## 1. An Illustration of the BOP-IIP Dynamics

## Source details

**Canonical URL:** [1. An Illustration of the BOP-IIP Dynamics](https://www.imf.org/-/media/files/publications/cr/2022/english/1cheea2022002.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2022/english/1cheea2022002.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2022/english/1cheea2022002.pdf.json)

---

### A. The Puzzle of Big Acquisition and Small Accumulation
- Key observation:
  - Switzerland’s cumulative financial account balance (CFAB) amounted to CHF1.2 trillion during 2000–2020.
  - The increase in Switzerland’s net IIP was only around CHF220 billion during 2000–2020.
  - The acquisition-accumulation gap was especially pronounced in 2008–20: net acquisition of foreign assets was over CHF600 billion, while net IIP increased by only CHF25 billion.
- Purpose:
  - Explain the acquisition-accumulation gaps using (i) the new Swiss integrated IIP statement and (ii) a cross-country comparison with economies that had large net foreign asset acquisitions but different IIP accumulation outcomes.

### B. Aggregate Drivers from the Integrated IIP Statement
- Revaluation and non-transactional drivers (2000–2020):
  - Acquisition-accumulation gap (aggregate): CHF977 billion.
  - Exchange rate revaluations: negative CHF704 billion — nearly 60 percent of the acquisition-accumulation gap.
  - Exchange rate revaluations were particularly large in 2008–2020: CHF558 billion loss versus transactions of CHF632 billion in that period.
  - “Other changes” (non-transactional stock changes, e.g., reporting population expansions, reclassifications) were also an important contributor to the gap.
  - Price revaluations contributed, but played a minor role relative to exchange rate revaluations and other changes.

### C. Analysis by IIP Component and Sector Behavior
- Exchange rate revaluation losses by component (2008–2020):
  - Direct investment lost CHF235 billion due to exchange rate revaluations.
  - Portfolio investment and reserve assets also experienced sizable negative exchange rate revaluations.
  - Other investment recorded a small exchange rate revaluation gain of CHF17 billion in 2008–2020.
- Sectoral shifts in willingness to invest abroad:
  - Swiss private sector:
    - 2000–2007: net acquisition of foreign assets averaged CHF73 billion a year (Direct investment +CHF27 billion; portfolio investment +CHF51 billion; other investment -CHF5 billion).
    - 2008–2020: turned into a reduction of foreign assets averaging CHF16 billion a year (Direct investment +CHF29 billion; portfolio investment +CHF14 billion; other investment -CHF59 billion).
  - Swiss National Bank (SNB):
    - 2000–2007: small net reduction of reserve assets averaging CHF2 billion a year.
    - 2008–2020: SNB acquired, on average, CHF65 billion reserve assets each year.
    - Result: non-SNB sectors’ net IIP declined by CHF846 billion during 2008–2020, while the SNB’s net IIP increased by CHF869 billion.
  - Implication: SNB’s additional foreign asset acquisitions only partially offset the decline in private investment abroad, contributing to persistent appreciation pressures on the Swiss franc despite large FX intervention operations.

### D. Investment Returns and Asset Composition Effects
- Observed return differentials (2008–2020):
  - Average return on Switzerland’s portfolio investment assets: 4.3 percent.
  - Average return on Switzerland’s portfolio investment liabilities: 5.3 percent.
  - Average return on reserve assets: 3.9 percent.
  - Counterfactual: Had portfolio investment assets and reserve assets achieved the same return rate as portfolio investment liabilities, the acquisition-accumulation gap during 2008–2020 (CHF609 billion, per Table 1) would have been narrowed by around CHF260 billion.
- Asset composition and risk profile:
  - During 2008–2020, average shares in Swiss portfolio investment assets abroad: 55 percent debt securities, 45 percent equities.
  - Over 90 percent of foreign portfolio investments in Switzerland were equities (liabilities side).
  - SNB foreign currency reserves during 2011–2020: equity shares averaged 17.5 percent.
  - Interpretation: Higher share of equities on the liabilities side and lower-risk profile of Swiss assets (and SNB reserves) contributed to lower returns on Switzerland’s foreign assets relative to liabilities, exacerbating the acquisition-accumulation gap.

### E. Implications Highlighted in the Paper
- Principal factors explaining the acquisition-accumulation gaps:
  - Negative exchange rate revaluations driven by franc appreciation.
  - Relatively low returns on foreign assets versus returns on foreign liabilities.
  - Measurement and “other changes” challenges in a small, open economy’s external accounts.
- Suggested directions to reduce future gaps:
  - Encourage robust global diversification by Swiss investors to relieve appreciation pressures on the franc.
  - Improve the currency composition of foreign assets to reduce exchange rate risks.
  - Raise foreign investment returns, for example through improved pension fund performance.

### Exchange rate revaluations and investment returns (post-GFC, 2008–2020)
- The 1-percentage-point return gap between portfolio investment assets and liabilities widens to 2.4 percentage points when exchange rate revaluations are considered.
- During 2008–2020, investment on foreign currency securities accounted for, on average, 80 percent of total reserve assets. This ratio was around 55 percent during 2000–2007.
- For reserve assets, the year of 2000 was excluded from calculations to exclude gains from an extraordinary revaluation of SNB's gold holdings in 2000.
- Negative exchange rate revaluations and lower returns on foreign assets are identified as important factors behind Switzerland’s large acquisition-accumulation gaps.
- The decline of private Swiss investors’ interest in overseas portfolio investment in the post-GFC era, interacting with exchange rate movements (lower return and depreciation of non-CHF holdings), may have contributed to a self-reinforcing appreciation pressure on the franc.

### Cross-country comparison: sample and focus
- Sample: Switzerland and ten other economies that acquired significant amounts of foreign assets during 2008–2020 (Germany, Israel, Japan, Korea, Macao SAR, Netherlands, Norway, Singapore, Sweden, and Taiwan Province of China).
- The 2008–2020 cumulative financial account balances (CFABs) of the economies in the group all exceed 30 percent of their respective 2020 GDPs.
- Two exchange rate–related questions analyzed:
  - Why Israel achieved a larger net IIP increase than its cumulative foreign assets acquisition despite a 45-percent nominal appreciation of the shekel?
  - Why some currencies (NOK, SGD, TWD) appreciated much less than the franc despite larger net acquisitions of foreign assets?

### Switzerland versus Israel: drivers of acquisition-accumulation gap differences
- Exchange rate revaluations:
  - For Switzerland, negative exchange rate revaluations amounted to CHF558 billion, offsetting nearly 90 percent of its net foreign assets acquisitions during 2008–2020.
  - For Israel, exchange rate revaluation loss was ILS4 billion versus net foreign assets acquisition of ILS130 billion.
  - When Switzerland’s IIP is denominated in USD, negative exchange rate revaluations become smaller relative to total transactions, but still sizable (USD381 billion).
- Exposure differences:
  - Israel’s net IIP averaged around 30 percent of GDP during 2008–2020.
  - Switzerland’s net IIP averaged almost 100 percent of GDP in the same period.
- Investment returns (2008–2020, investment return rate = (Price revaluation + investment income)/average stock):
  - Israel: Total assets return 5.0 percent; total liabilities return 4.6 percent.
  - Switzerland: Total assets return 4.0 percent; total liabilities return 4.1 percent.
  - Direct investment: Assets 6.3 percent (Israel) vs 5.3 percent (Switzerland); Liabilities 6.3 percent vs 5.5 percent.
  - Portfolio investment: Assets 8.2 percent (Israel) vs 4.3 percent (Switzerland); Liabilities 5.0 percent vs 5.3 percent.
  - Other investment: Assets 2.8 percent vs 1.5 percent; Liabilities 1.2 percent vs 1.4 percent.
  - Reserve assets: Assets 1.8 percent (Israel) vs 3.9 percent (Switzerland).
- Counterfactual: If Switzerland had Israel’s foreign asset and liability returns, Switzerland would have had around CHF300 billion additional price revaluation gains during 2008–2020.
- Possible explanation for Israel’s higher returns: higher equity share in portfolio investment assets (about 55 percent for Israel vs 45 percent for Switzerland) and potentially higher risk appetite among investors.

### Why some currencies appreciated less than the franc despite larger CFABs
- Inflation-adjusted (REER) relationship:
  - Excluding Switzerland, Israel, and Norway, a regression of REER changes against CFABs produced an R2 of almost 0.9.
- Norway exception:
  - Norwegian krone REER closely linked to oil prices; krone REER depreciated by 16 percent during 2014–2015, accounting for ¾ of total depreciation during 2008–2020.
- Role of non-central-bank sectors and FX interventions:
  - Only Switzerland and Israel recorded net capital inflows by non-central-bank sectors between 2008 and 2020 among the sample.
  - Adding the non-central-bank sectors’ shares in total capital outflows as a regressor increases R2 from 0.01 (full sample REER vs CFAB) to 0.74.
- Reserve assets accumulation and non-central-bank capital flows (2008–2020, % of 2020 GDP) — selected entries:
  - CHE reserve assets accumulation: 115.3; CHE non-central-bank capital flows: -37.9 (% of 2008–2020 CFAB) corresponding to -48.9 (%).
  - SGP reserve assets accumulation: 70.7.
  - TWN reserve assets accumulation: 42.2.
  - ISR reserve assets accumulation: 32.5.

### Summary findings (BOP–IIP section)
- Two important contributors to Switzerland’s large acquisition-accumulation gaps in the post-GFC era:
  - Exchange rate revaluation losses driven by franc strength, currency composition mismatch between foreign assets and liabilities, and large net IIP size.
  - Relatively low returns on foreign assets compared with liabilities, notably for portfolio investment assets.
- Private Swiss sector became more reluctant to invest abroad in the post-GFC era; this lower private-sector outward investment helps explain stronger franc appreciation relative to peers with similar CA balances.
- Differences in investment returns partly attributable to portfolio risk profiles (debt securities vs equities allocation).

---

### Residential real estate market: evidence of overheating and policy responses
- Evidence of overheating and market imbalances:
  - Negative policy interest rates of the SNB since December 2014 coincided with price growth and imbalances in residential real estate and mortgage markets.
  - Mortgage rates hovered at around historically low levels of 1.1 percent during the pandemic.
  - Vacancy rates tightened and declined in 2021 to 1.54 percent from 1.72 percent in 2020.
  - Residential price-to-income and price-to-rent ratios for privately-owned apartments, single-family homes, and apartment buildings are 30 percent above long-term averages.
  - Swiss banking sector relative to domestic economy is the largest in Europe at 500 percent of GDP.
  - The size of the mortgage market in Switzerland is triple that of the euro area.
  - The gap between real house price and real household income growth rates rose from 60 bps pre-Covid to 130 bps.
- Policy responses and regulatory measures:
  - Countercyclical capital buffer (CCyB): activated in February 2013 at 1 percent; increased to 2 percent in January 2014.
  - Self-regulation (SBA rules): minimum down-payment 10 percent from July 2012, compulsory amortization to two thirds within max 20 years; September 2014 shortened amortization to 15 years; January 2020 for investment properties shortened mandatory amortization to 10 years and minimum down-payment of 75 percent for certain cases.
  - FINMA measures: Pillar 2 multipliers on IRB models; mortgages exceeding 80 percent of property value get risk-weight of 100 percent for part over threshold; full Basel III implementation expected in 2024.
- Affordability, leverage, and interest-rate sensitivity:
  - The 75th value of the LTI distribution widened by 40 bps from 2018 to reach 7.2 at end-2021.
  - Loan-to-rent ratios at the 75th percentile: 22 (household investors) and 19 (corporate investors).
  - As of Q4:2021, the LTV of rented-out property at the 90th percentile decreased to 75 percent for both households and companies from 80 and 82 percent in 2020.
  - Estimate: a quarter (half) of new mortgage production could become unaffordable if rates were to increase to 3 percent (5 percent) across all sub-segments.
  - The 5-year fixed mortgage rate was 1.7 percent as of March 2022 (a 60-bps increase since December 2021).
  - Long-term rates are 3.7 percent.
  - Mitigant: Swiss households with mortgage loans tend to have significant buffers in the form of financial wealth.
- Macroprudential transmission and tools:
  - Capital-based tools and borrower-based tools (LTV, DTI, DSTI) discussed; international examples of caps cited.
  - Modeling approach favors vintage structural models over purely statistical methods.

### Modeling approach, calibration, and stress testing
- Model features:
  - Vintage structural model with 252 vintage-LTV buckets (12 LTV buckets; 21 vintages) weighted by outstanding share at end 2021.
  - Loss event defined by a ‘double trigger’: borrower financially distressed (cash-flow test) and house value below loan (balance sheet test).
  - Monte Carlo simulation: for each bucket, 10,000 house price draws; bucket-specific PD equals defaults/10,000 × bucket-specific probability of financial distress; simulation repeated 2,000 times per bucket.
  - Outputs: portfolio-wide PD, LGD, and loss rate.
- Data and segmentation:
  - SNB survey data on new mortgages since 2017, quarterly; covers loans by Swiss banks with domestic mortgage lending volume ≥ CHF 6 billion.
  - Segmentation: SORE and IPRRE; 20 vintages; within each segment/vintage: 12 LTV buckets, 8 DTI buckets, 11 DSTI buckets.
  - Household per capita net worth in Switzerland around CHF 480,000.
- Calibration and validation:
  - Calibrated using Swiss early 1990s loan-loss experience and peer-country crises.
  - Model replicates long-term average default rates; weighted loss rate of 5 basis points replicates long-term average loss rates.
- Stress test scenario (adverse, cumulative changes over 2022–24):
  - Household disposable income/rent: Baseline g=7.5% ; Adverse g=-3.6%
  - Real estate price: Baseline N [0, sdev=15%] ; Adverse N [-25%, sdev=15%]
  - Unemployment: Baseline -0.03 pp ; Adverse +1.7pp
  - Benchmark rate: Baseline 1.33% ; Adverse +300bps
  - Price changes follow a stochastic process with fluctuations of up to 40 percent; unemployment rate increases by 60 percent in the scenario.
- Stress test results under current macroprudential tools:
  - Annualized loss rate of the mortgage portfolio could rise from 5 basis points under the baseline to 90 basis points under stressed conditions.
  - In the IPRRE segment default rates could jump to 4.5 percent.
  - Over the 3-year stress test horizon, losses could reach CHF 27.6 billion (14 percent of CET1 capital).
  - Aggregate CET1 ratio could decline by 230 basis points to 14 percent.
  - The sectoral CCyB would absorb 25 percent of losses.
  - Segment-level projected annualized loss rates:
    - Owner-occupied (SORE): annualized loss rate could increase to 60 bps in the stressed scenario.
    - Income-producing (IPRRE): loss rates could jump to 150 bps in the stressed scenario.
  - Example component changes:
    - SORE: Baseline PD 0.13 ; Baseline LGD 17.01 ; Baseline Loss rate 0.03 ; Adverse PD 2.42 ; Adverse LGD 20.97 ; Adverse Loss rate 0.62
    - IPRRE: Baseline PD 0.32 ; Baseline LGD 30.91 ; Baseline Loss rate 0.10 ; Adverse PD 4.53 ; Adverse LGD 33.13 ; Adverse Loss rate 1.52

### Policy design and effectiveness (borrower-based tools and amortization)
- Vulnerabilities:
  - Recent vintages more exposed to rising rates and falling prices; loan-loss rate of recent vintages between one third (IPRRE) and three quarters (SORE) more than portfolio average.
- Potential policy measures and calibrated impacts:
  - An LTV cap of 70 percent with a 20 percent ‘speed limit’, or increasing amortization requirements to half of the value of the collateral, would reduce by 40 percent the risk of new lending in the SORE segment.
  - A combination of LTV-DSTI limits of 75–25 percent with a 20 percent ‘speed limit’, or increasing amortization rate to 50 percent of lending value, would cut the risk of new vintages by 25 percent in the IPRRE segment.
  - Anchoring risk of new vintages to average portfolio using borrower-based limits would save about CHF 1 billion of regulatory CET1 per year.
- Policy implications and implementation considerations:
  - Swiss authorities re-activated CCyB buffer at maximum 2.5 percent in January 2022 (effective from September 2022); under the adverse scenario the 2.5 percent CCyB buffer could absorb 25 percent of cumulative losses over a 3-year stress period.
  - Use of ‘speed limits’ allows exemptions (e.g., first-time buyers) and flexibility; careful design needed to address distributional concerns.
  - Complementary measures: taxation adjustments, support for rental market, and public consultation with banking sector and stakeholders recommended.

---

### LGD definition, non-linear PD–LGD interaction, and dataset fundamentals
- LGD definition and drivers:
  - LGD defined as outstanding debt net of the discounted recovery value of repossessed collateral, accounting for foreclosure discount δ and discounting by a risk-adjusted spread (equation (4) provided in source).
  - δ denotes the foreclosure discount; n reflects time to sell collateral; spread is the risk-adjusted spread used to discount the risky asset.
- Non-linear PD–LGD interaction:
  - Large house-price corrections increase PD via negative equity and increase LGD when borrowers fail to maintain mortgage payments.
  - Upward shifts in interest rates increase affordability risk (PD) and lower recovery rates (LGD).
  - Policy implication: LTV restrictions or interest-rate affordability tests can help prepare households and banks.
- Dataset coverage and key statistics:
  - SNB survey on new mortgages since 2017; excludes rollover loans with same lender and construction loans.
  - Observations: high-LTV lending declined in IPRRE but remains elevated for SORE; share of high-risk LTV loans (>70 percent) in IPRRE decreased from 41 in Q1:2017 to 36 percent in Q4:2021.
  - Most common maximum DTI cap in peer countries is 4.5.
  - Financial wealth haircut assumptions: 0 percent for cash, 10 percent for debt, 20 percent for equity and collective investment schemes.
  - For SORE: household wealth at the 50th percentile used; for IPRRE: household wealth at the 75th percentile used.

---

### Reliance on natural gas for electricity and heating; no-Russian-gas scenario
- Overview and seasonality:
  - Shares of natural gas in EU (2020): 20 percent in electricity production and 37 percent in gross heat production.
  - About ¾ of natural gas used by households and industry in 2019; households mostly for heating.
  - Switzerland is an importer of gas and contracts storage space in neighboring countries; consumption higher in October–April.
- Import dependence and sources:
  - Russia was slightly more than a half of total gas imported to Switzerland in 2019; for the EU this share was 41 percent in 2019.
  - Share of gas originating in Russia in total imports to Switzerland declined in 2020 to 43 percent.
- No-Russian-gas scenario (assumptions and outcomes):
  - Baseline annual gas imports averaged about 36.6 TWh during 2017–2020.
  - About 80 percent of gas imports occurred in the winter season.
  - Assume half of imported gas comes from Russia.
  - Baseline implied imports from Russia: 3.5 and 14.8 TWh in summer and winter seasons, respectively.
  - Scenario assuming no flows from Russia and unchanged imports from other countries:
    - Switzerland would face a supply gap of 18.3 TWh of natural gas.
    - Fuel switching assumed to compensate for about 20 percent of gas consumption.
    - Remaining gaps after fuel switching: 2.1 TWh in summer and 8.9 TWh in winter.
  - Shortages/rationing may emerge if additional supplies not secured; timing of interruption materially affects impacts.
- Policy responses and recommendations:
  - Actions already taken:
    - Authorities enabled gas companies to jointly procure additional gas imports and LNG terminal and storage capacities for 2022/23 winter.
    - Federal Council required booking gas storage capacity for around 6 TWh (15 percent of annual consumption); about half booked in France.
    - 6TWh of natural gas accessible at short notice for fixed fee to be purchased in France, Germany, Italy, and the Netherlands.
    - Political declaration signed to coordinate gas storage regionally (Pentalateral Energy Forum).
    - Companies asked to prepare risk-management plans; incentives for house refurbishments and streamlining renewable project procedures underway.
  - Recommended immediate measures:
    - Develop and update contingency plans for sector and large consumers.
    - Implement high-frequency monitoring of supplies, distribution, and consumption.
    - Secure alternative gas supplies and increase use of other fuels where feasible.
    - Consider near-term demand-management: voluntary measures, price- or incentive-based, or via smart meters.
    - Consider solidarity agreements with neighboring countries and targeted support for vulnerable parts of the economy.
  - Recommended medium-term measures:
    - Progress with green energy transition by increasing renewable energy contribution and improving energy efficiency.

*Prepared by Li Zeng (EUR); excerpted from “BIG ACQUISITION, SMALL ACCUMULATION: WHY? — A LOOK AT SWITZERLAND’S BOP-IIP DYNAMICS,” IMF staff paper, May 25, 2022.*

### 1. An Illustration of the BOP-IIP Dynamics ________________________________________________ 4

### 1. An Illustration of the BOP-IIP Dynamics

### A. The Puzzle of Big Acquisition and Small Accumulation
- Key observation:
  - Switzerland’s cumulative financial account balance (CFAB) amounted to CHF1.2 trillion during 2000–2020.
  - The increase in Switzerland’s net IIP was only around CHF220 billion during 2000–2020.
  - The acquisition-accumulation gap was especially pronounced in 2008–20: net acquisition of foreign assets was over CHF600 billion, while net IIP increased by only CHF25 billion.
- Purpose:
  - The paper aims to explain the acquisition-accumulation gaps using (i) the new Swiss integrated IIP statement (detailed breakdown of changes in IIP stocks) and (ii) a cross-country comparison with economies that had large net foreign asset acquisitions but different IIP accumulation outcomes.

### B. Aggregate Drivers from the Integrated IIP Statement
- Revaluation and non-transactional drivers (2000–2020):
  - Acquisition-accumulation gap (aggregate): CHF977 billion.
  - Exchange rate revaluations: negative CHF704 billion — nearly 60 percent of the acquisition-accumulation gap.
  - Exchange rate revaluations were particularly large in 2008–2020: CHF558 billion loss versus transactions of CHF632 billion in that period.
  - “Other changes” (non-transactional stock changes, e.g., reporting population expansions, reclassifications) were also an important contributor to the gap.
  - Price revaluations contributed, but played a minor role relative to exchange rate revaluations and other changes.

### C. Analysis by IIP Component and Sector Behavior
- Exchange rate revaluation losses by component (2008–2020):
  - Direct investment lost CHF235 billion due to exchange rate revaluations.
  - Portfolio investment and reserve assets also experienced sizable negative exchange rate revaluations.
  - Other investment recorded a small exchange rate revaluation gain of CHF17 billion in 2008–2020 (interpreted as evidence of sound exchange rate risk management by banks).
- Sectoral shifts in willingness to invest abroad:
  - Swiss private sector:
    - 2000–2007: net acquisition of foreign assets averaged CHF73 billion a year (Direct investment +CHF27 billion; portfolio investment +CHF51 billion; other investment -CHF5 billion).
    - 2008–2020: turned into a reduction of foreign assets averaging CHF16 billion a year (Direct investment +CHF29 billion; portfolio investment +CHF14 billion; other investment -CHF59 billion).
  - Swiss National Bank (SNB):
    - 2000–2007: small net reduction of reserve assets averaging CHF2 billion a year.
    - 2008–2020: SNB acquired, on average, CHF65 billion reserve assets each year.
    - Result: non-SNB sectors’ net IIP declined by CHF846 billion during 2008–2020, while the SNB’s net IIP increased by CHF869 billion.
  - Implication: SNB’s additional foreign asset acquisitions only partially offset the decline in private investment abroad, contributing to persistent appreciation pressures on the Swiss franc despite large FX intervention operations.

### D. Investment Returns and Asset Composition Effects
- Observed return differentials (2008–2020):
  - Average return on Switzerland’s portfolio investment assets: 4.3 percent.
  - Average return on Switzerland’s portfolio investment liabilities: 5.3 percent.
  - Average return on reserve assets: 3.9 percent.
  - Counterfactual: Had portfolio investment assets and reserve assets achieved the same return rate as portfolio investment liabilities, the acquisition-accumulation gap during 2008–2020 (CHF609 billion, per Table 1) would have been narrowed by around CHF260 billion.
- Asset composition and risk profile:
  - During 2008–2020, average shares in Swiss portfolio investment assets abroad: 55 percent debt securities, 45 percent equities.
  - Over 90 percent of foreign portfolio investments in Switzerland were equities (liabilities side).
  - SNB foreign currency reserves during 2011–2020: equity shares averaged 17.5 percent.
  - Interpretation: Higher share of equities on the liabilities side and lower-risk profile of Swiss assets (and SNB reserves) contributed to lower returns on Switzerland’s foreign assets relative to liabilities, exacerbating the acquisition-accumulation gap.

### E. Implications Highlighted in the Paper
- Principal factors explaining the acquisition-accumulation gaps:
  - Negative exchange rate revaluations driven by franc appreciation.
  - Relatively low returns on foreign assets versus returns on foreign liabilities.
  - Measurement and “other changes” challenges in a small, open economy’s external accounts.
- Suggested directions to reduce future gaps (summary from paper’s abstract):
  - Encourage robust global diversification by Swiss investors to relieve appreciation pressures on the franc.
  - Improve the currency composition of foreign assets to reduce exchange rate risks.
  - Raise foreign investment returns, for example through improved pension fund performance.

*Prepared by Li Zeng (EUR); excerpted from “BIG ACQUISITION, SMALL ACCUMULATION: WHY? — A LOOK AT SWITZERLAND’S BOP-IIP DYNAMICS,” IMF staff paper, May 25, 2022.*

### 10.      If exchange rate revaluations are also considered, the returns on foreign assets in the

### 1cheea2022002 - 10.      If exchange rate revaluations are also considered, the returns on foreign assets in the

### Exchange rate revaluations and investment returns (post-GFC, 2008–2020)
- The 1-percentage-point return gap between portfolio investment assets and liabilities widens to 2.4 percentage points when exchange rate revaluations are considered.
- During 2008–2020, investment on foreign currency securities accounted for, on average, 80 percent of total reserve assets. This ratio was around 55 percent during 2000–2007.
- For reserve assets, the year of 2000 was excluded from calculations to exclude gains from an extraordinary revaluation of SNB's gold holdings in 2000.
- Negative exchange rate revaluations and lower returns on foreign assets are identified as important factors behind Switzerland’s large acquisition-accumulation gaps.
- The decline of private Swiss investors’ interest in overseas portfolio investment in the post-GFC era, interacting with exchange rate movements (lower return and depreciation of non-CHF holdings), may have contributed to a self-reinforcing appreciation pressure on the franc.

### Cross-country comparison: sample and focus
- Sample: Switzerland and ten other economies that acquired significant amounts of foreign assets during 2008–2020 (Germany, Israel, Japan, Korea, Macao SAR, Netherlands, Norway, Singapore, Sweden, and Taiwan Province of China).
- The 2008–2020 cumulative financial account balances (CFABs) of the economies in the group all exceed 30 percent of their respective 2020 GDPs.
- The analysis focuses on two exchange rate–related questions:
  - Why Israel achieved a larger net IIP increase than its cumulative foreign assets acquisition despite a 45-percent nominal appreciation of the shekel (second highest in the group)?
  - Why some currencies (NOK, SGD, TWD) appreciated much less than the franc despite larger net acquisitions of foreign assets?

### Question 1 — Switzerland versus Israel: drivers of acquisition-accumulation gap differences
- Exchange rate revaluations:
  - For Switzerland, negative exchange rate revaluations amounted to CHF558 billion, offsetting nearly 90 percent of its net foreign assets acquisitions during 2008–2020.
  - For Israel, exchange rate revaluation loss was ILS4 billion versus net foreign assets acquisition of ILS130 billion.
  - When Switzerland’s IIP is denominated in USD, negative exchange rate revaluations become smaller relative to total transactions, but still sizable (USD381 billion), indicating currency mismatch between foreign assets and liabilities.
- Exposure differences:
  - Israel’s net IIP averaged around 30 percent of GDP during 2008–2020.
  - Switzerland’s net IIP averaged almost 100 percent of GDP in the same period, increasing exchange rate revaluation risk.
- Investment returns (2008–2020, investment return rate = (Price revaluation + investment income)/average stock):
  - Israel: Total assets return 5.0 percent; total liabilities return 4.6 percent.
  - Switzerland: Total assets return 4.0 percent; total liabilities return 4.1 percent.
  - Direct investment: Assets 6.3 percent (Israel) vs 5.3 percent (Switzerland); Liabilities 6.3 percent vs 5.5 percent.
  - Portfolio investment: Assets 8.2 percent (Israel) vs 4.3 percent (Switzerland); Liabilities 5.0 percent vs 5.3 percent.
  - Other investment: Assets 2.8 percent vs 1.5 percent; Liabilities 1.2 percent vs 1.4 percent.
  - Reserve assets: Assets 1.8 percent (Israel) vs 3.9 percent (Switzerland).
- Counterfactual: If Switzerland had Israel’s foreign asset and liability returns, Switzerland would have had around CHF300 billion additional price revaluation gains during 2008–2020.
- Possible explanation for Israel’s higher returns: higher equity share in portfolio investment assets (about 55 percent for Israel vs 45 percent for Switzerland) and potentially higher risk appetite among investors.

### Question 2 — Why some currencies appreciated less than the franc despite larger CFABs
- Inflation-adjusted (REER) relationship:
  - Excluding Switzerland, Israel, and Norway, a regression of REER changes against CFABs produced an R2 of almost 0.9, indicating CFABs are strongly associated with REER changes after inflation adjustment.
- Norway exception:
  - Norwegian krone REER closely linked to oil prices; krone REER depreciated by 16 percent during 2014–2015, accounting for ¾ of total depreciation during 2008–2020.
- Role of non-central-bank sectors and FX interventions:
  - Only Switzerland and Israel recorded net capital inflows by non-central-bank sectors between 2008 and 2020 among the sample.
  - Adding the non-central-bank sectors’ shares in total capital outflows as a regressor increases R2 from 0.01 (full sample REER vs CFAB) to 0.74, suggesting low willingness of non-central-bank sectors to invest abroad helps explain larger appreciations of the franc and shekel.
- Reserve assets accumulation and non-central-bank capital flows (2008–2020, % of 2020 GDP) — selected entries from Table 7:
  - CHE reserve assets accumulation: 115.3 (% of 2020 GDP); CHE non-central-bank capital flows: -37.9 (% of 2008–2020 CFAB) corresponding to -48.9 (%).
  - SGP reserve assets accumulation: 70.7.
  - TWN reserve assets accumulation: 42.2.
  - ISR reserve assets accumulation: 32.5.

### Summary findings
- Two important contributors to Switzerland’s large acquisition-accumulation gaps in the post-GFC era:
  - Exchange rate revaluation losses driven by franc strength, currency composition mismatch between foreign assets and liabilities, and large net IIP size.
  - Relatively low returns on foreign assets compared with liabilities, notably for portfolio investment assets.
- Private Swiss sector became more reluctant to invest abroad in the post-GFC era; this lower private-sector outward investment helps explain stronger franc appreciation relative to peers with similar CA balances.
- Differences in investment returns partly attributable to portfolio risk profiles (debt securities vs equities allocation).

### Policy-relevant observations and recommendations (directions indicated by the study)
- Relieve appreciation pressures on the franc in a fundamental and sustainable way, including through more robust global diversification by Swiss investors.
- Swiss investors should be more active in assessing and improving the currency composition of their foreign assets to limit exchange rate revaluation losses.
- There appears to be room for private Swiss investors to improve returns on foreign investments; for example:
  - Pension funds are important in asset management; the large number and relatively small size of pension funds indicate room for efficiency gains through consolidations.

_Italic: Sources: Haver Analytics, Swiss National Bank Data Portal, Swiss National Bank annual reports, IMF World Economic Outlook database, IMF staff estimates._

### 3.      The residential real estate market shows clear signs of overheating. With negative policy

### 1cheea2022002 - 3.      The residential real estate market shows clear signs of overheating. With negative policy interest rates of the Swiss National Bank (SNB) since December 2014, price growth has resulted in imbalances in residential real estate and mortgage markets.

### Evidence of overheating and market imbalances
- Negative policy interest rates of the SNB since December 2014 have coincided with price growth and imbalances in residential real estate and mortgage markets.
- Investors have increasingly turned to residential real estate in search-for-yield.
- Mortgage rates hovered at around historically low levels of 1.1 percent during the pandemic.
- Pandemic-related effects: households built up savings and increasingly worked from home, accelerating prices.
- Regional concentration: Gains particularly strong in the Lake Geneva and Zurich regions.
- Self-owner-occupied segment (SORE): strong demand and surging preference for ownership have outpaced supply; construction remains below pre-Covid levels.
- Vacancy rates tightened and declined in 2021 to 1.54 percent from 1.72 percent in 2020.
- Residential price-to-income and price-to-rent ratios for privately-owned apartments, single-family homes, and apartment buildings are 30 percent above long-term averages.
- Financial sector scale and mortgage market size:
  - Swiss banking sector relative to domestic economy is the largest in Europe at 500 percent of GDP.
  - The size of the mortgage market in Switzerland is triple that of the euro area.
- Pension funds and insurers: investment property accounts for 20 and 7 percent of assets, while mortgages represent 3 and 5 percent of assets. These lenders represent 2 and 4 percent of the overall mortgage market, respectively.
- The gap between real house price and real household income growth rates rose from 60 bps pre-Covid to 130 bps.

### Policy responses and regulatory measures
- Swiss authorities involved: Federal Council, SNB, FINMA, and the banking sector.
- Countercyclical capital buffer (CCyB):
  - Switzerland activated a residential real estate CCyB in February 2013 at 1 percent.
  - The sectoral CCyB was increased to 2 percent in January 2014 as imbalances persisted.
  - Context at activation: policy rate was zero and inflation rate was -0.3 percent; house prices had risen sharply at an average annual rate of 7.5 percent over the previous two years; mortgage growth exceeded 5 percent.
- Self-regulation (banks) under professional rules of the Swiss Banking Association (SBA):
  - From July 2012: minimum down-payment of at least 10 percent of property value; compulsory amortization to two thirds of property value within a maximum of 20 years.
  - Revision in September 2014: mandatory amortization period shortened to 15 years.
  - January 2020 (investment properties): maximum period for mandatory amortization revised down from 15 to 10 years; borrowers now need to provide a minimum down-payment of 75 percent of the property value.
  - Adjusted rules generally applied to buy-to-let segment following FINMA’s recommendation.
- FINMA supervisory measures to build bank resilience:
  - Introduction of Pillar 2 multipliers on Internal Ratings-based (IRB) models.
  - Adjustment in capital adequacy ordinance so that mortgages exceeding 80 percent of property value have a risk-weight of 100 percent for the part of the loan over the threshold.
  - Full Basel III implementation, expected in 2024, will introduce an output floor and higher risk-weights for riskier loans.

### Affordability, leverage, and interest-rate sensitivity
- Affordability risks remain elevated despite tightening self-regulation.
- Owner-occupied segment:
  - The 75th value of the LTI distribution widened by 40 bps from 2018 to reach 7.2 at end-2021.
- Income-producing residential real estate (IPRRE):
  - Loan-to-rent ratios at the 75th percentile: 22 (household investors) and 19 (corporate investors).
- LTV metrics:
  - As of Q4:2021, the LTV of rented-out property at the 90th percentile decreased to 75 percent for both households and companies from the 80 and 82 percent ratios reached in 2020 but remains elevated.
- Interest-rate shock exposure:
  - Estimate: a quarter (half) of new mortgage production could become unaffordable if rates were to increase to 3 percent (5 percent) across all sub-segments.
  - The 5-year fixed mortgage rate was 1.7 percent as of March 2022 (a 60-bps increase since December 2021).
  - Long-term rates are 3.7 percent, making an increase to 3 percent non-negligible.
- Mitigant: Swiss households with mortgage loans tend to have significant buffers in the form of financial wealth that they could use to fill potential liquidity shortfalls.
- Affordability risk definition used: imputed costs from mortgage servicing (including amortization, interest, and maintenance) exceed one-third of income (owner-occupied segment) or rents (investment-led segment). Computation assumes average LTV is 80 percent and maintenance costs represent 1 percent of loan value.

### Macroprudential transmission and tools
- Capital-based tools:
  - Capital buffers on real estate exposures absorb unexpected losses and can reduce credit demand via price effects (pass-through to mortgage rates) and quantity effects (reduced credit supply).
  - EU example: CRD V increased flexibility of systemic risk buffer to target sectoral real estate risks.
- Borrower-based tools:
  - Instruments include LTV, DTI and DSTI regulatory limits on mortgage lending.
  - LTV restrictions reduce bank losses given default; debt serviceability requirements (DTI, DSTI) reduce borrower probabilities of default.
- International comparisons and policy choices:
  - Some EU countries set LTV caps on investment-led segments (examples cited: Ireland and Latvia 70 percent LTV caps; Czech Republic 60 percent pre-Covid; Belgium 80 percent threshold with first-time buyer exemption).
  - Germany announced a sectoral SyRB on residential exposures—2 percent from February 2023, in addition to a 0.75 percent CCyB. Other jurisdictions cited with CCyB rates or planned increases include Bulgaria, Czech Republic, Luxembourg, Slovakia, Norway, Estonia, Iceland, Sweden, and the UK (detailed rates and timing noted in source).

### Modeling approach to quantify banking system resilience
- Rationale:
  - A vintage approach is needed because borrower-based tools apply to new mortgage production while sectoral capital buffers apply to outstanding portfolios.
  - Vintage analysis captures loan maturation and external factor effects (real estate prices, interest rates, disposable income) and dispersion of delinquency across vintages.
  - Structural models are preferred over purely statistical approaches due to limited historical crisis data, structural changes in housing finance, and regulatory adjustments; average loss rates in the mortgage portfolio have been just 5 basis points over the last 20 years.
- Model features:
  - Use of a granular structural model by vintage to project mortgage risk.
  - Loss event defined by a ‘double trigger’: borrower is financially distressed (cash-flow test) and house value is lower than loan (balance sheet test).
  - Financial distress probability (Pr(FD)) is modeled as a function of:
    - Debt service-to-income ratio (DSTI);
    - Change in DSTI (ΔDSTI);
    - Likelihood of being unemployed (U) and change in unemployment (ΔU);
    - Demographic factor (D);
    - Non-linear (exponential) effects of affordability shocks across DSTI buckets.
  - Economic default occurs if financially distressed borrower cannot repay by selling the house (net of transaction cost C) or drawing down liquid financial assets (FinWealth); NPV of the loan includes outstanding loan amount and penalty for early prepayment tied to locked-in mortgage rate and remaining maturity.
  - Monte Carlo simulation structure:
    - Simulations run over more than 250 vintage-LTV buckets.
    - Final implementation uses 252 vintage-LTV buckets (12 LTV buckets; 21 vintages) weighted by outstanding share at end 2021.
    - For each vintage-LTV bucket: N borrowers in distress considered; for each borrower 10,000 house price draws are generated from distribution with mean equal to average house price level in the tail risk scenario; condition for default evaluated; bucket-specific PD equals defaults divided by 10,000 multiplied by bucket-specific probability of financial distress.
    - To reduce simulation noise, the simulation process is repeated 2,000 times for each bucket.
  - Outputs: portfolio-wide PD, LGD, and loss rate calculated by combining bucket-vintage outputs and weighting by outstanding shares.
- Calibration and conservative assumptions:
  - The model conservatively assumes banks charge a positive penalty for early prepayment even if market rates rise, calculated as the net present value of foregone interest payments for a specified number of months.
  - The 5-year fixed mortgage rate is used to discount future interest payments in penalty calculations.
  - The model does not consider strategic defaults (non-recourse loans are not relevant in Switzerland).

*Source: IMF staff analysis in the provided chapter content.*

### 18.      Upon default, a bank’s loss given default (LGD) is driven by the discounted sale price

### 18.      Upon default, a bank’s loss given default (LGD) is driven by the discounted sale price 

### LGD definition and drivers
- LGD is defined as the outstanding debt net of the discounted recovery value of repossessed collateral, taking into account the foreclosure discount and discounting by a risk-adjusted spread. The source formula is:
  퐿퐿퐿퐿퐷퐷
  푖푖,푡푡
  =푁푁퐵퐵 푁푁�퐿퐿
  푖푖,푡푡
  ,푟푟
  푡푡
  푡푡푡푡푡푡푡푡,푀푀
  ,푟푟
  푡푡
  푓푓
  ,푇푇
  푡푡,푠푠
  �−
  (
  1−훿훿
  )
  ⋅
  푃푃
  �
  �
  푖푖,푡푡+푛푛
  �
  1+푟푟
  푡푡
  푓푓
  +푠푠푡푡푟푟푡푡 푠푠푠푠
  �
  푛푛
  (equation (4))
- Definitions from the source:
  - The first term denotes the outstanding debt and the second term the recovery value.
  - δ denotes the foreclosure discount at which the bank sells the property at time t+n, where n reflects the time needed to sell off the collateral.
  - spread is the risk-adjusted spread used to discount the value of the risky asset.

### Non-linear PD–LGD interaction and policy implications (paragraph 19)
- Key mechanisms:
  - Large house-price corrections increase the probability of default (PD) via negative equity.
  - House-price declines also increase LGD when borrowers fail to maintain mortgage payments.
  - Upward shifts in interest rates increase affordability risk (PD) and lower the recovery rate of defaulted loans (LGD).
- Policy implication:
  - LTV restrictions or interest-rate affordability tests can help prepare households and banks for sharp housing-market reversals or rising interest rates.

### Dataset (paragraphs 20–24)
- Data source and coverage:
  - SNB survey data on new mortgages since 2017, quarterly, covering all loans granted by Swiss banks with a domestic mortgage lending volume of at least CHF 6 billion.
  - Includes new mortgages that finance purchase of real estate and commutations; excludes rollover loans with the same lender and construction loans; outliers removed for calibration.
- Recorded loan characteristics:
  - Borrower, type of loan, credit limit, income, value of collateral, rent, interest rate, down payment.
- Segmentation:
  - Type of business transaction: SORE and IPRRE segments.
  - Vintage: 20 vintages.
  - Within each segment/vintage: 12 LTV buckets, 8 DTI buckets (LTI basis), and 11 DSTI buckets.
  - Correlation-structure matrices constructed for 96 LTV/DSTI buckets, and 132 LTV/DTI buckets.
- Observations and statistics:
  - High-LTV lending declined in the IPRRE segment but remains elevated for self-occupiers (SORE).
  - Share of high-risk LTV loans (>70 percent) in IPRRE decreased from 41 in Q1:2017 to 36 percent in Q4:2021.
  - Share of loans in [70%, 80%] bucket with DTI exceeding 5 reached around 30 percent in both SORE and IPRRE.
  - Most common maximum DTI cap in peer countries is 4.5 (international comparison notes).
- Repricing and vintage reconstruction:
  - Mortgage stock and aggregate inflows available starting in 1990.
  - Interest-rate repricing data at issuance available starting in 2017 and extended back to 1990 assuming similar pattern to first vintage.
  - Around two-f ifths of new loans reprice within a year (source note).
  - From 1990: 6 repricing buckets used and a uniform distribution of quarterly repricing within each bucket assumed.
  - For each vintage, the repriced interest rate prevailing at end 2021 is estimated and interest-rate shocks applied with pass-through driven by repricing schedules.
- Financial wealth accounting:
  - Household per capita net worth in Switzerland around CHF 480,000.
  - Wealth data from Swiss Wealth Statistics by individual used; merged with SNB wealth breakdown and haircut by instrument: 0 percent for cash, 10 percent for debt, and 20 percent for equity and collective investment schemes; insurance, pension schemes, and real estate excluded.
  - For SORE segment: household wealth at the 50th percentile used.
  - For IPRRE segment: household wealth at the 75th percentile used.

### Model calibration and validation (paragraphs 25–28)
- Calibration targets and historical episodes:
  - Calibrated using loan-loss experience in Switzerland in the early 1990s and real estate crises in peer countries.
  - 1990s Swiss recession used to estimate default-rate elasticity to macroeconomic risk factors, allocation of stressed sales between unemployment and interest-rate shocks, and transaction costs.
  - UK real estate crisis of the 1990s used to calibrate relationship between default rates and financially distressed borrowers, sensitivity of financial distress to initial unemployment, and elasticity of financial distress to shifts in debt serviceability ratios by DSTI buckets.
- Point-in-Time (PiT) risk parameter computation:
  - PiT parameters computed for each LTV-vintage bucket, accounting for self-regulation amortization requirements, the housing cycle, macrofinancial fluctuations, and repricing schedule from issuance to December 2021.
  - Implied amortization rate estimated consistent with aggregate SNB mortgage stocks and flows.
  - Self-regulation minimum amortization requirements applied, depending on LTV at origination and vintage (adjustments in 2012, 2014, 2020 considered).
- Model validation:
  - Model replicates long-term average default rates under baseline conditions.
  - Weighted loss rate of the mortgage portfolio predicted by the model of 5 basis points replicates the long-term average of loss rates in the Swiss banking system.

### Stress test scenario and key assumptions (paragraphs 27 and Figure 5)
- Scenario design:
  - 3-year inflationary scenario combined with a domestic recession to test resilience to a deep recession, sharp real estate price falls, and higher global interest rates.
  - Baseline projections based on IMF’s WEO forecast; adverse projections based on an inflationary recession scenario using econometric techniques.
  - Size of shocks aligned with SNB scenarios referenced.
- Adverse scenario cumulative changes over 2022–24 (as presented in Figure 5):
  - Household disposable income/rent: Baseline g=7.5% ; Adverse g=-3.6%
  - Real estate price: Baseline N [0, sdev=15%] ; Adverse N [-25%, sdev=15%]
  - Unemployment: Baseline -0.03 pp ; Adverse +1.7pp
  - Benchmark rate: Baseline 1.33% ; Adverse +300bps
- Note on stochastic behavior:
  - Price changes follow a stochastic process with fluctuations of up to 40 percent.
  - The unemployment rate increases by 60 percent in the scenario.
  - Tenant rent payments are less affected due to Swiss regulation allowing rent adjustments when interest rates increase.

### Stress test results under current macroprudential tools (paragraphs 29, Figures)
- Aggregate resilience and losses:
  - Annualized loss rate of the mortgage portfolio could rise from 5 basis points under the baseline to 90 basis points under stressed conditions.
  - In the IPRRE segment default rates could jump to 4.5 percent.
  - Over the 3-year stress test horizon, losses could reach CHF 27.6 billion (14 percent of CET1 capital).
  - Aggregate CET1 ratio could decline by 230 basis points to 14 percent.
  - The sectoral CCyB would absorb 25 percent of losses.
- Distributional impacts and second-round effects:
  - Some banks could breach their capital conservation buffers and could curtail credit to financially-distressed borrowers, amplifying the shock.
  - Consumption would be affected by rising living costs, reduced real household disposable income, higher debt-servicing costs, and lower household wealth, creating negative feedback loops and deepening the recession.
- Segment-level projected annualized loss rates (figures from stress results):
  - Owner-occupied (SORE) segment: annualized loss rate could increase to 60 bps in the stressed scenario.
  - Income-producing (IPRRE) segment: loss rates could jump to 150 bps in the stressed scenario.
- Example component changes presented (Stress Test Results charts):
  - Self-Occupied Segment (Percent): Baseline PD 0.13 ; Baseline LGD 17.01 ; Baseline Loss rate 0.03 ; Adverse PD 2.42 ; Adverse LGD 20.97 ; Adverse Loss rate 0.62
  - Income Producing Segment (Percent): Baseline PD 0.32 ; Baseline LGD 30.91 ; Baseline Loss rate 0.10 ; Adverse PD 4.53 ; Adverse LGD 33.13 ; Adverse Loss rate 1.52

### Calibration and design of borrower-based macroprudential tools (paragraphs 30–33)
- Vulnerabilities:
  - Recent vintages are more exposed to rising interest rates and declining house prices because they have repaid less principal, experienced smaller home equity gains, and benefited less from income growth.
  - Loan-loss rate of recent vintages is between one third (IPRRE segment) and three quarters (SORE segment) more than that of the portfolio on average.
- Potential policy measures:
  - Reduce maximum period of amortization from 15 to 10 years in the owner-occupied or buy-to-let segment.
  - Increase amortization rate by requiring borrowers to pay down the mortgage to less than two thirds of the lending value of the property.
  - Introduce borrower-based instruments: LTV, DTI, DSTI, or a combination.
- Design considerations:
  - Whether caps are hard limits applied to all new issuances or speed limits reducing the volume of high-risk lending to below a specific share of new commitments.
  - Whether restrictions apply temporarily when financial risk is elevated or address structural risk more permanently.
  - Account for lags between policy announcements and effects by introducing macroprudential tools before the adverse scenario materializes.
- Implementation assumptions used in counterfactual analysis:
  - In absence of new measures, mortgage flows in the next 8 quarters share the average risk characteristics of issuance observed in the last four quarters of data.
  - Macroprudential limits reduce the relative mass of high-LTV or DSTI above the limits to zero if ‘hard’ limits are introduced, or to a specific percentage if ‘speed limits’ are applied.
  - Assumed bunching of new loans just below regulatory limits, as observed empirically.
  - During the 8 quarters before the stress horizon, house prices and macrofinancial factors follow April 2022 WEO assumptions.
  - Losses on the aggregate portfolio under the adverse scenario are compared with losses on new vintages under ‘No limits’ and under each policy intervention.

*Source: IMF staff analysis as presented in the provided chapter text.*

### 34.      Introduction of an LTV-DSTI cap with a ‘speed limit’ or an increase in amortization

### 34.      Introduction of an LTV-DSTI cap with a ‘speed limit’ or an increase in amortization

### Impact on expected losses by vintage and tool calibrations
- Different combinations of macroprudential limits can have a similar impact on expected losses of new mortgages (Figure 8).
- A simple rule to select tools/limits: target the risk of new vintages to the average risk of older vintages.
- Specific calibrations and impacts reported:
  - An LTV cap of 70 percent with a 20 percent ‘speed limit’, or an increase of amortization requirements to half of the value of the collateral (from two thirds under current self-regulation), would reduce by 40 percent the risk of new lending in the SORE segment.
  - A combination of LTV-DSTI limits of 75–25 percent with a 20 percent ‘speed limit’, or an increase in the amortization rate to 50 percent of the lending value of the property, would cut the risk of new vintages by 25 percent in the IPRRE segment.
- Figure 8 note: The red bars represent the cumulative loss rate under the adverse scenario over the 3-year stress test horizon, and current macroprudential policy settings. The blue bars denote the stressed loss rate under alternative policy tools and calibrations (new instruments). The grey bars show the stressed loss rates under adjustments of self-regulation rules (current instruments).

### Estimated regulatory capital effects and methodology
- Anchoring the risk of new vintages to the average portfolio using borrower-based limits would save about CHF 1 billion of regulatory CET1 by year.
- Methodology to estimate average impact:
  - Use the Basel III IRB supervisory formula to back out the effective maturity to match Swiss banks risk-weights on the mortgage portfolio to the reported PDs and LGDs.
  - Use the structural model to project PDs and LGDs of new vintages (no limits) and plug them into the IRB formula—adjusted to create Through-the-Cycle (TTC) projections—and calculate capital requirements.
  - Compute capital required on new vintages (limits targeting the risk of the portfolio), assuming that 10 percent of the portfolio is being added yearly.
  - The difference between the two amounts equals the savings in banks’ regulatory capital from applying the borrower-based tools calibrated in the paper.

### Policy implications and macroprudential context
- A real estate crisis could have a substantial impact on macrofinancial stability in Switzerland; the mortgage market is very large relative to the size of the economy and banks are highly exposed with geographical concentration at the canton level.
- Despite historically low default rates, rising vulnerabilities across mortgage and real estate sub-segments exist; a sharp price correction would reduce household wealth, affect investment and consumption, and could lead banks to accentuate downturns through credit rationing.
- The Swiss authorities have re-activated the CCyB buffer:
  - The Federal Council approved the SNB’s proposal to reactivate the buffer at its maximum 2.5 percent level in January 2022 (effective from September 2022).
  - Under the adverse scenario presented, the 2.5 percent CCyB buffer could absorb 25 percent of cumulative losses over a 3-year stress period.
- Borrower-based tools (LTV/DSTI caps, ‘speed limits’, increased amortization) could:
  - Guard against the build-up of vulnerabilities in new lending flows by reducing the share of homebuyers that could become financially stretched.
  - Decrease losses under severe stress and save regulatory capital under baseline conditions.
  - Contribute to more sustainable mortgage lending and dampen house price fluctuations.

### Implementation considerations and complementary measures
- Use of ‘speed limits’:
  - A ‘speed limit’ allows for a certain proportion of the volume of new loans to be exempt from a particular measure; these limits could be unconditional or targeted to specific types of loans (e.g., first-time home buyers, green mortgage loans, etc.).
  - Speed limits allow flexibility in implementation and can be easier to enforce, minimizing distortions to allocative efficiency and mortgage-market access.
- Distributional concerns:
  - Borrower-based measures could have a disproportionate impact on first-time home buyers or lower-income households; careful design (including targeted exemptions via speed limits) is important.
- Additional policy levers beyond macroprudential tools:
  - Adjustments to taxation (e.g., abolition of imputed-rent taxation or phase-out of mortgage interest relief).
  - Actions to support the rental market (e.g., targeted subsidies, social housing) to help decrease affordability risk and support sustainable house prices.
- Process recommendation:
  - The Swiss authorities could start public consultation with the banking sector and relevant stakeholders to complete the macroprudential framework in case affordability risks keep increasing and new tools need to be activated.

*Source: IMF staff analysis in "Introduction of an LTV-DSTI cap with a ‘speed limit’ or an increase in amortization" (section text and Figure 8).*

### 6.      Reliance on natural gas for electricity and heating in Switzerland is relatively less than

### 6.      Reliance on natural gas for electricity and heating in Switzerland is relatively less than in the EU.

### Overview
- The shares of natural gas in electricity production and gross heat production were 20 and 37 percent, respectively, in the EU in 2020.
- Reliance on natural gas varies significantly across countries in Europe and among Switzerland’s neighbors.

### Natural gas usage and seasonality
- About ¾ of natural gas was used by households and industry in 2019. Households mostly used natural gas for heating.
- Among other sectors in 2019, about a third of natural gas was used by manufacturing; public administration and real estate activities, professional, and scientific, technical and administrative activities consumed about 5 percent each.
- Switzerland’s natural gas consumption and imports vary significantly across seasons:
  - Consumption is higher during winter months (October–April) and low during summer (May–September).
  - Switzerland is an importer of gas from EU countries and contracts for storage space in neighboring countries, notably in France, but does not maintain sizable supplies domestically.
  - Storage arrangements in EU countries may give priority to EU consumers during periods of tight supply.

### Import dependence and sources
- Russia was the largest natural gas supplier for both Switzerland and the EU in 2019:
  - Slightly more than a half of total gas imported to Switzerland originated in Russia in 2019.
  - For the EU, this share was 41 percent in 2019.
  - The share of gas originating in Russia in total imports to Switzerland declined in 2020 to 43 percent.
- Other important gas importers for Switzerland are Norway and the EU countries (transshipment).
- Most imported gas is bought at a spot market.

### Scenario analysis: impact of a stop in Russian gas imports
- Baseline assumptions:
  - Switzerland’s annual gas imports averaged about 36.6 TWh during 2017–2020.
  - About 80 percent of gas imports occurred in the winter season.
  - It is assumed that half of the imported gas comes from Russia.
  - Baseline implied imports from Russia of 3.5 and 14.8 TWh of natural gas in the summer and winter seasons, respectively.
- No-Russian-gas scenario assumptions and outcomes:
  - Scenario assumes no flows from Russia and imports from other countries kept unchanged.
  - Switzerland would face a supply gap of 18.3 TWh of natural gas.
  - It is assumed fuel switching could compensate for about 20 percent of gas consumption.
  - Remaining gaps after fuel switching: 2.1 TWh in summer and 8.9 TWh in winter.
  - The magnitude of the shock suggests shortages/rationing may emerge if additional supplies are not secured.
  - Dual-fuel plants may help in industry, but less so for households.
  - Timing of interruption would materially affect impacts; substitution from alternative suppliers would reduce mismatches, but neighboring EU countries would likely face similar pressures.

### Policy implications and measures underway
- Actions already taken:
  - Authorities created conditions for gas companies to jointly procure additional gas imports and LNG terminal and storage capacities for the 2022/23 winter season.
  - In late May 2022, the Federal Council required booking gas storage capacity for around 6 TWh (15 percent of Switzerland’s annual consumption), about a half of which has already been booked in France.
  - In addition, 6TWh of natural gas that can be accessible at short notice for a fixed fee will be purchased in France, Germany, Italy, and the Netherlands.
  - Switzerland and Austria, Belgium, France, Germany, Luxembourg, and the Netherlands signed a political declaration to coordinate gas storage on a regional level (Pentalateral Energy Forum).
  - Authorities have asked key companies to prepare risk-management plans in case of electricity-supply shortages.
  - Incentives for house refurbishments are being implemented to reduce energy consumption over time.
  - Streamlining procedures for renewable energy projects is underway to tilt the energy mix towards green energy, recognizing lead times and potential equipment bottlenecks.
- Recommended immediate measures:
  - Develop and update contingency plans for the sector and large consumers.
  - Implement high-frequency monitoring of natural gas supplies, distribution, and consumption.
  - Secure alternative gas supplies and increase use of other fuels where feasible.
  - Consider near-term demand-management efforts: initially voluntary, potentially price- or incentive-based, or through use of smart meters.
  - Consider solidarity agreements with neighboring countries to coordinate supply and demand responses.
  - Consider temporary and targeted measures to support parts of the economy most vulnerable to interruptions in gas supplies.
- Recommended medium-term measures:
  - Progress with green energy transition by increasing contribution from renewable energy sources and improving energy efficiency.

*Source: IMF staff summary of 1cheea2022002 - 6.      Reliance on natural gas for electricity and heating in Switzerland is relatively less than*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2022/english/1cheea2022002.pdf_