, allowing continued emissions in industry, energy, and transport while investing in offsets.
  - Risk: key sectors may reach net zero (expected late-2030s on current trends) but a national gap to 2050 path could remain if ETS does not incentivize further mitigation.
- Surplus and overhang:
  - CCC estimates a surplus of 144 million units issued and banked; restricting supply of new units is suggested to address surplus which otherwise depresses prices and weakens the ETS.
- Government response and timeline:
  - March 2023 review of ETS to ensure better market functioning and greater emphasis on gross reductions; review also to manage exotic forestry (non-native species).
  - First recommendations for public consultation scheduled in Q2 2023; proposals to strengthen incentives and manage exotic forest planting expected to be rolled out by end-2024.

### Pricing agriculture emissions: proposals, design, and implications
- Timeline and proposals:
  - First proposals October 2022; revised December 2022.
  - Government set 2025 as start date for pricing agriculture emissions (He Waka Eke Noa consultations); formal proposals expected later in 2023.
- December 2022 proposal highlights:
  - Farm-level, split-gas levy pricing biogenic methane and nitrous oxide separately, imposed at individual farm level; interim processor-level levy proposed only if farm-level system not operational by 2025.
  - Emissions assessed indirectly through reportable metrics (e.g., herd sizes) with adjustments for emissions-mitigating measures.
  - Levies apply to farms above threshold (~200 tCO2e per year).
  - Initial levies set low with a 5-year path and review after 3 years; nitrous oxide price capped so sector would not face higher price than if it had entered ETS at that point.
  - Payments for sequestration; compensation would transition to ETS prices; sequestration strategy to account for and reward sequestration.
  - Levy revenues to be redistributed to farmers via incentive and sequestration payments and a dedicated fund for Māori farmers.
- Unresolved detail: no specific starting price announced for biogenic methane and nitrous oxide.
- International and CPAT context:
  - IMF staff: global methane price of US$70 per tCO2e consistent with holding global temperature rises to under 2 ̊C.
  - CPAT preliminary: assuming ETS price returns to peak in real terms by 2030, a real methane price of US$5 per tCO2e would get New Zealand to its NDC for 2030; higher prices needed for further improvement to 2050.
- Policy design recommendations for agriculture pricing:
  - Use proceeds to support farmers adopting emissions-reducing technologies and practices.
  - Careful price-setting for incentive and sequestration payments to ensure fiscal sustainability; reimbursement converging to ETS prices quicker than levy could create unfunded contingent liabilities.

### Climate Emergency Response Fund (CERF), ERP/NAP financing and investments
- CERF establishment and flows:
  - CERF established in Budget Policy Statement for Budget 2022 with initial down payment NZ$4.5 billion (projected ETS cash proceeds 2022/23–2025/26).
  - Government adjusted CERF size in Economic and Fiscal Updates based on 4-year rolling ETS cash forecasts.
  - Budget 2022: NZ$2.9 billion allocated to implement first ERP, with NZ$1.5 billion immediate.
  - Subsequently government topped up CERF by net NZ$2.1 billion and allocated NZ$5.7 billion in gross climate-related spending.
  - Government confirmed commitments will be supported in event of shortfall in ETS auction proceeds.
- Financing instruments and capacity:
  - Green bonds: first issuance November 2022, NZ$3 billion securities maturing in 2034 placed.
  - NZGIF capital increased by NZ$300 million under Budget 2023, bringing it to NZ$700 million; NZGIF invested over NZ$160 million to-date.
  - NZ Super announced intention to allocate NZ$100 million to suitable climate-related investments.
- Investment need estimates:
  - CCC (2021): additional NZ$10.3 billion between 2021–2030 in five sectors (road transport, electricity generation, energy efficiency in buildings, food processing, native afforestation) to achieve Commission’s 2021 advice.
  - Boston Consulting Group (2022): by 2030, real investment of NZ$42 billion needed to support rapid scaled-up electrification and switch to zero-carbon energy sources (large share to replace and upgrade transmission infrastructure).
- Policy actions to strengthen financing:
  - Build on CERF, green bonds, NZGIF.
  - Strengthen responsible investment framework in public sector.
  - Support high-quality investment decisions, improve climate reporting and accounting.
  - Make it easier for savers to avoid fossil-fuel investments.
  - Implement Carbon Neutral Government program and use procurement to create demand for green products/services.
- Potential financial instruments and innovation:
  - Expand sustainability-linked bonds and loans, sustainability/social bonds, green asset-backed securities, and other instruments.
  - Sustainability-linked instruments can supplement ETS price mechanisms for projects without climate-driven cash flows (e.g., adaptation).
  - Regulatory efforts on disclosures and data sharing could support financial innovation and investment in climate start-ups.

### Offshore abatement and international cooperation
- Offshore abatement included in New Zealand’s NDC; authorities exploring offshore mitigation partnerships in Asia Pacific, but no decisions yet.
- Near-term usefulness limited due to underdeveloped offshore abatement market and lack of robust international accounting/verification standards; CCC (2022) recommended setting offshore mitigation at zero in near-term.
- New Zealand climate finance commitments 2022–25:
  - Intend to invest NZ$1.3 billion in climate projects, at least 50 percent to the Pacific region.
  - At least 50 percent of NZ$1.3 billion to be set aside for adaptation projects.
- Rationale: Pacific Island Countries face capacity constraints in preparing bankable projects; New Zealand can assist given its experience.

### Adaptation framework, National Adaptation Plan (NAP), and implementation capacity
- NAP cycle and governance:
  - 6-year cycle: risk assessment (NCCRA), preparation/adoption of NAP, biannual reviews.
  - First NCCRA (2020) identified 43 critical risks; future NCCRAs to be undertaken by CCC, next completion slated for 2026 feeding into second NAP by 2028.
  - CCC to prepare biennial progress reports on government actions’ effectiveness.
- First NAP (2022–28) goals:
  - Reduce vulnerability to climate impacts.
  - Enhance adaptive capacity and integrate climate into decisions at all levels.
  - Strengthen resilience to climate change.
- NAP actions:
  - Total actions: 111 actions—39 critical, 50 supportive, 22 proposed—to be implemented over 6-year Plan period.
  - Coordination framework within government proposed to support implementation.
  - Emphasis on filling information gaps and building institutional frameworks to support multi-level decision-making (local governments, regions, private sector).
- Gaps and priorities:
  - NAP does not yet set out framework for assessing climate readiness of critical infrastructure—identified as priority.
  - Need to advance planning processes in light of rebuilding after Cyclone Gabrielle and January Auckland floods.
  - Need to link NAP actions to Budget framework to ensure progress.
- Adaptation areas of action (eight focal areas): risk-informed decisions; climate-resilient development location planning; adaptation options (avoid/protect/invest/retreat); natural environment; homes/buildings/places; infrastructure; communities; economic/financial system (accounting/disclosure improvements).

### Costs, damages, and fiscal implications of extreme weather
- Historical partial estimate (2007–17): droughts and flooding events costs (including income foregone) estimated at NZ$840 million (0.4 percent of average annual GDP over that period), NZ$720 million from droughts alone.
- Early-2023 extreme weather events damage and rebuilding estimated at NZ$9-14.5 billion (2.2-3.6 percent of GDP).
- Authorities’ forward estimates:
  - Increasing drought frequency: GDP in 2061 could be 0.5 percent lower than historical counterfactual.
  - More frequent storms/flooding: GDP could be 0.7 percent lower relative to counterfactual.
  - More frequent storms and droughts would leave net Crown debt around 3.8 percent of GDP higher.
- International context: IPCC estimates a 1.5–2 ̊C rise by 2100 would lower global GDP by 0.3–0.5 percent.
- Recommendation: further work to refine estimates to build political consensus.

### Key conclusions and policy recommendations (summarized)
- Emissions and pricing:
  - Progress made but gap remains between 2030 NDC and projected path.
  - Allowing emissions prices to rise further would contribute to greater reductions, but relying solely on price increases likely requires politically difficult higher prices.
  - Recent decline in emissions prices is concerning for both emissions outcomes and auction proceeds; sustained low prices could make ERP/NDC targets out of reach.
  - Policy recommendations:
    - Address weaker price sentiment and bolster credibility of emissions budgets with detailed policies.
    - Allow recent policy changes that raise fossil-fuel demand to expire as planned; if needed, replace with targeted support.
    - Consider tightening planned supply of new NZ units in ETS to signal commitment and address overhang of previously issued and banked units.
- Agriculture pricing:
  - Proposed framework welcome but details remain; initial proposed price is too low to effect needed changes—allowing prices to rise gradually is appropriate.
  - Current recommendations commit to bringing nitrous oxide to parity with ETS but leave out biogenic methane.
  - Goal: price all forms of agriculture emissions consistent with NDC, with flexible timing to build political consensus.
  - Recycle proceeds to support on-farm investment and sequestration; reimbursement levels must be sustainable.
- Rebuilding and adaptation:
  - NAP is expansive; urgency heightened by early-2023 disasters.
  - Preliminary estimate: around 700 homes assessed unsafe to rebuild; government/local councils examining buyout options.
  - Need to strengthen cost assessments for central government responsibilities and integrate adaptation package costs into Budget.
- Financing and private participation:
  - Financing needs exceed available resources; create mechanisms and instruments to crowd in private investment and innovation.
  - Sustainability-linked instruments could complement ETS price mechanisms, especially for adaptation.
  - Identify and remove barriers to adoption of existing tools/instruments.
- Limitations and further work:
  - Need framework to assess climate readiness of critical infrastructure.
  - Refine domestic and global cost estimates and incorporate global spillovers.
  - Develop robust data, modeling, verification, and monitoring standards to make offshore abatement viable.
  - Continue strengthening disclosures, accounting, and verification to support financial innovation.

### Labor market constraints, wage dynamics, and inflation linkages (summary)
- Labor market tightness:
  - Historically tight labor market with record high participation and historically low unemployment/underemployment; job vacancies very high and underemployment declined.
  - These dynamics have put upward pressure on wages and short-term wage growth expectations, feeding into inflation.
  - Inflation was 6.7 percent y/y in 2023Q1; tradeable inflation easing while non-tradeable inflation remains high and sticky.
- Wage measures:
  - LCI (Labor Cost Index): declined to 1.6 percent during pandemic; rose to 4.3 percent in 2023Q1 (RBNZ target 1–3).
  - QES average hourly earnings: increased 7.6 percent in 2023Q1, with services and construction particularly rapid.
- Baseline wage Phillips curve (2000–2019) (selected coefficients):
  - Unemployment gap: -0.626*** (0.0953)
  - Expected inflation: 0.386*** (0.0826)
  - Trend labor productivity growth: 0.408* (0.215)
  - Constant: 2.130*** (0.0555)
  - Observations: 80; R-squared: 0.371
- Alternative NAIRU-gap specification (selected):
  - Unemployment gap: -0.473*** (0.0523)
  - Expected inflation: 0.285*** (0.0956)
  - Constant: 2.160*** (0.0495)
  - Observations: 80; R-squared: 0.507
- Extended specifications:
  - Lagged net migration (4 quarters) significant negative impact on wage growth (effects highest 4–8 quarters).
  - Lagged export prices associated with higher wage growth after ~4 quarters.
  - Underemployment/underutilization inclusion does not materially change results.
- Cross-country panel (23 countries):
  - Higher unemployment relative to trend associated with lower wage growth; New Zealand wages somewhat less sensitive to unemployment than cross-country average.
- Labor market institutions and wage sensitivity:
  - Interaction analysis: greater collective bargaining share and longer contract length associated with wages less sensitive to unemployment and inflation expectations.
  - New Zealand institutional context:
    - Trade union participation <20 percent; collective bargaining share 18.6 percent in 2020 versus >60 percent in Australia.
    - Fair Pay Agreements Act (December 2022) will increase sectoral bargaining role and likely contribute to less volatile wage growth going forward.
- Minimum wage and distribution:
  - As of April 1, 2023:
    - Adult minimum wage (before tax): NZ$22.70/hour.
    - Starting-out and training: NZ$18.16/hour.
  - Minimum wage growth:
    - 2019–2023 average growth ~6.6 percent yearly.
    - Real minimum wage ~55 percent higher than OECD average; grew ~70 percent since 2000 (OECD average +54 percent).
    - Minimum wage binding moved from 10th percentile (2001) to 30th percentile (2022).
  - Pass-through estimates:
    - A one percent increase in minimum wages would lift hourly real wage by 0.5 percentage point on average.
    - Effects stronger for women: at median a one percent minimum increase: 0.74 percentage point for women vs 0.5 percentage point for men.
- Industry, gender, ethnicity patterns:
  - Industries closest to minimum: services, community services, retail, accommodation, agriculture, transport.
  - Women earn consistently less than men across industries; average gap relative to minimum: women NZ$8.6/hr above minimum, men NZ$11.9/hr above minimum (women earn about 88 cents for each dollar men earn).
  - Ethnic disparities: Pacific peoples, Māori, and Asian women earn less on average than European peers.
- Industry-level empirical evidence (2009–2022):
  - Main regression (log average − log median on log minimum − log median) shows positive and significant coefficients (examples):
    - Coefficient on Log(min. wage) − log(median): 0.587*** (col 1), 0.564*** (col 2), 0.822*** (col 3), 0.364*** (col 4), 0.408*** (col 5), 0.332*** (col 6).
  - Quadratic terms show mixed signs and significance, indicating nonlinearity in some specifications.
  - Robustness: results hold with lag controls; OLS may bias upward; lack of granular lower-tail data limits direct tests below median.
- Policy implications:
  - Minimum wage increases uplift low-wage earners but spill over beyond binding percentile, potentially contributing to overall wage pressures.
  - Minimum wage policy can reduce inequality at lower tail and between genders in some cases but is insufficient alone to address ethnic wage gaps.
  - Interaction with collective bargaining and contract length matters for wage dynamics and volatility.

*Source: IMF staff chapter on Addressing Climate Change—Updates and Emerging Challenges and related sections (provided content).*

### 1. Historical and Projected Greenhouse Gas Emissions and Targets ____________________ 4

### 1. Historical and Projected Greenhouse Gas Emissions and Targets

### Overview and recent developments
- Gross GHG emissions in 2021 were estimated to be about 76.8 million metric tons of CO2-equivalent (MtCO2e), about 4 percent below the 2019 level.  
- Net emissions in 2021: about 0.6 percent lower than in 2019 and 3.1 percent higher than in 2020, as the carbon captured by land use, land use change and forestry (LULUCF) declined.  
- Net emissions are projected to rise until around 2030 before stabilizing and then decline rapidly as recently planted forests mature and absorption from LULUCF increases.  
- Updated projections indicate New Zealand is on track to meet its 2050 commitments and is almost matching the ambitious scenario for biogenic methane abatement.  
- The gap to the 2030 Nationally Determined Contribution (NDC) has narrowed from 24 MtCO2e estimated in 2022 to 17 MtCO2e, but additional abatement is still required to meet the 2030 NDC.

### Profile of emissions and sectoral drivers
- Methane from agriculture remains the most significant GHG emitted in New Zealand.  
- Sectoral shares and notable figures (Net emissions in 2020, MtCO2e): agriculture accounts for around 62 percent of net emissions. Energy and transport are the next most significant sources.  
- Electricity generation is largely zero carbon due to reliance on hydropower, but overall energy use remains dependent on fossil fuels, particularly in the transportation sector.  
- Vulnerability: reliance on hydropower increases exposure to droughts, which could force reliance on fossil fuels unless renewables are scaled up significantly.

### Emissions pricing, market developments, and risks
- Secondary market prices peaked at NZ$88.50 per NZ unit (1 tCO2e) in mid-November 2022, then fell to NZ$76 per unit by end-December 2022. The official auction clearing price declined to NZ$79 in December 2022.  
- Prices continued to fall in 2023: by end-May 2023 the market price stood at NZ$55 per unit; after the ETS review announcement the secondary market price fell to NZ$41.  
- The March 2023 ETS auction failed to clear (resulting in no new units released); the unsold units were rolled into the June auction, which also did not clear; unsold units were rolled into September. If September and December auctions also fail to clear, unsold units will expire.  
- Concerns cited for auction failures and price weakness: policy uncertainty, possible over-provision of units for auction, and independent Climate Change Commission (CCC) concerns about an overhang of allocated NZUs.  
- The government launched an ETS review (public comment period ending early August) presenting options including: decrease units offered at auction; increase demand by allowing government and foreign investor purchases; introduce restrictions/conditions on removal activities; bar emitters from purchasing units directly from foresters.

### Emissions Reduction Plan (ERP), budgets, and ETS role
- The ERP (released June 2022) sets emissions budgets up to 2035: first budget covers 2022-25 (total emissions under first budget restricted to 290 MTCO2e, a reduction of around 11.5 MtCO2e relative to the 2017-21 period); subsequent budgets cover 2026-30 and 2031-35.  
- The ETS is envisaged to contribute around a third of the emissions reduction in the first emissions budget period using an average price of NZ$75 per NZ unit; other cross-sectoral and sectoral initiatives expected to deliver the remaining two-thirds. Energy and industry, the ETS, transport, and in some scenarios agriculture are the biggest contributors to emissions reduction.  
- The ERP proposes five ETS focus areas:
  - The ETS settings: ensure unit limit and price controls are consistent with emissions budgets and the NDC.  
  - Adjust ETS to balance net and gross emissions reductions.  
  - Market governance of the ETS: develop robust governance and regulatory framework to protect participants and ensure reliable price signals.  
  - Address the risk of emissions leakage: domestic overallocation to industry and external competitiveness concerns.  
  - Voluntary market framework: support a voluntary carbon market and develop a market policy framework to complement the ETS.

### Policy scenarios and impact of higher emissions prices
- Staff used the IMF Climate Policy Assessment Tool (CPAT) to assess a higher real emissions price scenario: increase from 2023 levels to US$100 (NZ$165.40) by 2030.  
- The revised assessment suggests around two-thirds of the gap between New Zealand’s projected emissions and its NDC for 2030 could be closed by this higher price scenario.  
- The analysis notes the scenario is ambitious: it requires the real emissions price to more than double in 7 years and may be politically difficult to deliver, but confirms that a price-centered approach can deliver substantial gains.

### ETS settings, CCC recommendations, and government decisions
- CCC (July 2022) recommended:
  - Reserve price increase to NZ$60 (from NZ$30 in 2022).  
  - Cost containment reserve trigger prices of NZ$171-214 (compared to NZ$70 in 2022).  
  - Make a total of 16.3 million units available for sale (2022: 26.3 million NZUs, including the cost containment reserve).  
- Government decisions (December): set a reserve price of NZ$33.06, a cost containment reserve trigger price of NZ$80.64, and make 17.9 million units available for auction.  
- The government’s more modest 2023 price and unit supply settings are below CCC recommendations and were already below secondary market prices at the time of announcement; subsequent convergence of market and auction prices and rapid price declines followed.

### Key risks and financing implications
- Continued declines in emissions prices risk undermining the achievement of emissions reductions envisaged by the ERP and NDC.  
- Failure of auctions reduces proceeds intended for climate investments, potentially putting planned mitigation and adaptation investments at risk.  
- A robust and rising price floor is identified as necessary to ensure the government’s targets can be met; ETS adjustments may be required to align settings with emissions budgets.  
- Mechanisms to fund public mitigation and adaptation investments and to crowd in private investments and risk sharing are being developed, including a further boost to the Climate Emergency Response Fund, an increase in capital of New Zealand Green Investment Finance, and issuance of green bonds to finance climate projects.

*Source: IMF staff chapter on Addressing Climate Change—Updates and Emerging Challenges (provided content).*

### 11.      A related issue that remains is the need to ensure the ETS drives gross emissions

### 1nzlea2023002 - 11.      A related issue that remains is the need to ensure the ETS drives gross emissions

### Emissions Trading System (ETS): gross versus net emissions
- Concern: the ETS may be biased toward afforestation and net emissions abatement rather than driving gross emissions reductions.
- Modelling by the CCC suggests afforestation becomes a less costly option under net-emissions focus, allowing continued emissions in industry, energy, and transport while investing in offsets.
- Risk: when key sectors reach net zero (expected by the late-2030s on current trends), a national gap against the path to the 2050 targets could remain if the ETS does not sufficiently incentivize further mitigation.
- Emitters outside the ETS (e.g., agriculture) and those benefiting from industrial allocations of units may not face pressure to reduce emissions under current policies.
- Surplus issue: the CCC estimates there is a surplus of 144 million units issued and banked that needs to be addressed by restricting the supply of new units by some other means; these surplus units act as an overhang and could depress prices, weakening the ETS.
- Government response: announced in March 2023 a review of the ETS to ensure better market functioning and greater emphasis on gross reductions of GHG emissions; the review also intends to manage the amount of “exotic forestry” (planting of non-native plant and tree species).
- Timeline: the review is scheduled to put forward its first recommendations for public consultation in the second quarter of 2023; proposals to strengthen incentives in the ETS and manage exotic forest planting are expected to be rolled out by end-2024.

### Pricing agriculture emissions: proposals and design features
- Timeline and process:
  - First proposals released in October 2022; revised proposal published in December 2022.
  - Government had set 2025 as the start date for pricing agriculture emissions to guide consultations under He Waka Eke Noa.
  - Formal proposals are expected to be introduced later in 2023.
- Key highlights of December 2022 proposals:
  - Price based on a farm-level, split-gas levy, pricing biogenic methane and nitrous oxide separately, imposed at the individual farm level; an interim processor-level levy proposed only if farm-level system could not be operationalized by 2025.
  - Emissions assessed indirectly through reportable metrics such as herd sizes, with adjustments for emissions-mitigating measures (use of feedstock, integrated farming techniques).
  - Levies apply to farms meeting a pre-determined emissions threshold (approximately 200 tCO2e per year).
  - Initial levies set at a low level with a path for 5 years and a review after 3 years; the price of nitrous oxide will be capped so the sector would not face a higher price than if it had entered the ETS at that point.
  - Payments available for those undertaking sequestration; compensation would transition to ETS prices; a sequestration strategy would be formulated to account for and reward sequestration efforts.
  - Revenues from the levy redistributed to farmers through incentive and sequestration payments, and a dedicated fund for Māori farmers.
- Unresolved: no specific starting price announced for biogenic methane and nitrous oxide.

### Implications of agriculture pricing and price level considerations
- If implemented, New Zealand would be the first country to introduce a direct price on agriculture emissions; experience could serve as reference internationally.
- IMF staff work indicates a global methane price of US$70 per tCO2e is consistent with holding global temperature rises to under 2 ̊C.
- For New Zealand’s NDC: preliminary CPAT estimates suggest assuming the ETS price returns to its peak in real terms by 2030, a real methane price of US$5 per tCO2e would get New Zealand to its NDC for 2030; higher prices would be needed for further improvement to 2050 targets.
- Policy design recommendations:
  - Use of proceeds from levies should support farmers adopting technologies and practices that reduce emissions (feedstock, integrated farming techniques).
  - Careful price-setting for incentive and sequestration payments is critical for fiscal sustainability; a reimbursement scheme that converges to ETS prices quicker than the levy could create a mismatch where outflows outpace revenues, generating unfunded contingent liabilities.

### Climate Emergency Response Fund (CERF), ERP, and sectoral policies
- CERF establishment and funding:
  - CERF established in the Budget Policy Statement for Budget 2022 with an initial down payment of NZ$4.5 billion (projected cash proceeds from ETS auctions between 2022/23-2025/26).
  - Government has regularly adjusted CERF size at each Economic and Fiscal Update based on four-year rolling forecast of ETS cash proceeds.
  - In Budget 2022, NZ$2.9 billion allocated to implement the first ERP, with NZ$1.5 billion made available immediately.
  - Since initial setup, government topped up CERF by a net total of NZ$2.1 billion and has allocated NZ$5.7 billion in gross climate-related spending.
  - Government confirmed commitments will be supported in the event of a shortfall in ETS auction proceeds.
- ERP cross-sectoral goals and focus areas:
  - Reform planning and infrastructure to improve resource management, support emissions reduction, strengthen financing, and promote innovations.
  - Encourage research, science, innovation and technology to generate and adopt options to reduce or capture emissions.
  - Equitable transition measures to ensure emissions reduction is compatible with economic growth and improved living standards.
  - Support circular economy and bioeconomy activities to strengthen resource management and address biodiversity and other environmental concerns.
- Sector-specific policies (expected impacts):
  - Transport: promote mode-sharing and accelerate electrification.
  - Energy and industry: invest in renewables, support electrification, upgrade energy efficiency.
  - Building and construction: lower embodied emissions, improve energy efficiency (including Building Code amendments), shift from fossil fuels.
  - Agriculture: price agriculture emissions, accelerate mitigation technology development and adoption, support producers to invest, and transition to lower-emissions land uses and systems.
  - Forestry: maintain role as GHG sink while addressing biodiversity and equity concerns.
  - Waste: waste contributes to 4 percent of New Zealand’s total emissions; focus on reducing organic waste, diverting from landfills, increasing emissions capture, and strengthening governance of waste management licensing.
  - Fluorinated gases: encourage alternatives and regulate to eliminate the most potent gases.
- Expected sectoral policies impact: total emissions reduction of 6.2-13.3 MtCO2e over the period of the first emissions budget on top of the impact of the ETS.

### Shifts in announced policies, fiscal trade-offs, and emissions implications
- Reprioritization in 2022 to address cost-of-living: several changes with implications for emissions if maintained.
  - 36 percent cuts in the fuel excise and road user charges announced in March 2022 have been extended but expected to expire in July 2023.
  - Scrap-and-replace scheme wound down to focus on immediate household concerns.
  - 50 percent discount for public transport extended to June 2023, to be replaced on July 1 by a more targeted scheme prioritizing community services cardholders and Total Mobility users.
  - Biofuel mandate scrapped.
- Staff assessment:
  - Impacts of deviations from previously announced policies judged limited: scrap-and-replace would mainly accelerate purchases by households who would have bought ZEVs anyway; public transit subsidies likely reduce emissions only marginally; cuts in fuel excise and road user charge likely have limited impact given expiry in July 2023.
- Scenario analysis:
  - If cuts in fuel excise and road user charge announced in March 2023 are made permanent, emissions by 2033 would wipe out improvements from a sustained increase in the ETS price, resulting in almost no emissions reductions.
  - Staff estimate making the tax cuts permanent will result in about a 6.7 percent increase in gross emissions relative to the ETS baseline by 2033.

### Adaptation framework and National Adaptation Plan (NAP)
- Governance and cycle:
  - Climate adaptation framework centered on a 6-year cycle: risk assessment, preparation and adoption of a National Adaptation Plan (NAP), and biannual reviews.
  - First National Climate Change Risk Assessment (NCCRA) prepared in 2020 identified 43 critical risks and assessed urgency.
  - Future NCCRAs to be undertaken by the CCC, with the next one slated to be completed in 2026; this will feed into the second NAP to be prepared by the government by 2028.
  - CCC will prepare biennial progress reports on effectiveness of government actions.
- First NAP (covering 2022-28): three goals
  - Reduce vulnerability to impacts of climate change.
  - Enhance adaptive capacity and consider climate change in decisions at all levels.
  - Strengthen resilience to climate change.
- Eight areas of action in the NAP and focal issues:
  - Enabling better risk-informed decisions: strengthen data, tools and methodologies to assess exposure and vulnerability.
  - Driving climate-resilient development in the right locations: integrate data and tools into planning and development; ensure climate-resilient public housing.
  - Adaptation options, including managed retreat: identify options—avoid, protect, invest to accommodate, or retreat when costs are too high.
  - Natural environment: restore native ecosystems and avoid spread of invasive species to reduce climate pressures and maintain biodiversity.
  - Homes, buildings and places: integrate climate risks into siting and construction; strengthen planning framework for resilience.
  - Infrastructure: ensure new infrastructure incorporates climate considerations and review capacity of existing infrastructure.
  - Communities: support vulnerable communities; develop strategies, guidance and tools to ensure economic, cultural and social wellbeing.
  - Economic and financial system: improve accounting and disclosure standards to enable better risk assessments and possible pricing of risks.

*Source: 1nzlea2023002 - 11.      A related issue that remains is the need to ensure the ETS drives gross emissions (https://www.imf.org/-/media/files/publications/cr/2023/english/1nzlea2023002.pdf).*

### 23.      Measures to build capacity to implement the NAP are underway. The Plan sets out 3

### 23.      Measures to build capacity to implement the NAP are underway. The Plan sets out 3

### NAP implementation framework and actions
- The Plan categorizes actions into: critical, supporting, and proposed.
- Total actions: 111 actions—39 critical, 50 supportive and 22 proposed—are to be implemented over the 6 years of the Plan period.
- The NAP proposes to set up a coordination framework within the government to support implementation.

### Focus on information, institutional frameworks, and planning
- Emphasis is skewed towards filling information gaps and building institutional frameworks to support multi-level decision-making (local governments, regions, private sector).
- Central government role: long-term planning and developing decision-support tools; more direct roles expected in infrastructure planning and development, and provision of healthcare and social services.
- Gaps noted:
  - The NAP does not yet set out a framework for assessing the climate readiness of critical infrastructure — identified as a priority.
  - Need to advance planning processes in light of rebuilding after Cyclone Gabrielle and January floods in Auckland.
  - Need to link NAP actions to the Budget framework to ensure progress.

### Accounting for the cost of climate change: findings and projections
- Partial historical estimate: costs associated with droughts and flooding events between 2007-17, including income foregone, are estimated at NZ$840 million (0.4 percent of average annual GDP over that period), of which NZ$720 million is estimated to have come from droughts alone.
- Early-2023 extreme weather events damage and rebuilding costs estimated at NZ$9-14.5 billion (2.2-3.6 percent of GDP).
- Authorities’ forward estimates:
  - Increasing frequency of droughts would lead to GDP in 2061 to be 0.5 percent lower than a counterfactual based on historical precedent.
  - A scenario with more frequent storms and flooding was expected to lower GDP by 0.7 percent relative to the counterfactual.
  - More frequent storm and drought events would leave net Crown debt around 3.8 percent of GDP higher.
- International context: the International Panel on Climate Change estimates that a 1.5-2  ̊C increase in average global temperatures by 2100 would lower global GDP by 0.3-0.5 percent.
- Recommendation: further work to refine these estimates to build political consensus.

### Financial system measures and disclosures
- Financial Sector (Climate-related Disclosures and Other Matters) Amendment Act 2021 requires around 200 large financial firms to provide climate-related disclosures per External Reporting Board guidelines.
- Reserve Bank of New Zealand (RBNZ) has started incorporating climate change impacts into its financial supervision framework and included climate disaster scenarios in its latest annual stress testing exercise.
- Plans to extend disclosure expectations to government at all levels and the private sector regarding natural hazards.

### Financing climate action: current instruments and gaps
- Public instruments and initiatives:
  - Climate Emergency Response Fund (CERF).
  - Issuance of green bonds: first issuance in November 2022, when NZ$3 billion in securities maturing in 2034 were placed.
  - New Zealand Green Investment Finance (NZGIF) established in 2019; capital increased by NZ$300 million under Budget 2023, bringing it to NZ$700 million.
  - NZGIF has invested over NZ$160 million to-date in renewables, energy efficiency, transport and agriculture projects.
  - NZ Super has announced that it would look to allocate NZ$100 million to suitable climate-related investments.
- Investment need estimates:
  - CCC (2021) estimate: between 2021-2030, an additional NZ$10.3 billion increase in investment in 5 sectors (road transport, electricity generation, increasing energy efficiency in buildings, food processing and native afforestation) is needed to achieve emissions targets in the Commission’s 2021 advice.
  - Boston Consulting Group (2022) estimate: by 2030, there would need to be a real investment of NZ$42 billion to support rapid, scaled up electrification and switch to zero-carbon energy sources (large share to replace and upgrade transmission infrastructure).
- Policy actions identified in ERP and NAP to strengthen financing framework:
  - Build on CERF, green bonds, NZGIF.
  - Strengthen responsible investment framework in public sector, particularly Crown Agencies.
  - Support high-quality investment decisions.
  - Improve climate reporting and accounting.
  - Make it easier for individual savers to avoid investment in fossil fuels and other polluting industries.
  - Implement the Carbon Neutral Government program.
  - Use procurement rules to create demand for green investment products and services.

### Potential instruments and financial innovation
- Consider expanding instruments: sustainability-linked bonds and loans, sustainability and social bonds, green asset-backed securities, green asset-backed securities, and other financial investments.
- Sustainability-linked instruments could supplement ETS price mechanisms for projects without climate-driven cash flows (e.g., adaptation).
- Regulatory efforts to enable better disclosures and data sharing could support financial innovation.
- Investments in climate start-ups and innovations could complement government R&D and create positive externalities.

### Offshore abatement and international cooperation
- Offshore abatement incorporated into New Zealand’s NDC; authorities exploring options for offshore mitigation partnerships in the Asia Pacific region, but no decisions yet.
- Near-term usefulness limited due to underdeveloped offshore abatement market and lack of robust international accounting and verification standards — CCC (2022) recommended setting emissions reductions from offshore mitigation at zero in the near-term.
- New Zealand’s climate finance commitments:
  - Between 2022-25, New Zealand intends to invest NZ$1.3 billion in climate projects, of which at least 50 percent will go to the Pacific region.
  - At least 50 percent of this NZ$1.3 billion is to be set aside to support adaptation projects.
- Rationale: Pacific Island Countries face considerable difficulties accessing climate finance due to lack of capacity in preparing bankable project portfolios; New Zealand can assist given its experience in accounting for and verifying emissions.

### Key conclusions and policy recommendations
- Emissions and pricing
  - New Zealand is making progress on reducing emissions but a gap remains between the 2030 NDC and projected emissions path.
  - Allowing emissions prices to rise further than currently projected would contribute to greater emissions reductions, but relying solely on price increases would likely require politically difficult higher prices.
  - Recent decline in emissions prices is a concern; sustained low prices could make envisaged emissions reductions out of reach and reduce auction proceeds intended for climate investments.
  - Policy recommendations:
    - Address weaker sentiment on prices and bolster credibility of emissions budgets with fleshed-out policies.
    - Allow recent policy changes that raise demand for fossil fuels to expire as planned and, if needed, replace them with targeted support.
    - Consider tightening planned supply of new NZ units into the ETS to signal commitment and address overhang of previously issued and banked units.
- Agriculture pricing
  - Proposed framework for pricing agriculture emissions is welcome but details remain:
    - Initial proposed price is too low to effect needed changes; allowing prices to rise gradually is appropriate.
    - Current recommendations commit to bringing nitrous oxide to parity with the ETS but leave out biogenic methane.
    - Goal should be pricing all forms of agriculture emissions consistent with New Zealand’s NDC, with flexible timing to build political consensus.
    - Recycling proceeds from agriculture levies to support on-farm investment and sequestration is appropriate.
    - Reimbursement levels should be sustainable; overly generous reimbursement relative to levy levels risks making the scheme unsustainable and harming long-term viability.
- Rebuilding and adaptation
  - NAP is expansive and links NCCRA risks to sector-level actions, but urgency to flesh out specific policies is heightened as rebuilding from early-2023 natural disasters begins.
  - Preliminary estimate: around 700 homes assessed as being unsafe to rebuild; government and local councils are examining buyout options.
  - Need to strengthen cost assessments for central government responsibilities (replacement of utilities and roads) and integrate adaptation package costs into the Budget.
- Financing and private participation
  - Financing needs exceed available resources; setting up financing mechanisms and new instruments can crowd in private investment and innovation.
  - Sustainability and sustainability-linked instruments could complement price mechanisms, especially for adaptation.
  - Identify and remove barriers to adoption of existing tools and instruments.

### Limitations and areas for further work
- Need for a framework to assess climate readiness of critical infrastructure.
- Further refinement of domestic and global cost estimates of climate change and stronger incorporation of spillovers from global economic conditions.
- Development of robust data collection, modeling, verification, and monitoring standards to make offshore abatement a viable pathway for the NDC.
- Continued strengthening of disclosures, accounting, and verification to support financial innovation.

*Source: 1nzlea2023002 - 23.      Measures to build capacity to implement the NAP are underway. The Plan sets out 3 (PDF chapter/section).*

### 1.      Significant labor market constraints and high wage growth are feeding into inflation,

### 1.      Significant labor market constraints and high wage growth are feeding into inflation

### Labor market conditions and inflation dynamics
- Labor markets in New Zealand are historically tight, with record high labor force participation and historically low levels of unemployment and underemployment rates.
- Job vacancies are at very high levels and underemployment and underutilization have gone down as hiring workers has become increasingly challenging.
- These dynamics have put upward pressure on wages and short-term wage growth expectations, increasingly feeding into inflation.
- High inflation has persisted since the second half of 2021 and is increasingly driven by non-tradeables.
- Inflation was 6.7 percent y/y in 2023Q1.
- When inflation first picked up in the second half of 2021 it was mostly driven by a spike in international food and energy prices and supply chain and shipping disruptions; pressures from food and commodity prices are easing and tradeable inflation has declined, while non-tradeable inflation has remained high and sticky.

### Key empirical findings on wages and inflation
- Wage inflation (Labor Cost Index, LCI):
  - LCI declined to a low of 1.6 percent during the pandemic.
  - LCI increased to 4.3 percent in 2023Q1, above the Reserve Bank of New Zealand’s (RBNZ) inflation target of 1–3.
- Quarterly Employment Survey (QES) measure:
  - Average hourly earnings (QES) increased 7.6 percent in 2023Q1, with services and construction showing particularly rapid wage growth.
- International context:
  - New Zealand’s wage inflation in 2022 was on par with other advanced economies but higher than in Australia.
  - Median wage inflation in peers was about 4 percent.
  - Both Australia and New Zealand had similar unemployment gaps at end-2022, but wage pressures were higher in New Zealand.

### Wage Phillips curve estimates (New Zealand)
- Baseline quarterly wage Phillips curve specification (2000–2019) relates y-o-y wage growth (LCI) to unemployment gap, 1-year ahead inflation expectations, and trend labor productivity growth.
- Table 1 (selected coefficients, HP filter gap; robust standard errors in parentheses):
  - Unemployment gap: -0.626*** (0.0953)
  - Expected inflation: 0.386*** (0.0826)
  - Trend labor productivity growth: 0.408* (0.215)
  - Constant: 2.130*** (0.0555)
  - Observations: 80
  - R-squared: 0.371
- Table 1 (selected coefficients, NAIRU gap):
  - Unemployment gap: -0.473*** (0.0523)
  - Expected inflation: 0.285*** (0.0956)
  - Constant: 2.160*** (0.0495)
  - Observations: 80
  - R-squared: 0.507
- Interpretation:
  - A decline in the unemployment gap is associated with higher wage growth.
  - Higher expected inflation is associated with higher wage growth.
  - Trend labor productivity is positively associated with wages, though less strongly.

### Extended wage Phillips curve (additional controls)
- Adding lagged migration and export prices (quarterly data):
  - Lagged net migration (4 quarters): -0.623*** (0.177) in one specification; -0.375** (0.150) and -0.968*** (0.160) in others (see Table 2).
  - Lagged export prices: 1.100*** (0.390), 1.240*** (0.394), 1.096*** (0.357), 0.954 (standard error reported) in various specifications (see Table 2).
  - Selected model fit statistics:
    - Observations: 72 or 80 depending on specification.
    - R-squared ranges: 0.583 to 0.760 across specifications.
- Findings:
  - Lagged migration (4 quarters) has a statistically significant negative impact on wage growth, with effect highest between 4 and 8 quarters.
  - Higher export prices are associated with higher wage growth after around 4 quarters.
  - Inclusion of underemployment and underutilization rates in New Zealand does not have a large or significant impact in these specifications.

### Cross-country panel results and comparison with peers
- Panel wage Phillips curve estimated for 23 countries (quarterly, 2000–2019) using HP-filtered unemployment gap:
  - Higher unemployment relative to trend is associated with lower wage growth; higher inflation expectations associated with higher wage growth.
  - New Zealand wages are somewhat less sensitive to unemployment compared to the cross-country average.
- Sample notes:
  - Data assembled for a group of 24 advanced economies for descriptive analysis; panel estimation uses 23 countries (Greece excluded).
  - Harmonized wage price indices are not available for all countries; various national sources used (Wage Price Index in Australia, LCI in New Zealand, Labor Cost Indices for Europe, etc.).
- Table 3 (summary):
  - Cross-country baseline: unemployment gap negative and significant.
  - New Zealand and Australia: unemployment gap remains negative and significant but with smaller coefficients than the cross-country panel, suggesting wages in NZ and AU are less sensitive to labor market tightness than the average among advanced economies.
  - Results robust to inclusion of lags of the dependent variable, q-o-q wage growth, and NAIRU-based gaps.

### Role of labor market arrangements: collective bargaining and contract length
- Literature and theoretical expectations:
  - “Corporatism” view (Bruno and Sachs; Calmfors and Drifill): higher collective bargaining coverage can moderate wage demands as unions internalize macroeconomic implications of wage increases, especially for supply shocks.
  - Recent literature: decline in union density and collective bargaining may have contributed to a flattening Phillips curve because workers with lower bargaining power were less able to negotiate large wage increases in tight markets.
  - Contract length: longer contract lengths lead to stickier wages that react less to domestic and external conditions.
- Cross-country evidence (summary from paper):
  - Greater share of collective bargaining and longer contract length is associated with wages being less sensitive to unemployment and inflation expectations.
  - Collective bargaining is more prevalent in Australia, which may help explain higher wage pressures in New Zealand relative to Australia at the current juncture despite similarly tight labor markets in both countries.
  - The passage of the Fair Pay Agreements Act in New Zealand is discussed in the context of lessons for the future.

### Minimum wage and wage distribution (preview of Section D)
- Industry-level hourly wage analysis indicates:
  - Minimum wages in New Zealand, which have risen rapidly over the past decade and sit above the OECD average, have a positive pass-through to overall wages and could shift the entire wage distribution to the right (higher).
  - Increases in the minimum wage can reduce overall inequality for low earners, but do not address inequality between genders and ethnic groups.

### Robustness and additional notes
- Baseline specification excludes the COVID-19 pandemic period to avoid noise; results hold when including the pandemic period (until 2023Q1).
- Results are robust to alternate wage measures (QES vs LCI), inclusion of lags of the dependent variable, q-o-q growth specifications, and NAIRU-based unemployment gaps.
- For trend unemployment, HP filter with smoothing parameter of 14400 used for cross-country comparability; NAIRU from WEO used as robustness.

*Source: IMF staff analysis in chapter "1. Significant labor market constraints and high wage growth are feeding into inflation" (New Zealand).*

### 14.      To assess the role of labor market arrangements in wage dynamics, we modify the

### 1nzlea2023002 - 14. To assess the role of labor market arrangements in wage dynamics, we modify the

### Methodology and Model
- Modified Phillips curve estimated with interaction terms to allow heterogeneity based on labor market characteristics (equation 3).
- Estimated specification (equation 3):
  - πw i,t = αi + αt + β1 u i,t gap + γ (Xi * u i,t gap) + β2 π i,t e + β3 g i,t LP + ε i,t
  - Xi captures labor market institutions: (i) average share of workers under collective bargaining agreements, (ii) average length of collective bargaining agreements, (iii) trade union density.
- Data: country-quarter panel; regressions include country and time fixed effects; standard errors clustered at the country level.
- Tables reported:
  - Table 4: interactions of unemployment gap with collective bargaining share, average contract length, and trade union density.
  - Table 5: interactions of expected inflation with the same institutional variables.
  - Lagged dependent variables included in columns 1–4; omitted in columns 5–8.

### Key Empirical Findings — Unemployment Interactions (Table 4)
- Unemployment gap coefficients remain negative and significant across specifications:
  - Examples: Unemployment gap coefficients reported include -0.515***, -0.715***, -0.364**, -0.720*** in columns with lags; -1.608***, -2.030***, -1.274***, -2.027*** in columns without lags.
- Interaction effects indicate less sensitivity of wages to unemployment where institutional measures are stronger:
  - Coll Bargaining * U gap: positive coefficients reported, e.g., 0.00485**, 0.00156, 0.0155***, 0.00657 (standard errors shown).
  - Avg length of contract * U gap: positive and significant coefficients, e.g., 0.0189***, 0.0148*, 0.0535***, 0.0349*.
  - Trade union density * U gap: mixed/weaker results, examples 0.00522, 0.00108, 0.0205*, 0.00516.
- Interpretation:
  - Negative unemployment gap coefficient combined with positive interactions implies that higher collective bargaining share and longer contract length are associated with wages being less sensitive to changes in unemployment.
  - Length of contract is most robustly associated with reduced wage sensitivity to unemployment when all three interactions are included together.

### Key Empirical Findings — Inflation Expectations Interactions (Table 5)
- Expected inflation coefficients positive and significant across specifications:
  - Examples: 0.292***, 0.322***, 0.286***, 0.323*** in lagged regressions; 1.110***, 1.135***, 1.206***, 1.112*** without lags.
- Interaction effects with expected inflation:
  - Coll Bargaining * Inf exp: negative coefficients in some specifications, e.g., -0.00327* and -0.0156**, -0.0145** indicating that greater collective bargaining is associated with less wage sensitivity to inflation.
  - Avg length of contract * Inf exp: negative and significant in several specifications, e.g., -0.0195**, -0.0182**, -0.0632**, -0.0436**.
  - Trade union density * Inf exp: mixed results, e.g., -0.00107, 0.00736*, -0.0165, 0.0132.
- Interpretation:
  - Greater collective bargaining and longer contracts are associated with less sensitivity of wages to inflation developments.

### Implications for New Zealand: Institutional Context and Recent Dynamics
- Trade union density and collective bargaining in New Zealand:
  - Trade union participation currently stands at less than 20 percent in New Zealand.
  - Share of collective bargaining in New Zealand stood at 18.6 percent in 2020 versus above 60 percent in Australia.
  - Historical note: before the 1980s the share of collective bargaining was close to 70 percent in New Zealand; legal changes in the late 1980s and early 1990s led to a dramatic decline and a more decentralized labor system.
- Recent wage dynamics comparison (as of 2023Q1):
  - Australia: WPI growing at around 3.6 percent.
  - New Zealand: LCI growing at 4.3 percent; Quarterly Employment Survey (QES) shows 7.6 percent in New Zealand.
- Interpretation for New Zealand vs Australia:
  - Australia’s greater prevalence of collective wage bargaining and an increase in average duration of wage agreements to around 3 years in 2022 may help explain why wages in Australia have responded less to labor market tightness relative to New Zealand since the pandemic.
  - Wage increases under enterprise agreements are slow moving; approved enterprise bargaining agreements in Australia in 2022 remained subdued.
  - Policy change: passage of the Fair Pay Agreements Act in December 2022 will increase the role of sectoral bargaining in New Zealand and is likely to contribute to less volatile wage growth going forward.

### Structural Drivers of Wages: The Minimum Wage
- Types and levels (as of April 1, 2023):
  - New Zealand has three types of minimum hourly wages: adult, starting-out, and training.
  - As of April 1, 2023:
    - Adult minimum wage (before tax): NZ$22.70/hour.
    - Starting-out and training: NZ$18.16/hour (sub-minimum rates typically 80 percent of adult minimum wage).
- Historical evolution and relative standing:
  - Minimum wage increases moderated between 2010 and 2017, averaging between 2 to 3 percent.
  - From 2019 through 2023, minimum wage growth averaged around 6.6 percent yearly.
  - Historically elevated relative to OECD average:
    - In 2000, real minimum wage was about 40 percent higher than the OECD average.
    - Currently the real minimum wage is 55 percent higher than the OECD average.
  - In real terms, the minimum wage grew by about 70 percent since 2000; OECD average increased by 54 percent.
- Distributional binding and coverage:
  - Large minimum wage increases over 20 years have closed the gap with the median wage.
  - Assuming a full-time earner at 40 hours/week and 52 weeks/year, the minimum wage was binding at the 10th percentile in 2001 and at the 30th percentile in 2022.
  - Implication: increases in the minimum wage are likely to affect workers at and around the 30th percentile and potentially those below it.
- Regression evidence on pass-through (industry-year regressions; Table 6):
  - Specification: ln(Wage i,t) = β ln(MinimumWage t) + σi + σi1*time + γt + εi,t.
  - Estimated pass-through:
    - A one percent increase in minimum wages would lift the hourly real wage by 0.5 percentage point on average.
    - Effects stronger for women compared to men.
    - At the median percentile, the impact of a one percent minimum wage increase: 0.74 percentage point for women vs 0.5 percentage point for men.
  - Table 6 reported coefficients (examples):
    - Minimum wage (log) coefficients: 0.522*** (Hourly Wage Total), 0.783*** (Hourly Wage Male), 0.411** (Hourly Wage Female), 0.629*** (Median Wage Total), 0.742*** (Median Wage Female), 0.508*** (Median Wage Male).
    - All regressions include year fixed effects, industry fixed effects, and industry x time trends. Number of industries: 18. Number of observations vary (examples: 252, 240).
- Mechanisms and literature:
  - Spillovers and distributional effects:
    - Minimum wage hikes can spike wage distribution density around the minimum wage and generate spillovers to workers earning above the minimum (higher-skilled substitution) and below it.
    - Findings in literature: increases can reduce inequality at lower tail; spillovers may extend to percentiles above the minimum wage (affecting both tails).
    - References in text include Lee (1999); Autor, Manning and Smith (2016); DiNardo, Fortin, and Lemieux (1996); Neumark and Wascher (2006); Gramlich (1976).

*Source: 1nzlea2023002, IMF.*

### 24.      Given the recent increases in the minimum wage, we explore its potential effects on

### 24.      Given the recent increases in the minimum wage, we explore its potential effects on 

### Methodology
- Data: annual industry-level data covering 17 industries for 2009 to 2022.
- Dependent variable: log hourly real average (mean) wage minus log hourly median (50th percentile) wage in industry i at time t.
- Key regressor: log minimum real wage for that year; bindingness defined as log(minimum wage) − log(median) (the effective minimum wage).
- Empirical specification: quadratic in the effective minimum wage to capture stronger impacts where the minimum is more binding; includes industry fixed effects (σi0), industry-specific time trends (σi1 ∗ time), time fixed effects (γtσ), and, as a robustness check, lags of the dependent variable to mitigate reverse causality (Annex tables 5-12).
- Followed literature: Autor, Manning, and Smith (2016); Lee (1999).

### Main empirical results (industry-level)
- Table 7 estimates (equation 5) show positive and statistically significant relationships between the effective minimum wage and log(average) − log(median) of real hourly wages:
  - Coefficient on Log(min. wage) − log(median): 0.587*** (column (1)), 0.564*** (column (2)), 0.822*** (column (3)), 0.364*** (column (4)), 0.408*** (column (5)), 0.332*** (column (6)).
  - Coefficient on [Log(min. wage) − log(median)]^2: 0.388** (col (1)), 0.233 (col (2)), 0.446*** (col (3)), 0.111 (col (4)), −0.277* (col (5)), 0.058 (col (6)).
  - Constant terms reported (examples): 0.357*** (col (1)), 0.305*** (col (2)), 0.483*** (col (3)).
  - R-squared values reported: 0.154, 0.221, 0.194, 0.161, 0.271, 0.185.
  - Number of industries in regressions: 18, 18, 18, 18, 17, 18 (respectively).
  - Number of observations: 252, 240, 252, 234, 221, 234 (respectively).
- Interpretation: the minimum wage in New Zealand is binding beyond the 50th percentile and has spillover effects into the upper part of the wage distribution.

### Industry, gender, and ethnicity patterns
- Industries closest to the minimum wage:
  - Services-focused industries and community services (retail sales employees, laborers, community workers) have median hourly wages closest to the minimum, with the difference between their median hourly wages and hourly minimum wage ranging between NZ$2-5.
  - Workers in retail trade and accommodation, agriculture, and transport also earn closer to the minimum wage.
- Gender disparities:
  - Female workers earn consistently less than men across almost all industries.
  - Average gap relative to the minimum wage across industries: women earn about NZ$ 8.6 per hour more than the minimum wage; men earn about NZ$ 11.9 per hour more than the minimum wage.
  - Expressed as a ratio: women earn what amounts to 88 cents in wages for women per each dollar that men earn.
  - Women in services, agriculture, transport, and manufacturing earn relatively closer to the minimum wage.
  - Disparities within female workers by ethnicity: Pacific peoples, Māori, and Asian women on average earn less compared to their European peers.
- Ethnicity patterns (Tables 8–10, industry-year level):
  - Table 8 (pooled by ethnicity) highlights larger estimated impacts of the effective minimum wage on Māori workers compared to European, Pacific, or Asian workers in some specifications:
    - Example coefficients (level specifications): Log(min. wage) − log(median): Māori 0.350**, European 0.333, Pacific Peoples 0.202, Asian 0.106.
    - Alternate specifications (first-differenced or FD): coefficients include 0.409*** (Māori), 0.272*** (European), 0.357*** (Pacific Peoples), 0.307*** (Asian).
  - Table 9 (women by ethnicity) and Table 10 (men by ethnicity) show heterogeneity:
    - For men, examples include Log(min. wage) − log(median) coefficients of 0.176** (Māori), 0.227*** (European), 0.488*** (Pacific Peoples), 0.672*** (Asian) in one specification.
    - Nonlinear terms (squared) sometimes enter negatively and significantly for certain ethnic-gender cells (examples reported in tables).
  - Overall interpretation: estimated spillovers vary across ethnic groups; some specifications show stronger pass-through for Māori or for European men depending on the regression. Results are not uniformly conclusive across all ethnicity–gender cells.
  - Note: some regressions report low R-squareds, indicating unexplained variation.

### Robustness
- Results remain robust when:
  - Adding lags of the dependent variable and lags of the hourly real average wage (annex tables 5-6).
  - Controlling for overall wages.
- Caution: OLS estimations may at times lead to an upward bias; pass-through of the minimum wage may be overestimated.
- Data limitations: lack of granular data at lower percentiles prevents direct empirical testing of effects on the left tail (below the median); inference about lower-tail effects relies on international literature (Lee, 1999; Autor, Manning and Smith, 2016).

### Policy-relevant findings and implications
- Spillovers and distributional effects:
  - The minimum wage lifts wages for low-wage earners but has spillover effects beyond the percentile where it binds (the minimum is expected to bind at the 30th percentile in 2022), potentially raising wages at the higher end of the distribution.
  - As a result, minimum wage increases can feed into overall wage pressures.
  - Minimum wage increases can reduce inequality for low-wage earners and between male and female workers in some cases, but do not provide concrete evidence of reducing inequities embedded across ethnicities.
- Targeting and equity considerations:
  - Because pay disparities across ethnicities persist (European workers generally earn close to or higher than the total average, while Māori or Pacific Peoples generally earn less than the total average), minimum wage policies alone may be insufficient to address ethnic wage gaps.
- Interaction with labor market institutions:
  - Separate section conclusions (Key Takeaways) emphasize that collective bargaining and longer contract lengths are associated with wages being less sensitive to unemployment; sectoral collective arrangements (e.g., Fair Pay Agreements) could weaken wage-employment sensitivity and produce less volatile but stickier wage growth.

*Source: IMF staff estimates, New Zealand chapter (annual industry-level analysis covering 2009–2022).*

### Annex Table 3. New Zealand: Growth in Average Hourly Earnings Using QES Data

### 1nzlea2023002 - Annex Table 3. New Zealand: Growth in Average Hourly Earnings Using QES Data

### Annex Table 3 — Estimates using QES data (quarterly)
- Notes: Data at quarterly frequency. Estimates of equation 1 augmented with lags of dependent variable. Unemployment gap calculated by estimating trend unemployment using an HP filter with a smoothing parameter of 14400. Robust standard errors with *, **, and *** indicating significance at the 10, 5 and 1 percent level respectively.
- Unemployment gap (columns 1–6):
  - Column (1): -0.632***
  - Column (2): -0.527***
  - Column (3): -0.528***
  - Column (4): -0.543***
  - Column (5): -0.542***
  - Column (6): -0.538***
  - Standard errors (cols 1–6): (0.0823), (0.0845), (0.0817), (0.0790), (0.0725), (0.0705)
- Expected inflation (columns shown):
  - Column (1): 0.379***
  - Column (2): 0.282***
  - Column (3): 0.150
  - Column (4): 0.318***
  - Column (5): 0.218**
  - Standard errors: (0.0612), (0.102), (0.104), (0.0937), (0.0931)
- Trend labor productivity growth (selected columns):
  - Column (1): 0.298
  - Column (2): 0.704***
  - Column (3): 0.210
  - Column (4): 0.676***
  - Standard errors: (0.216), (0.207), (0.213), (0.206)
- Lagged net migration:
  - Column (1): -0.357*
  - Column (2): -0.114
  - Standard errors: (0.208), (0.177)
- Lagged export prices:
  - Column (1): 1.094***
  - Column (2): 1.388***
  - Standard errors: (0.392), (0.392)
- Constant (columns 1–6):
  - 2.191***, 1.312***, 1.242***, 1.312***, 1.221***, 1.108***
  - Standard errors: (0.0564), (0.133), (0.165), (0.200), (0.161), (0.166)
- Observations (first block): 929, 184, 768, 476, 476
- R-squared (first block): 0.331, 0.563, 0.521, 0.638, 0.553, 0.681
- Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1

- Second reported block (alternative specification / lag structure — variables repeated):
  - Unemployment gap (columns 1–6): -0.948***, -0.871***, -0.860***, -0.953***, -0.896***, -0.928***
    - Standard errors: (0.169), (0.172), (0.170), (0.166), (0.138), (0.147)
  - Expected inflation (columns shown): 0.533***, 0.427**, 0.166, 0.517***, 0.295**
    - Standard errors: (0.134), (0.173), (0.170), (0.145), (0.133)
  - Trend labor productivity growth (selected): 0.450, 0.953**, 0.225, 0.883**
    - Standard errors: (0.383), (0.405), (0.356), (0.386)
  - Lagged net migration: -1.163***, -0.651**
    - Standard errors: (0.347), (0.312)
  - Lagged export prices: 2.723***, 2.557***
    - Standard errors: (0.712), (0.769)
  - Constant (cols 1–6): 3.133***, 1.947***, 1.752***, 2.207***, 1.703***, 1.830***
    - Standard errors: (0.0969), (0.286), (0.252), (0.344), (0.226), (0.299)
  - Observations (second block): 8080 8080 7280 72 (note: printed as 808080728072)
  - R-squared (second block): 0.316, 0.427, 0.437, 0.542, 0.512, 0.598
  - Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1

### Annex Table 4 — Using quarter-on-quarter wage growth (q-o-q)
- Notes: Data at quarterly frequency. Estimates of equation 1 augmented with lags of dependent variable. Unemployment gap calculated by estimating trend unemployment using an HP filter with a smoothing parameter of 14400. Robust standard errors with *, **, and *** indicating significance at the 10, 5 and 1 percent level respectively.
- Unemployment gap (columns 1–6):
  - Column (1): -0.195***
  - Column (2): -0.178***
  - Column (3): -0.178***
  - Column (4): -0.184***
  - Column (5): -0.180***
  - Column (6): -0.182***
  - Standard errors (cols 1–6): (0.0309), (0.0266), (0.0269), (0.0263), (0.0254), (0.0256)
- Expected inflation (columns 1–6): 0.117***, 0.116***, 0.0921***, 0.123***, 0.103***
  - Standard errors: (0.0269), (0.0334), (0.0326), (0.0345), (0.0342)
- Trend labor productivity growth (selected): 0.00618, 0.0759, -0.0108, 0.0701
  - Standard errors: (0.0670), (0.0754), (0.0715), (0.0770)
- Lagged net migration: -0.0904, -0.0478
  - Standard errors: (0.0674), (0.0666)
- Lagged export prices: 0.205*, 0.212
  - Standard errors: (0.119), (0.128)
- Constant (cols 1–6): 0.531***, 0.270***, 0.267***, 0.287***, 0.264***, 0.256***
  - Standard errors: (0.0168), (0.0565), (0.0595), (0.0674), (0.0598), (0.0636)
- Observations: 8080 8080 7280 72 (printed as 808080728072)
- R-squared: 0.375, 0.526, 0.526, 0.572, 0.538, 0.583
- Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1

### Key takeaways from Tables 3–4 (empirical patterns)
- The unemployment gap enters with a consistently negative and statistically significant coefficient across specifications in both year-on-year and q-o-q wage growth regressions (examples: -0.632***; -0.195***).
- Expected inflation generally enters with a positive and often statistically significant coefficient (examples: 0.379***; 0.117***).
- Trend labor productivity growth coefficients are positive in several specifications but not uniformly significant (examples: 0.704***; 0.00618).
- Lagged net migration and lagged export prices show varying impacts across specifications, with some coefficients large and significant in alternative specifications (examples: lagged net migration -1.163***; lagged export prices 2.723***).

---

### Annex Tables 5–12 — Robustness: Relationship between log(average) – log(median) and log(min. wage) - log(median)
- Notes common to these tables: Data at the industry-year level. Dependent variable in each regression is log(average) - log(median) of real hourly wages (variously for total, female, male, and by ethnicity). All regressions include industry and time fixed effects, industry-time trends, and either the lag of the dependent variable or the lag of real average hourly wages as a control (depending on table). Standard errors clustered at the industry level. *, **, and *** indicate significance at the 10, 5 and 1 percent level respectively.

- Annex Table 5 (lag dependent variable as control; total/female/male; levels and FD specifications)
  - Log(min. wage) - log(median):
    - Hourly Wage (total) (col 1): 0.702***
    - Hourly Wage Female (col 2): 0.736***
    - Hourly Wage Male (col 3): 0.877***
    - Hourly Wage (FD) (col 4): 0.145**
    - Hourly Wage Female (FD) (col 5): 0.001
    - Hourly Wage Male (FD) (col 6): -0.098
    - Standard errors example: (0.217), (0.248), (0.204), (0.066), (0.087), (0.065)
  - Log(min. wage) - log(median), sq:
    - Examples: 0.442**, 0.359, 0.466***, -0.014, -0.410***, -0.914***
  - Lag dependent variable:
    - Examples: 0.047, 0.139*, 0.077, -0.950***, -0.916***, 0.085
  - R-squared: 0.170, 0.236, 0.228, 0.562, 0.583, 0.562
  - Number of industries: 18, 17, 18, 17, 18, 17
  - Number of observations: 234, 221, 234, 234, 221, 234

- Annex Table 6 (lag average wage (log) as control)
  - Log(min. wage) - log(median) (columns 1–6 examples): 0.675***, 0.631**, 0.859***, 0.941***, 0.780***, 0.894***
  - Log(min. wage) - log(median), sq: 0.412**, 0.275, 0.452***, 0.623***, 0.424***, 0.481***
  - Lag average wage (log): 0.062, 0.027, 0.015, -0.910***, -0.886***, -0.935***
  - R-squared: 0.173, 0.221, 0.223, 0.572, 0.605, 0.598
  - Number of industries and observations mirror Table 5.

- Annex Tables 7–8 (by ethnicity: Māori, European, Pacific Peoples, Asian)
  - Table 7 (lag of real average hourly wage (log) reported; levels and FD):
    - Log(min. wage) - log(median) (cols 1–4 level specs): 0.431***, 0.486**, 0.149, 0.177
    - (cols 5–8 FD specs): 0.144***, 0.124*, 0.026, 0.025
    - Log(min. wage) - log(median), sq (examples): 0.074, 0.275, -0.231, -0.117; and -0.195***, -0.012, -0.374***, -0.258***
    - Lag of real average hourly wage (log) examples: 0.117**, 0.062, -0.093, 0.105; -0.790***, -0.935***, -1.085***, -0.886***
    - R-squared range: 0.129 to 0.727
    - Number of industries examples: 18, 17, 18, 17
    - Number of observations examples: 216, 234, 198, 213
  - Table 8 (lag of real hourly average wage as control; alternative reported coefficients):
    - Level columns (1–4) Log(min. wage) - log(median): 0.440***, 0.464*, 0.181, 0.171
    - FD columns (5–8) Log(min. wage) - log(median): 0.862***, 0.917***, 0.884***, 0.773***
    - Squares and lags reported with positive FD coefficients for many ethnic groups (examples: 0.482***, 0.606***, 0.606***, 0.430***)
    - R-squared range: 0.134 to 0.686

- Annex Tables 9–10 (women, by ethnicity)
  - Table 9 (lag dependent variable as control; female workers):
    - Level columns (1–4) Log(min. wage) - log(median): 0.074, 0.137, 0.081, 0.660***
    - FD columns (5–8) Log(min. wage) - log(median): 0.041, -0.152**, 0.094, 0.271***
    - Squares and lag dependent variable reported; R-squared range: 0.214 to 0.691
    - Number of observations examples: 189, 221, 144, 183
  - Table 10 (lag real average hourly wage as control; female workers):
    - Level columns (1–4) Log(min. wage) - log(median): 0.022, 0.007, 0.089, 0.763***
    - FD columns (5–8) Log(min. wage) - log(median): 0.690***, 0.666***, 0.874***, 1.043***
    - R-squared range: 0.210 to 0.696

- Annex Tables 11–12 (men, by ethnicity)
  - Table 11 (lag dependent variable as control; male workers):
    - Level columns (1–4) Log(min. wage) - log(median): -0.220, 0.57***, -0.070, 0.28*
    - FD columns (5–8) Log(min. wage) - log(median): 0.050, 0.11*, -0.00, -0.01
    - Log(min. wage) - log(median), sq examples: -0.52***, 0.18, -0.47**, -0.12; and -0.28***, -0.12**, -0.40***, -0.34***
    - R-squared range: 0.218 to 0.709
    - Number of observations examples: 204, 234, 175, 206
  - Table 12 (lag of real average hourly wage as control; male workers):
    - Level columns (1–4) Log(min. wage) - log(median): -0.13, 0.58***, -0.08, 0.280.92*** (table prints mixed formatting)
    - FD columns (5–8) Log(min. wage) - log(median): 1.00***, 0.87***, 0.85*** (printed across)
    - Log(min. wage) - log(median), sq examples show both negative and positive significant squares (examples: -0.44***, 0.18, -0.48**, -0.13; and 0.47***, 0.44***, 0.52***, 0.30***)
    - Lag of real average hourly wage (log) FD columns: -0.92***, -0.99***, -0.96***, -0.84***
    - R-squared range: 0.216 to 0.646
    - Number of observations examples: 204, 234, 175, 206

### Summary of robustness patterns across Annex Tables 5–12
- The relationship between log(average) - log(median) and log(min. wage) - log(median) is consistently estimated with positive level coefficients in many level specifications (examples: 0.702***, 0.675***, 0.440***).
- Quadratic terms (log(min. wage) - log(median), sq) show mixed signs and statistical significance across specifications, indicating nonlinearity in some groups (examples: 0.442**, -0.914***, 0.482***).
- Results differ by gender and ethnicity:
  - Female and male specifications produce different coefficient magnitudes and significance patterns (examples: female FD: 0.690***; male FD: 1.00*** in Table 10/12 reporting).
  - Ethnic subgroup estimates (Māori, European, Pacific Peoples, Asian) exhibit heterogeneous coefficients and R-squared values (examples: Māori level: 0.431***; European FD: 0.917***).
- Lag controls (lag dependent variable or lag real average hourly wage (log)) materially affect coefficient magnitudes and signs in FD specifications (examples: lag average wage (log) -0.910*** in Table 6; lag of real average hourly wage (log) -0.935*** in Table 7 FD columns).

*Source: IMF staff estimates (Annex Tables 3–12, New Zealand).*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2023/english/1nzlea2023002.pdf_