## EXECUTIVE SUMMARY (cr17118)

## Source details

**Canonical URL:** [EXECUTIVE SUMMARY (cr17118)](https://www.imf.org/-/media/files/publications/cr/2017/cr17118.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2017/cr17118.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2017/cr17118.pdf.json)

---

### Institutional framework
- The Reserve Bank of New Zealand (RBNZ) is the single prudential regulator with responsibilities and powers for the supervision of financial institutions and macroprudential policies.
- A Memorandum of Understanding (MoU) between the RBNZ and the Minister of Finance:
  - Legitimized the macroprudential use of existing prudential instruments.
  - Provided additional clarity on objectives, governance, and instruments for macroprudential policy.
  - Attributes to the RBNZ the responsibility to monitor emerging risks and intervene when needed.
  - Focuses on increasing the resilience of the domestic financial system and countering instability arising from credit, asset prices or liquidity shocks (time-dimension is primary; dampening extremes in the financial cycle is a secondary role).
- Strengths:
  - Clear mandate for financial stability and operational clarification via the MoU.
  - Independent decision making, transparent communication, and external accountability form a strong base for macroprudential policy.
- Weaknesses / areas to strengthen:
  - The MoU created a relatively narrow framework limiting actions to banks and including four instruments in the macroprudential toolkit, effectively constraining powers to take broader macroprudential actions.
  - Lack of clear and transparent procedures for adjusting the MoU creates the risk of a lengthy process that may undermine timely prudential action.
  - Recommendation: Make the process for adjusting the MoU more transparent; publicly disclose RBNZ advice and the Minister’s opinions when adjustments are considered.
  - Maintain an effective accountability framework for the RBNZ that does not jeopardize the integrity and independence of its macroprudential decision making process.

### Financial cycle risks and tools (housing sector focus)
- Use of LVR restrictions:
  - In mid-2013, rules required banks to reduce the volume of high loan-to-value ratio (LVR) lending to below 10 percent of new commitments (from around 30 percent prior to the new rules).
  - Real estate prices and credit expansion continued to grow strongly, particularly in the Auckland area, prompting a tightening of the LVR restriction to Auckland investors in late 2015.
  - October 2016 restrictions:
    - Eliminate the difference between Auckland and the rest of the country.
    - Impose tighter limits on investor loans and owner-occupier lending.
    - On investor loans, only 5 percent of a bank’s total lending to residential property investors can be to borrowers with a deposit less than 40 percent of the house value.
    - On owner-occupier lending, only 10 percent of a bank’s total lending to owner-occupiers can be to borrowers with a deposit less than 20 percent of the house value.
- Remaining imbalances and systemic risks:
  - Improvements in loan-to-value indicators occurred after the LVR restrictions, but imbalances have not been fully addressed.
  - Credit and price growth remained strong for years following the initial measures.
  - Long term declining rental yields and a persistently rising price-to-income ratio suggest the possibility of overvaluation.
  - Household debt-to-income levels have reached record highs.
  - Housing loans represent more than half of banking sector assets.
  - Authorities are encouraged to continue close monitoring and to promote further actions to increase system resilience if October 2016 LVR adjustments do not substantially change the balance of risks.
- Recommendation on DTI limits:
  - Limits on debt-to-income (DTI) should become part of the macroprudential toolkit; the RBNZ is currently discussing with the Minister of Finance the introduction of limits to total DTI.
  - Caps on DTI or similar measures (e.g., debt servicing to total income (DSI)) can complement LVR restrictions by:
    - Addressing risks from record high household indebtedness.
    - Increasing resilience by reducing probability of defaults (PD) and eventual losses.
    - Contributing to stabilization by reducing demand for loans.
  - Rationale and implementation guidance:
    - Risks build up relatively quickly; legal and operational basis should be in place before the need to use prudential instruments fully materializes.
    - If LVR measures do not substantially reduce risks, DTI limits should be considered.
    - Price increases reduce the effectiveness of LVR restrictions via refinancing; DTI caps can act as automatic stabilizers because they become more binding when house prices grow faster.
    - Reliance on multiple tools reduces distortions compared to the use of a single conservatively calibrated tool.
    - Authorities should take a gradual approach when implementing DTI restrictions to mitigate uncertainties and reduce burdens on lenders and borrowers.
    - Consider measures of similar nature (DSI) if appropriate.

### Structural risks and tools (financial sector concentration)
- Concentration and vulnerabilities:
  - The four biggest banks in New Zealand hold 87 percent of the banking system’s assets.
  - These banks have very similar business models: the majority of their assets are housing loans and they are partly reliant on foreign funding.
  - Direct exposures among them are relatively limited, but spillover potential is elevated through reliance on overseas funding and the size of banks (deleverage or fire sales by one bank could depress asset values/system-wide activity).
- Capital requirements and SIFI treatment:
  - Distress of systemically important financial institutions (SIFIs) causes significant disruption to the wider financial system and economic activity.
  - The largest banks in New Zealand are systemically important and should be required to hold capital commensurate with the magnitude of the externalities of their eventual failure.
  - Given RBNZ’s prudential approach emphasizing market discipline and the practice that smaller banks maintain larger capital ratios, differentiating capital buffer sizes by bank does not seem necessary.
  - The current review of capital requirements by the RBNZ should increase capital thresholds for the SIFIs or the whole system to better reflect risks posed by concentration.
- Liquidity considerations:
  - Review liquidity regulatory requirements aiming to reduce banks’ reliance on the provision of liquidity by the RBNZ.

### Main recommendations (selected)
- Strengthen institutional arrangements for macroprudential policy by increasing communication efforts and transparency when adjustments to the framework are being considered.
  - Timing: C (continuous); Responsible: RBNZ/Treasury
- Maintain an effective accountability of the RBNZ that does not jeopardize the integrity and independence of its macroprudential decision making process.
  - Timing: C (continuous); Responsible: RBNZ/Treasury
- Enhance the operational basis for macroprudential policy by including DTI limits in the macroprudential toolkit.
  - Timing: I (immediate) = within one year; Responsible: RBNZ/Treasury
- Implement DTI measures if the changes to the LVR do not substantially reduce the risks in the housing sector.
  - Timing: I (immediate); Responsible: RBNZ
- Increase capital requirements to reflect the risks posed by the concentration of the financial sector in four main banks.
  - Timing: I (immediate); Responsible: RBNZ
- Review liquidity regulatory requirements aiming to reduce banks reliance on the provision of liquidity by the RBNZ.
  - Timing: I (immediate); Responsible: RBNZ

*Prepared by IMF staff as presented in the source content.*

### Governance and dialogue between RBNZ, Minister of Finance, and Treasury
- Role and consultation:
  - The RBNZ is, in practice, the authority responsible for macroprudential policy.
  - The RBNZ should keep the Minister of Finance and the Treasury regularly informed on its thinking on significant policy developments and emerging risks.
  - The RBNZ should consult with the Minister and the Treasury when macroprudential intervention is under active consideration and inform them prior to making any decision on triggering a macroprudential policy instrument.
  - The Treasury provides advice on the impact of decisions taken by the RBNZ on broader Government objectives.
  - Instruments in the toolkit are expected to be subject to enhanced consultation with the Treasury due to potential economy-wide impacts.
  - The Minister of Finance and the Treasury are consulted on all decisions to impose, amend or revoke prudential requirements.
- Decision-making bodies and internal governance:
  - The Governor of the RBNZ is legally the sole decision maker for prudential policy.
  - Prudential policies are discussed and approved by the Governing Committee (Governor, two Deputy Governors, Assistant Governor).
  - For macroprudential policy, a key advisory role is undertaken by the Macro-Financial Committee (MFC) which meets monthly.
- External oversight and transparency:
  - The Board of Directors reviews how well the RBNZ meets legal obligations and publishes conclusions in an annual report.
  - All prudential measures are subject to public consultation and the publication of an impact assessment (consultations have been relatively short; room to improve impact assessment).
  - The RBNZ publishes the Financial Stability Report (FSR) twice a year; the FSR is reviewed by Parliament and the Board and must be presented to Parliament by the Minister of Finance.

### Ability to act: powers and toolkit
- RBNZ powers and toolkit under the MoU:
  - The MoU included four instruments in the macroprudential toolkit:
    - (i) a countercyclical capital buffer;
    - (ii) adjustments in sectoral capital requirements;
    - (iii) restrictions to the share of high loan-to-value ratio (LVR) residential mortgages; and
    - (iv) adjustments to the minimum core funding requirement.
  - As a prudential regulator, the RBNZ can also implement tools not expected to fluctuate over the cycle (leverage ratio, limits on interbank exposures, liquidity requirements, D-SIFIs buffers).
  - Footnote: Leverage ratio and D-SIFIs buffers haven’t been implemented in New Zealand for the time being.
  - The MoU will be reviewed after 5 years, in 2018, and provides room to add instruments with agreement of the Minister.

### Cooperation arrangements and checks and balances
- Integrated supervision and coordination:
  - RBNZ acts as supervisor, resolution authority, macroprudential authority and central bank, facilitating timely information sharing.
  - Governing Committee discusses major monetary and financial decisions.
  - Internal mitigation of groupthink: MFC debates and staff analytical papers; Governing Committee as additional forum.
  - External arrangements: Treasury discussions prior to implementation, mandatory public consultation, Board oversight.
- Council of Financial Regulators (CoFR):
  - Formed in 2013. Members: Financial Markets Authority (FMA), RBNZ, Treasury, Ministry of Business, Innovation and Employment (MBIE).
  - CoFR meets quarterly; chaired on a rotating basis between the Governor of the RBNZ and the Chief Executive of the FMA.
  - A CoFR subcommittee (Banking Forum) discusses regulatory matters relating to registered banks; includes MoJ and IRD among others.
- Bilateral MoUs exist between the FMA and the RBNZ and between the Treasury and RBNZ for information sharing.

### Assessment of systemic risks and data
- Macro-Financial Department:
  - Dedicated department with two teams and 10 staff leads systemic risk analysis and macroprudential policy discussions.
  - Responsible for the Financial Stability Report (FSR), published twice a year.
  - The 2013 MoU requires the FSR to detail reasons for, and impact of, any macroprudential intervention undertaken by the RBNZ.
- Data limitations:
  - Reports are not sufficiently standardized across financial entities.
  - Ongoing work to increase standardization and improve data quality of reports of insurance firms and debt-to-income for mortgage customers.
  - Information is not always granular enough for detailed asset/liability analysis or solvency/liquidity stress tests.
  - Data on market risks is limited and liabilities data does not cover counterparty types or existence of collateral.
  - Ongoing efforts to improve commercial real estate information.

### Credit expansion and housing sector imbalances — key indicators and findings
- Credit and household indicators:
  - After a period of rapid expansion, credit growth decelerated after the GFC; credit growth has accelerated recently but credit gap indicators suggest overall credit is still aligned with its long term trend.
  - Recent developments:
    - Credit has recently outpaced deposit growth.
    - Property-related DTI increased to reach 116 percent.
    - Aggregate debt servicing ratio (DSR) has been relatively stable due to a substantial reduction in the cost of loans.
    - Household debt-to-income ratio (DTI) has been on a long term increasing trend.
- Housing market imbalances and risks:
  - Mortgage lending increased sharply following a slowdown after the GFC despite restrictive macroprudential measures.
  - Housing prices have grown strongly, particularly in the Auckland area.
  - Long term declining rental yields suggest the possibility of overvaluation.
  - A significant proportion of borrowers are investors and the vast majority of mortgages are repriced within 3 years.
  - Repricing could increase DSR and lead to defaults when interest rates normalize.

### Housing market fundamentals and overvaluation (selected findings)
- Supply and demand:
  - Housing supply in New Zealand does not appear to have kept up with demand.
  - Strong immigration has resulted in population growth rapid and well above OECD average.
  - Real housing investment as percent of GDP has remained below the OECD average.
  - Supply problem particularly acute in Auckland; supply measures (e.g., the Unitary Plan) are expected to moderate price dynamics only gradually.
- Overvaluation estimates and recent price changes:
  - International comparisons and fundamentals-based analysis suggest overvaluation in the 20 to 30 percent range.
  - A fundamentals-based model (estimating real house price changes as a function of real disposable income, working-age population, equity prices and interest rates, estimation period starting in 2000) suggests house price overvaluation has recently increased further to more than 20 percent.
  - Regional price increases (year to September): New Zealand as a whole picked up 14.3 percent; Auckland experienced 15.0 percent; Wellington experienced 21.2 percent.
- Investor behavior and leverage:
  - Rising housing prices allowed investors to use equity gains to increase leverage and buy new properties.
  - Tax benefits that largely exempt capital gains and limited alternatives have made housing a popular investment.
  - Proportion of investors in the market has increased; household DTI reached record highs.
  - Housing loans represent more than half of banks’ assets.

### Use of prudential tools by the RBNZ — capital and LVR adjustments
- LVR chronology and calibrations:
  - October-2013: Restrict new residential mortgage lending at LVRs over 80 percent to no more than 10 percent of new commitments.
  - November-2015: Three groups of restrictions:
    - For investor loans in Auckland, only 5 percent of total new loans could be granted with a LVR above 70 percent.
    - For other loans in Auckland, only 10 percent of total new loans with a LVR above 80 percent could be granted.
    - For loans outside Auckland, only 15 percent of total new loans with a LVR above 80 percent could be granted.
  - October-2016: Geographical differences removed and investor restrictions tightened (see above).
- Capital adjustments for housing loans (2013–2015):
  - Average risk weights for banks following the standardized approach are 38 percent.
  - Mortgages in IRB banks are risk weighted in the 25–31 percent range.
  - Changes in correlation increased the requirements by approximately 12 percent considering the mortgage portfolio and 3 percent considering overall capital.
  - Despite more conservative calibration, risk weights remained substantially below the ones in force until 2008 (50 percent, following Basel I rules).

### Impact and effectiveness of measures (LVR evidence)
- LVR effects:
  - LVR restrictions had a substantial impact on strengthening new loan origination standards and gradually reducing the share of high mortgages in the credit portfolio.
- Counterfactual VAR analysis (summary):
  - For 2013 LVR limits: VAR estimated on 1993Q1–2013Q1; forecasts for 2013Q3–2016Q1.
  - For 2015 LVR limits: VAR estimated on 1993Q1–2015Q1; forecasts for 2015Q4–2016Q2.
  - Findings:
    - Forecasted household credit growth is above actual credit growth following 2013 and 2015 LVR limits, but the difference is not statistically significant at the 95 percent confidence level.
    - No evidence that 2013 LVR limits had a prolonged impact on house prices.
    - 2015 LVR limits do not seem to have affected rates of credit growth nor growth of average house prices.
    - 2015 LVR limits likely contributed to a reduction of the growth of house prices in the Auckland area, at least in the short term.
    - Both 2013 and 2015 LVR limits appear to have reduced the number of house sales in New Zealand.
  - Mortgage lending from nonbank institutions has increased but remains small at approximately 1 percent of outstanding mortgages.
  - Behavior of other consumer credit does not suggest substantial substitution away from bank mortgage lending.
- Notes on uncertainty:
  - Confidence intervals are very wide due to high volatility of credit and price growth in recent years.
  - For 2015 LVR limits the short forecast period increases forecast uncertainty further.

### Systemic importance, capital framework, and sectoral concentration
- Identification and approach:
  - The RBNZ has not formally identified a set of systemically important institutions or imposed additional capital requirements based on systemic importance.
  - Decision reflects desire to avoid moral hazard associated with perception of institutions being ‘too big to fail’.
  - RBNZ considers applying higher requirements to all banks if necessary given concentration and the tendency of smaller banks to hold larger capital ratios.
  - RBNZ is conducting a wider review of the capital framework in light of Basel Committee proposals.
- Sectoral concentration and agriculture/dairy risks:
  - Dairy exposures represent around 10 percent of total bank lending.
  - Total agriculture exposures are around 15 percent of total bank lending.
  - All dairy farms exposed to same commodity price → high correlation in cash flows and farm land values; market prone to lack of liquidity during downturns.
  - Current NPLs in the sector remain at around 1.5 percent.
  - Stress tests suggest NPLs could increase significantly but in isolation could be absorbed from underlying earnings across other bank assets.
- Mortgage borrower DTI distribution (New Zealand share of mortgages by borrower type and DTI, in percent):
  - First home buyers:
    - DTI ≤ 3: 18.9
    - DTI > 3: 81.0
    - DTI > 4: 59.1
    - DTI >5: 34.6
    - DTI > 6: 16.1
    - DTI > 7: 6.5
    - DTI unknown: 0.1
  - Investors:
    - DTI ≤ 3: 14.2
    - DTI > 3: 84.1
    - DTI > 4: 74.6
    - DTI >5: 59.6
    - DTI > 6: 45.0
    - DTI > 7: 32.4
    - DTI unknown: 1.7

### Funding structure, liquidity risk, and prudential liquidity policy
- Structural funding reliance:
  - New Zealand has chronically low savings → economy dependent on borrowing from abroad; banking system has structural reliance on offshore wholesale funding.
  - Offshore funding remains material and likely to increase due to credit outpacing deposit growth → structural vulnerability to deterioration in global funding conditions.
- RBNZ liquidity policy and requirement metrics:
  - RBNZ prudential liquidity policy (BS13) introduced in 2010 to address rollover risk by requiring more long-term debt-funding.
  - Liquidity buffer composition: significant part comprised of residential mortgage-backed securities usable in repo operations with the RBNZ.
  - RBNZ liquidity quantitative requirements:
    - One-week mismatch ratio: cannot be less than zero percent at the end of each business day.
    - One-month mismatch ratio: cannot be less than zero percent at the end of each business day.
    - One-year CFR: cannot be less than 75 percent at the end of each business day.
  - Comparability with international standards:
    - One-month mismatch ratio conceptually comparable to LCR but definitions and calibrations differ; mismatch ratio includes most securities rated BBB- or better.
    - Applied haircuts (including on New Zealand government securities: one or three percent depending on maturity) tend to be lower than Basel Committee haircuts.
    - Treatment of contractual inflows and committed facilities is more lenient than LCR (no 75 percent cap on inflows; allows banks to consider 75 percent undrawn committed lines).
    - Run-off factors are less granular than the LCR.
- Core Funding Ratio (CFR) history and rationale:
  - CFR design analogous to NSFR: includes funding > 1 year, 50 percent of tradable debt with residual maturities between 6 months and 1 year, tier 1 capital, plus a percentage of non-marketable funding < 1 year depending on liability size.
  - RBNZ view: a 75 percent CFR results in liquidity requirements broadly equivalent to a 100 percent NSFR.
  - CFR timeline:
    - Initially set in 2010 at 65 percent.
    - Increased to 70 percent in mid-2011.
    - Increased to 75 percent in early 2013.
  - The CFR requirement is permanent; the MoU allows temporary adjustments in response to the financial cycle (tool not yet used for this purpose).
  - In response to increases in CFR, banks increased proportion of retail deposits and long-term funding.
- Stress tests concluded the liquidity profile of New Zealand banks is resilient.
- Regulatory observations:
  - Mismatch ratio is less conservative than LCR due to broader liquid asset definition.
  - Regulatory approach lacks sufficient granularity to differentiate asset/liability structures for liquidity purposes.
  - Broad liquid asset range is necessary given shortage of high-quality liquid assets in New Zealand but results in significant holdings of assets that will be liquid in stress only with central bank support.
  - Further convergence with international standards could reduce distortions, recognizing limited market volume of liquid assets constrains full alignment.

### Recommendations (capital and liquidity)
- Capital:
  - Capital requirements should better reflect systemic risk posed by concentration of the financial system.
  - The largest banks are systemically important and should be required to hold capital commensurate with the magnitude of externalities of their eventual failure.
  - Given market discipline and smaller banks’ higher capital ratios, it does not seem necessary to differentiate the size of the capital buffer by bank.
  - The current RBNZ review of capital requirements is recommended to increase capital thresholds for the SIFIs or the whole system to better reflect concentration risks.
- Liquidity:
  - Liquidity requirements should provide stronger incentives for banks to reduce excessive reliance on provision of liquidity by the Reserve Bank.
  - While New Zealand standards improved liquidity profiles, they do not provide the same incentives as international standards for banks to hold genuinely liquid assets.
  - Regulatory approach should be more granular to differentiate asset and liability structures for liquidity purposes.
  - Further convergence with international standards could potentially reduce distortions, recognizing constraints from limited market volume of liquid assets.

### Appendix I — Impact of LVR restrictions (counterfactual analysis) — key points
- Methodology:
  - VAR estimated on quarterly data starting 1993 with 9 variables (net migration non-NZ citizens per 1000 NZ residents; net migration of NZ citizens per 1000 NZ residents; output gap calculated using one-sided HP filter; floating rate on new mortgage for new customers; number of house sales per 1000 NZ residents; number of new building consents per 1000 NZ residents; annual real growth rate of average NZ house prices; annual real growth rate of house prices in the Auckland region; annual real growth rate of household credit).
  - Conditional forecasts of housing-specific variables (variables 5–9) use actual paths of macroeconomic variables (variables 1–4) after LVR introduction.
  - For 2013 LVR limits: VAR estimated on 1993Q1–2013Q1; forecasts for 2013Q3–2016Q1.
  - For 2015 LVR limits: VAR estimated on 1993Q1–2015Q1; forecasts for 2015Q4–2016Q2.
- Findings:
  - Forecasted household credit growth is above actual credit growth following introduction of 2013 and 2015 LVR limits, but the difference is not statistically significant at the 95 percent confidence level.
  - Impact on New Zealand average house price growth is not statistically significant for either 2013 or 2015 LVR limits.
  - The 2015 LVR limits seem to have reduced price growth in the Auckland area (at least in the short-term).
  - Both 2013 and 2015 LVR limits appear to have reduced the number of house sales in New Zealand.
- Notes on uncertainty:
  - Confidence intervals are very wide due to high volatility of credit and price growth in recent years.
  - For 2015 LVR limits the short forecast period increases forecast uncertainty further.

*International Monetary Fund — cr17118 (excerpts provided in content unit).*

### EXECUTIVE SUMMARY ___________________________________________________________________________ 5

### EXECUTIVE SUMMARY

### Institutional framework
- The Reserve Bank of New Zealand (RBNZ) is the single prudential regulator with responsibilities and powers for the supervision of financial institutions and macroprudential policies.
- A Memorandum of Understanding (MoU) between the RBNZ and the Minister of Finance:
  - Legitimized the macroprudential use of existing prudential instruments.
  - Provided additional clarity on objectives, governance, and instruments for macroprudential policy.
  - Attributes to the RBNZ the responsibility to monitor emerging risks and intervene when needed.
  - Focuses on increasing the resilience of the domestic financial system and countering instability arising from credit, asset prices or liquidity shocks (time-dimension is primary; dampening extremes in the financial cycle is a secondary role).
- Strengths:
  - Clear mandate for financial stability and operational clarification via the MoU.
  - Independent decision making, transparent communication, and external accountability form a strong base for macroprudential policy.
- Weaknesses / areas to strengthen:
  - The MoU created a relatively narrow framework limiting actions to banks and including four instruments in the macroprudential toolkit, effectively constraining powers to take broader macroprudential actions.
  - Lack of clear and transparent procedures for adjusting the MoU creates the risk of a lengthy process that may undermine timely prudential action.
  - Recommendation: Make the process for adjusting the MoU more transparent; publicly disclose RBNZ advice and the Minister’s opinions when adjustments are considered.
  - Maintain an effective accountability framework for the RBNZ that does not jeopardize the integrity and independence of its macroprudential decision making process.

### Financial cycle risks and tools (housing sector focus)
- Use of LVR restrictions:
  - In mid-2013, rules required banks to reduce the volume of high loan-to-value ratio (LVR) lending to below 10 percent of new commitments (from around 30 percent prior to the new rules).
  - Real estate prices and credit expansion continued to grow strongly, particularly in the Auckland area, prompting a tightening of the LVR restriction to Auckland investors in late 2015.
  - Despite benefits, growth in house prices and credit remained elevated, leading the RBNZ to extend restrictions nationwide starting in October 2016. The October 2016 restrictions:
    - Eliminate the difference between Auckland and the rest of the country.
    - Impose tighter limits on investor loans and owner-occupier lending.
- Remaining imbalances and systemic risks:
  - Improvements in loan-to-value indicators occurred after the LVR restrictions, but imbalances have not been fully addressed.
  - Credit and price growth remained strong for years following the initial measures.
  - Long term declining rental yields and a persistently rising price-to-income ratio suggest the possibility of overvaluation.
  - Household debt-to-income levels have reached record highs.
  - Housing loans represent more than half of banking sector assets.
  - Authorities are encouraged to continue close monitoring and to promote further actions to increase system resilience if October 2016 LVR adjustments do not substantially change the balance of risks.
- Recommendation on DTI limits:
  - Limits on debt-to-income (DTI) should become part of the macroprudential toolkit; the RBNZ is currently discussing with the Minister of Finance the introduction of limits to total DTI.
  - Caps on DTI or similar measures (e.g., debt servicing to total income (DSI)) can complement LVR restrictions by:
    - Addressing risks from record high household indebtedness.
    - Increasing resilience by reducing probability of defaults (PD) and eventual losses.
    - Contributing to stabilization by reducing demand for loans.
  - Rationale for introducing DTI limits:
    - Risks build up relatively quickly; legal and operational basis should be in place before the need to use prudential instruments fully materializes.
    - If LVR measures do not substantially reduce risks, DTI limits should be considered.
    - Price increases reduce the effectiveness of LVR restrictions via refinancing, potentially requiring successive tightening; DTI caps can act as automatic stabilizers because they become more binding when house prices grow faster.
    - Reliance on multiple tools reduces distortions compared to the use of a single conservatively calibrated tool.
  - Implementation guidance:
    - Authorities should take a gradual approach when implementing DTI restrictions to mitigate uncertainties over transmission strength and reduce burdens on lenders and borrowers.
    - Consider measures of similar nature (DSI) if appropriate.

### Structural risks and tools (financial sector concentration)
- Concentration and vulnerabilities:
  - The four biggest banks in New Zealand hold 87 percent of the banking system’s assets.
  - These banks have very similar business models: the majority of their assets are housing loans and they are partly reliant on foreign funding.
  - Direct exposures among them are relatively limited, but spillover potential is elevated through:
    - Reliance on overseas funding, which could lead to system-wide tightening if one bank faces problems.
    - The size of banks: deleveraging or fire sales by one bank could depress asset values and economic activity, contaminating the whole system.
- Capital requirements and SIFI treatment:
  - Distress of systemically important financial institutions (SIFIs) causes significant disruption to the wider financial system and economic activity.
  - The largest banks in New Zealand are systemically important and should be required to hold capital commensurate with the magnitude of the externalities of their eventual failure.
  - Given RBNZ’s prudential approach emphasizing market discipline and the practice that smaller banks maintain larger capital ratios, differentiating capital buffer sizes by bank does not seem necessary.
  - The current review of capital requirements by the RBNZ should increase capital thresholds for the SIFIs or the whole system to better reflect risks posed by concentration.
- Liquidity considerations:
  - Review liquidity regulatory requirements aiming to reduce banks’ reliance on the provision of liquidity by the RBNZ.

### Main recommendations (from Table 1)
- Strengthen institutional arrangements for macroprudential policy by increasing communication efforts and transparency when adjustments to the framework are being considered.
  - Timing: C (continuous); Responsible: RBNZ/Treasury
- Maintain an effective accountability of the RBNZ that does not jeopardize the integrity and independence of its macroprudential decision making process.
  - Timing: C (continuous); Responsible: RBNZ/Treasury
- Enhance the operational basis for macroprudential policy by including DTI limits in the macroprudential toolkit.
  - Timing: I (immediate) = within one year; Responsible: RBNZ/Treasury
- Implement DTI measures if the changes to the LVR do not substantially reduce the risks in the housing sector.
  - Timing: I (immediate); Responsible: RBNZ
- Increase capital requirements to reflect the risks posed by the concentration of the financial sector in four main banks.
  - Timing: I (immediate); Responsible: RBNZ
- Review liquidity regulatory requirements aiming to reduce banks reliance on the provision of liquidity by the RBNZ.
  - Timing: I (immediate); Responsible: RBNZ

*Source: EXECUTIVE SUMMARY (cr17118).*

### 8.      The New Zealand framework provides channels to promote a dialogue between the

### 8.      The New Zealand framework provides channels to promote a dialogue between the 

### Governance and dialogue between RBNZ, Minister of Finance, and Treasury
- The RBNZ is, in practice, the authority responsible for macroprudential policy.
- The RBNZ should keep the Minister of Finance and the Treasury regularly informed on its thinking on significant policy developments and emerging risks to the financial system.
- The RBNZ should consult with the Minister and the Treasury when macroprudential intervention is under active consideration and inform them prior to making any decision on triggering a macroprudential policy instrument.
- The Treasury provides advice on the impact of decisions taken by the RBNZ on broader Government objectives.
- Instruments included in the toolkit are expected to be subject to enhanced consultation and discussion procedures with the Treasury due to their potential impact on the overall economy and synergies with other government policies (i.e., tax policies).
- The Minister of Finance and the Treasury are consulted on all decisions to impose, amend or revoke prudential requirements, regardless of the status of the instrument in the macroprudential toolkit.
- These procedures are designed to allow opportunities for the government to express their views while maintaining a necessary and appropriate level of independence of the macroprudential decision making process.

### Decision-making bodies and internal governance
- The Governor of the RBNZ is legally the sole decision maker for prudential policy.
- All prudential policies are discussed and approved by the Governing Committee (which also undertakes monetary policy decisions), comprised of the Governor, the two Deputy Governors and the Assistant Governor.
- For macroprudential policy, a key advisory role is undertaken by the Macro-Financial Committee (MFC).
- The MFC includes the members of the Governing Committee and meets monthly to assess financial stability and potential policy responses to risks.

### External oversight and transparency
- The Board of Directors acts as an agent of the Minister of Finance in reviewing how well the RBNZ meets its legal obligations.
- The Board reviews the efforts of the RBNZ to maintain a sound and efficient financial system and monitors the regulatory process of the RBNZ, publishing its conclusions in an annual report.
- Oversight of the Board has been enhanced recently after the Minister of Finance asked the Board to explicitly monitor the regulatory process and how the objectives of soundness and efficiency in the financial sector are balanced.
- All prudential measures are subject to public consultation and the publication of an impact assessment, although consultations have been relatively short and there is room for improving the impact assessment.
- The RBNZ publishes twice a year a financial stability report (FSR) with its views on the risks to the financial sector; the report is reviewed by the Parliament and the Board of Directors.
- The RBNZ has published several papers discussing the macroprudential framework in New Zealand and discloses twice a year an updated data series of the main indicators used to assess imbalances and the need for macroprudential actions.

### Ability to act: powers and toolkit
- The RBNZ has extensive powers to require regulated entities to supply information and has formal arrangements to facilitate information sharing with other agencies including the Treasury and members of the Council of Financial Regulators (CoFR).
- The MoU signed with the Minister of Finance included four instruments in the macroprudential toolkit of the RBNZ:
  - (i) a countercyclical capital buffer;
  - (ii) adjustments in sectoral capital requirements;
  - (iii) restrictions to the share of high loan-to-value ratio (LVR) residential mortgages; and
  - (iv) adjustments to the minimum core funding requirement.
- As a prudential regulator, the RBNZ is also empowered to implement a variety of other tools whose calibration is not expected to fluctuate over the financial cycle such as, the leverage ratio, limits on interbank exposures, liquidity requirements and D-SIFIs buffers.
- Footnote: Leverage ratio and D-SIFIs buffers haven’t been implemented in New Zealand for the time being.
- The MoU is described as a voluntary and effective constraint agreed by the RBNZ to clarify and legitimize a different approach to use of traditional policy instruments.
- The MoU will be reviewed after 5 years, in 2018.
- The MoU provides room for adding instruments to the toolkit with the agreement of the Minister of Finance and requires the RBNZ to advise the Minister of Finance on changes to the macroprudential framework that would extend the use of macroprudential instruments to nonbanks.

### Cooperation arrangements and checks and balances
- The RBNZ operates as an integrated supervisor, resolution authority, macroprudential authority and central bank, facilitating timely sharing of information and collaboration across policy functions.
- The Governing Committee discusses all major monetary and financial decisions falling under the responsibilities of the RBNZ, ensuring coordination across monetary and macroprudential policy actions.
- Internal arrangements to mitigate group thinking include the Macro-Financial Committee debating macroprudential policy and analytical and policy papers prepared by staff; the Governing Committee is another forum to debate competing views.
- External arrangements involve discussions with the Treasury prior to implementation of measures, mandatory public consultation and oversight by the Board.
- The Council of Financial Regulators (CoFR) was formed in 2013 to share information, identify important trends and issues and ensure coordination to respond to financial market events and developments.
  - CoFR members: the Financial Markets Authority (FMA), the RBNZ, the Treasury and the Ministry of Business, Innovation and Employment (MBIE).
  - CoFR meets quarterly and is chaired on a rotating basis between the Governor of the RBNZ and the Chief Executive of the FMA.
  - A CoFR subcommittee (the Banking Forum) discusses regulatory matters relating specifically to registered banks; members include the four permanent CoFR members and other public agencies with interest in the banking sector such as the Ministry of Justice (MoJ) and the Inland Revenue Department (IRD).
- There are bilateral MoUs between the FMA and the RBNZ and between the Treasury and RBNZ dealing with issues such as information sharing.

### Assessment of systemic risks and data
- The RBNZ has a dedicated Macro-Financial Department with two teams and 10 staff that leads systemic risk analysis and macroprudential policy discussions.
- The Macro-Financial Department is in charge of the Financial Stability Report (FSR), published twice a year.
- The FSR covers risks to the financial system, international financial markets, financial risks to the New Zealand economy (such as housing market, agriculture and external sector) and strengths and weaknesses of financial institutions and infrastructure; it also includes boxes on special topics (e.g., implications of the LVR restrictions and dairy farm land valuation).
- The FSR is a key accountability document under the RBNZ Act (section 165A); after receiving the FSR from the RBNZ, the Minister of Finance must present it to Parliament.
- The 2013 MoU requires the FSR to detail the reasons for, and impact of, any macroprudential intervention undertaken by the RBNZ.
- The RBNZ Board reviews each FSR shortly after publication.
- Macroprudential work is done in close collaboration with the Prudential Supervision, Economic and Financial Markets Departments and through cooperation with other government and supervision authorities.
- Data limitations identified:
  - Reports are not sufficiently standardized to allow consistent collection of data across financial entities.
  - Ongoing work to increase standardization and improve data quality of reports of insurance firms and debt-to-income for mortgage customers.
  - Information is not always granular enough for detailed analysis of asset and liability structures or more detailed solvency and liquidity stress tests.
  - Data on market risks is limited and information on liabilities does not cover the type of counterparties and the existence of collateral.
  - Ongoing efforts to improve commercial real estate information.

### Recommendations
- Clarify objectives and role of macroprudential policy:
  - Strengthen efforts to explain the framework and technical matters; communication efforts should continue.
  - Emphasize that the primary role of macroprudential policy is to increase the resilience of the financial sector and contain the procyclical feedback between asset prices and credit (as opposed to controlling the financial cycle per se).
- Strengthen institutional arrangements while preserving independence:
  - Maintain effective accountability of the RBNZ that does not jeopardize the integrity and independence of macroprudential decision making.
  - The Board’s work to develop a framework to enhance monitoring of the RBNZ’s responsibility to maintain a sound and efficient financial system is welcome; the framework should allow effective accountability without jeopardizing decision-making integrity by reviewing actions already taken and not aiming to influence them in advance.
- Enhance cooperation and information sharing:
  - CoFR should work as an active and timely forum where all members express their views and share their analysis.
  - A more detailed description of the objectives and expected role of each CoFR member should help increase Council effectiveness.
  - Develop guidelines for consultation procedures between the Treasury and the RBNZ to ensure both parties are aware of each other’s views without creating obstacles to timely action.
- Improve transparency for extending and updating the macroprudential framework:
  - Procedures to extend the macroprudential framework beyond banks and improve the toolkit should be more transparent.
  - The MoU requires RBNZ and the Minister to agree on changes, but the lack of clear and transparent procedures for adjusting the MoU risks a lengthy process that may undermine timely prudential action.
  - To mitigate this risk, the RBNZ advice and the opinions of the Minister on the need for adjustment should be publicly disclosed and the Minister should provide a response within an appropriate timeframe.
  - A more regular review of the macroprudential toolkit prescribed in MoU than the current five years (e.g., biannually) may be useful.

### Financial cycle risks and tools — key findings
- Stress tests suggest that the financial system in New Zealand is resilient, but some market imbalances remain a source of vulnerability.
- Banks dominate the financial system and have capital and liquidity buffers well above the regulatory minima.
- Despite the expectation of increasing NPLs in the dairy industry, asset quality is high and relatively high profitability provides another cushion to absorb shocks.
- Imbalances that remain sources of vulnerabilities:
  - Housing market imbalances,
  - Increasing household leverage, and
  - Concentration in the banking sector.

### Credit expansion and housing sector imbalances — key findings and indicators
- After a period of rapid expansion, credit growth in New Zealand decelerated after the GFC; credit growth has accelerated recently but credit gap indicators suggest overall credit is still aligned with its long term trend.
- The low interest rate environment after the GFC reduced banks’ funding costs, but increased spreads smoothed the reduction in effective mortgage rates.
- Recent developments:
  - Credit has recently outpaced deposit growth.
  - Household debt-to-income ratio (DTI) has been on a long term increasing trend.
  - Aggregate debt servicing ratio (DSR) has been relatively stable due to a substantial reduction in the cost of loans.
  - Property-related DTI increased to reach 116 percent.
- Housing market imbalances and risks:
  - Mortgage lending increased sharply following a slowdown after the GFC despite restrictive macroprudential measures.
  - Housing prices have grown strongly, particularly in the Auckland area.
  - Long term declining rental yields suggest the possibility of overvaluation.
  - A significant proportion of borrowers are investors and the vast majority of mortgages are repriced within 3 years.
  - Repricing could increase the DSR and lead to defaults when interest rates normalize.

*NEW ZEALAND — INTERNATIONAL MONETARY FUND*

### 27.      Housing price inflation is partially

### 27.      Housing price inflation is partially

### Housing market fundamentals and overvaluation
- Housing supply in New Zealand does not appear to have kept up with demand (Figure 4).
- On the back of strong immigration, population growth has been rapid and well above OECD average.
- The share of real housing investment as percent of GDP has remained below the OECD average.
- The supply problem seems particularly acute in Auckland; supply measures, such as the Unitary Plan, are expected to have a moderating impact on price dynamics only gradually over time.
- International comparisons and fundamentals-based analysis suggest overvaluation in the 20 to 30 percent range.
- Using a fundamentals-based approach (modeling real house price changes as a function of real disposable income, the working-age population, equity prices and the level of short- and long-term interest rates) and an estimation period starting in the year 2000 suggests house price overvaluation in New Zealand has recently increased further to more than 20 percent.
- Regional price increases (year to September): New Zealand as a whole picked up 14.3 percent; Auckland experienced 15.0 percent; Wellington experienced 21.2 percent.

### Investor behavior and leverage
- Rising housing prices have allowed investors to use the equity gain to increase leverage and buy new properties.
- Tax benefits that largely exempt capital gains and relatively limited alternatives for investments have made housing a popular asset category for investments.
- Higher prices have allowed households to use the equity gain to finance the investment of new properties.
- Despite conservative restrictions implemented in recent years, the proportion of investors in the market has increased.
- Household’s debt-to-income have reached record highs and investors are increasing their share of the market.
- Housing loans represent more than half of banks’ assets.

### Use of prudential tools by the RBNZ
- The RBNZ has been actively using LVR tools to address systemic risks from the housing sector, introducing and adjusting exposure limits to high-LVR loans:
  - October-2013: Banks were required to restrict new residential mortgage lending at LVRs over 80 percent (a deposit of less than 20 percent) to no more than 10 percent of the new commitments.
  - November-2015: Three groups of restrictions were created:
    - For investor loans in Auckland, only 5 percent of total new loans could be granted with a LVR above 70 percent.
    - For other loans in Auckland, only 10 percent of total new loans with a LVR above 80 percent could be granted.
    - For loans outside Auckland, only 15 percent of total new loans with a LVR above 80 percent could be granted.
  - October-2016: Geographical differences were removed and restrictions to investors were tightened:
    - On investor loans, only 5 percent of a bank’s total lending to residential property investors can be to borrowers with a deposit less than 40 percent of the house value.
    - On owner-occupier lending, only 10 percent of a bank’s total lending to owner-occupiers can be to borrowers with a deposit less than 20 percent of the house value.
- The RBNZ has adjusted capital requirements for housing loans between 2013 and 2015 to better align risk weights with underlying risk:
  - Average risk weights for banks following the standardized approach are 38 percent.
  - Mortgages in IRB banks are risk weighted in the 25–31 percent range.
  - Changes in correlation increased the requirements by approximately 12 percent considering the mortgage portfolio and 3 percent considering overall capital.
  - Despite the more conservative calibration, risk weights remained substantially below the ones in force until 2008 (50 percent, following Basel I rules).
- Table 2 (Residential Mortgages Capital Requirements) reports average risk weight (percentage) classifications by LVR and loan type (preserved in source).

### Impact and effectiveness of measures
- The LVR restrictions had a substantial impact on strengthening new loans origination standards and gradually reducing the share of high mortgages in the credit portfolio (Figure 5).
- Counterfactual analysis (VAR-based, projecting housing-specific variables conditional on actual macroeconomic behavior) provides mixed evidence on the impact of LVR restrictions:
  - The difference between counterfactual and actual growth of household credit after the 2013 LVR limits is positive but not statistically significant.
  - There is no evidence that LVR limits introduced in 2013 had a prolonged impact on house prices.
  - The 2015 LVR limits do not seem to have affected the rates of credit growth nor the growth of average house prices.
  - The 2015 LVR limits likely contributed to a reduction of the growth of house prices in the Auckland area, at least in the short term.
- Mortgage lending from nonbank institutions has increased but remains small at approximately 1 percent of outstanding mortgages.
- The behavior of other consumer credit does not suggest substantial substitution away from bank mortgage lending.

### Recommendations and policy considerations
- Limits on DTI should become part of the macroprudential toolkit:
  - The RBNZ is discussing introduction of limits DTI in the macroprudential toolkit with the Minister of Finance.
  - Caps on DTI, or measures such as DSI, can usefully complement LVR restrictions, directly targeting risks from record high indebtedness of households.
  - DTI limits can increase resilience by reducing PDs and eventual losses and could contribute to stabilization by reducing demand for loans.
  - DTI and other affordability metrics are operationally challenging due to definition issues; engaging banks long in advance is key.
- If October 2016 LVR adjustments do not substantially reduce current risks, authorities should consider gradually establishing limits on DTI.
  - Price increases tend to reduce effectiveness of LVR restrictions due to refinancing operations, potentially requiring successive tightening; DTI caps may act as automatic stabilizers as they become more binding when house prices grow faster.
  - Reliance on multiple tools reduces distortions compared to using only a single tool; first time home buyers tend to be more affected by LVR restrictions but may be in a better position in terms of income availability relative to investors.
  - When implementing restrictions, authorities should take a gradual approach to mitigate uncertainties and reduce burdens on lenders and borrowers.
- Authorities are encouraged to maintain efforts to reduce distortionary tax benefits and facilitate housing supply:
  - Tax measures announced in the 2015/16 budget apply income tax on profits from property sales for non-primary residences if the house is bought and sold within two years; the government also announced a tightening of reporting and taxation rules for foreign buyers.
  - Authorities are encouraged to continue reducing tax benefits that distort the housing market.
  - Demand-side sectoral tools should be complemented with housing supply measures; supply measures operate with greater lags but are key to reducing the mismatch between supply and demand in the long term.

*Prepared by IMF staff as presented in the source content.*

### 40.      The RBNZ has not formally identified a set of systemic important institutions or

### cr17118 - 40.      The RBNZ has not formally identified a set of systemic important institutions or

### Systemic importance and capital framework
- The RBNZ has not formally identified a set of systemically important institutions or imposed additional capital requirements based on systemic importance.
- The decision reflects a desire to avoid moral hazard associated with the perception that a set of institutions are ‘too big to fail’.
- Given system concentration and the observation that smaller banks tend, in practice, to hold larger capital ratios, the RBNZ considers it more efficient to apply higher requirements to all banks if necessary.
- The RBNZ is conducting a wider review of the capital framework considering changes in international standards proposed by the Basel Committee.
- Regulatory adjustment noted:
  - Removal of the firm size adjustment allowed in the IRB framework for small farm loans.
  - Establishment of LGD floors as a function of the LVR.
  - Requirement for banks to use a term of 2.5 years for the purpose of the maturity adjustment.

### Sectoral concentration and credit risk (housing and agriculture)
- The sectoral concentration of the credit portfolio is high; the majority of bank assets are housing loans.
- Dairy sector is a key systemic risk:
  - Dairy exposures represent around 10 percent of total bank lending.
  - Total agriculture exposures are around 15 percent of total bank lending.
- Agriculture/dairy risk characteristics:
  - All dairy farms are exposed to the same underlying commodity price → high degree of correlation in cash flows across farms.
  - Farm land values are highly correlated with commodity prices and the market is prone to a lack of liquidity during downturns.
  - There is a risk of a significant number of farm loans becoming nonperforming during periods of cash flow pressures.
- Nonperforming loans (NPLs):
  - Current NPLs in the sector remain at around 1.5 percent.
  - Stress tests suggest NPLs could increase significantly, but in isolation should be absorbed from underlying earnings across other bank assets.

### Mortgage borrower DTI distribution (Table 3)
- New Zealand Share of Mortgages by Type of Borrower and DTI (in percent):
  - First home buyers:
    - DTI ≤ 3: 18.9
    - DTI > 3: 81.0
    - DTI > 4: 59.1
    - DTI >5: 34.6
    - DTI > 6: 16.1
    - DTI > 7: 6.5
    - DTI unknown: 0.1
  - Investors:
    - DTI ≤ 3: 14.2
    - DTI > 3: 84.1
    - DTI > 4: 74.6
    - DTI >5: 59.6
    - DTI > 6: 45.0
    - DTI > 7: 32.4
    - DTI unknown: 1.7
- Sources: New Zealand authorities and IMF staff estimates.

### Funding structure, liquidity risk, and prudential liquidity policy
- New Zealand has chronically low savings → economy dependent on borrowing from abroad; banking system has structural reliance on offshore wholesale funding.
- RBNZ prudential liquidity policy introduced in 2010 to address rollover risk by requiring a greater share of debt-funding from long-term instruments.
- Offshore funding remains material and likely to increase due to credit outpacing deposit growth → structural vulnerability to deterioration in global funding conditions.
- Banks tend to hedge currency risk; foreign currency loans are not material.
- Liquidity buffer composition:
  - Significant part of banks’ liquidity buffer is comprised of residential mortgage-backed securities usable in repo operations with the RBNZ.
- RBNZ liquidity framework (BS13) quantitative requirements:
  - One-week mismatch ratio: cannot be less than zero percent at the end of each business day.
  - One-month mismatch ratio: cannot be less than zero percent at the end of each business day.
  - One-year CFR: cannot be less than 75 percent at the end of each business day.
- Comparability with international standards:
  - One-month mismatch ratio conceptually comparable to LCR but definitions of liquid assets, limits, and calibrations differ.
  - Mismatch ratio includes most securities rated BBB- or better (commercial paper, corporate bonds, asset backed securities, registered bank securities, subordinated debt) which may not be eligible in Reserve Bank’s Domestic Market Operations.
  - Applied haircuts (including on New Zealand government securities: one or three percent depending on maturity) tend to be lower than Basel Committee haircuts.
  - The RBNZ’s treatment of contractual inflows and committed facilities is more lenient than the LCR (no 75 percent cap on inflows; allows banks to consider 75 percent undrawn committed lines).
  - Run-off factors in mismatch ratio are less granular than the LCR and based on liability size rather than type of institution or collateral.
- Core Funding Ratio (CFR):
  - Designed to address over-exposure to short-term wholesale funding by lengthening tenor of funding; objectives similar to NSFR.
  - CFR calculation includes: all funding with residual maturity > 1 year, plus 50 percent of tradable debt with residual maturities between 6 months and 1 year, plus tier 1 capital, plus a percentage of non-marketable funding < 1 year depending on liability size.
  - RBNZ view: a 75 percent CFR results in liquidity requirements broadly equivalent to a 100 percent NSFR.
  - CFR timeline:
    - Initially set in 2010 at 65 percent.
    - Increased to 70 percent in mid-2011.
    - Increased to 75 percent in early 2013.
  - The CFR requirement is permanent, though the MoU on macroprudential policy allows temporary adjustments in response to the financial cycle (tool not yet used for this purpose).
  - In response to increases in CFR, banks increased proportion of retail deposits and long-term funding.
- Stress tests concluded the liquidity profile of New Zealand banks is resilient.

### Use of prudential tools and regulatory adjustments
- RBNZ strengthened regulatory framework to reduce reliance on short-term wholesale funding (CFR increases noted above).
- Capital framework adjustments to account for high homogeneity and correlation of farm loans (see “Regulatory adjustment” in Systemic importance and capital framework).
- Liquidity policy (BS13) imposes both quantitative and qualitative requirements; mismatch ratio less conservative than LCR due to broader liquid asset definition.
- The regulatory approach lacks sufficient granularity to differentiate appropriately the structure of assets and liabilities for liquidity purposes.

### Recommendations
- Capital:
  - Capital requirements should better reflect systemic risk posed by concentration of the financial system.
  - The largest banks are systemically important and should be required to hold capital commensurate with the magnitude of externalities of their eventual failure.
  - Considering the importance of market discipline and that smaller banks traditionally maintain larger capital ratios, it does not seem necessary to differentiate the size of the capital buffer by banks.
  - The current RBNZ review of capital requirements is recommended to increase capital thresholds for the SIFIs or the whole system to better reflect concentration risks.
- Liquidity:
  - Liquidity requirements should provide stronger incentives for banks to reduce excessive reliance on provision of liquidity by the Reserve Bank.
  - While New Zealand standards improved liquidity profiles, they do not provide the same incentives as international standards for banks to hold genuinely liquid assets.
  - The broad range of assets considered liquid is necessary given shortage of high-quality liquid assets in New Zealand but results in significant holdings of assets that will be liquid in stress only with central bank support.
  - The regulatory approach should be more granular to differentiate asset and liability structures for liquidity purposes.
  - Further convergence with international standards could potentially reduce existing distortions, recognizing the limited market volume of liquid assets constrains full alignment.

### Appendix I — Impact of LVR restrictions (counterfactual analysis) — key points
- Methodology:
  - VAR estimated on quarterly data starting 1993 with 9 variables (net migration non-NZ citizens per 1000 NZ residents; net migration of NZ citizens per 1000 NZ residents; output gap calculated using one-sided HP filter; floating rate on new mortgage for new customers; number of house sales per 1000 NZ residents; number of new building consents per 1000 NZ residents; annual real growth rate of average NZ house prices; annual real growth rate of house prices in the Auckland region; annual real growth rate of household credit).
  - Conditional forecasts of housing-specific variables (variables 5–9) use actual paths of macroeconomic variables (variables 1–4) after LVR introduction.
  - For 2013 LVR limits: VAR estimated on 1993Q1–2013Q1; forecasts for 2013Q3–2016Q1.
  - For 2015 LVR limits: VAR estimated on 1993Q1–2015Q1; forecasts for 2015Q4–2016Q2.
- Findings:
  - Forecasted household credit growth is above actual credit growth following introduction of 2013 and 2015 LVR limits, but the difference is not statistically significant at the 95 percent confidence level.
  - Impact on New Zealand average house price growth is not statistically significant for either 2013 or 2015 LVR limits.
  - The 2015 LVR limits seem to have reduced price growth in the Auckland area (at least in the short-term).
  - Both 2013 and 2015 LVR limits appear to have reduced the number of house sales in New Zealand.
- Notes on uncertainty:
  - Confidence intervals are very wide due to high volatility of credit and price growth in recent years.
  - For 2015 LVR limits the short forecast period increases forecast uncertainty further.

*International Monetary Fund — cr17118 (excerpts provided in content unit)*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2017/cr17118.pdf_