## wpiea2024079-print-pdf - Section IV offers concluding remarks with policy implications

## Source details

**Canonical URL:** [wpiea2024079-print-pdf - Section IV offers concluding remarks with policy implications](https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024079-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2024/english/wpiea2024079-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2024/english/wpiea2024079-print-pdf.pdf.json)

---

### Executive Summary — Context and objective
- The world economy experienced a series of unprecedented shocks over the past three years, disrupting supply chains, causing a deep recession, and pushing inflation to the highest level since the 1970s.
- Inflation declined from a peak of 11.6 percent in 2022 to 5.3 percent in 2023, but long-term vulnerabilities persist.
- Climate change is identified as one of the most significant risks to the global economy and financial markets due to greater frequency and severity of extreme weather conditions.
- Objective: Investigate the impact of weather anomalies—measured by unseasonal deviation in temperature from the historical average—on global supply chain pressures and inflation dynamics.

### Data and methods
- Monthly data for six large and well-diversified economies: China, the Euro area, Japan, Korea, the United Kingdom, and the United States.
- Sample period: 1997-2021.
- Primary econometric approach: structural vector autoregressive (SVAR) model estimated for each country to trace contemporaneous effects of weather anomalies on supply chains and inflation.
- Robustness check: local projection (LP) method to trace dynamic effects of temperature anomalies over time.
- Endogenous variables: weather anomalies, Supply Chain Pressures Index (SCPI), and consumer price inflation (headline or core CPI).
- Inflation definition: 휋_{i,t} = (CPI_{i,t} / CPI_{i,t−12}) * 100; transformed via inverse hyperbolic sine: ln(휋_{i,t} + √(휋_{i,t}^2 + 1)).
- SCPI: constructed by the Federal Reserve Bank of New York, measured in standard deviations from the sample average and normalized to have an average of zero.
- Weather anomalies: monthly mean temperature deviation from historical average for that month during the reference period 1901-1996; for Euro area a GDP-weighted average of 19 countries is used.
- Extreme weather events defined as anomalies large relative to historical variation.
- Stationarity: Augmented Dickey-Fuller (1981) tests indicate all series are stationary after logarithmic transformation (Appendix Table A1).

### Mechanisms linking temperature shocks to inflation
- Three principal transmission channels highlighted:
  - Higher energy demand and prices leading to second-round effects on non-energy prices.
  - Lower individual productivity in climate-sensitive sectors altering inflation dynamics.
  - Supply chain disruptions causing productivity shortfalls, higher transportation and production costs, and consequently higher inflation.
- Conceptual emphasis that opposing effects on demand and supply can operate simultaneously.

### Key empirical findings (overview)
- Weather anomalies can contribute to supply chain disruptions and subsequently lead to inflationary pressures.
- Results are based on high-frequency data and are robust to alternative estimation methodologies.
- Significant heterogeneity across countries in the sample, attributed to differences in the severity of weather shocks and vulnerability to supply chain disruptions.
- The impact of weather shocks on supply chains and inflation dynamics is likely to become more pronounced with accelerating climate change that can have non-linear effects.
- The empirical results do not always show a strong positive link between weather shocks and supply chain disruptions; partial explanation is the use of aggregated supply-side disruption measures.
- Example of real-world disruption: a severe drought reduced water level considerably in the Panama Canal, disrupting the trade route connecting Asia and North America.
- Supporting empirical context from literature and data:
  - Global land temperature increase reached more than 1.1 degrees Celsius (°C) compared with the preindustrial average.
  - Projections: mean temperature may rise above the 1.5°C threshold in the near term and by as much as 4°C over the next century absent a global green transition.
  - Supply chain disruptions persisting a month or longer already occur every 3.7 years on average.

### Descriptive statistics (Table 1: summary statistics reproduced exactly)
- Temperature (Mean):
  - China: 0.9592
  - EA: 1.2417
  - Japan: 1.2687
  - Korea: 1.1846
  - UK: 0.8603
  - US: 0.9203
- Temperature (Variance):
  - China: 0.7228
  - EA: 1.3944
  - Japan: 0.7898
  - Korea: 1.3405
  - UK: 0.9623
  - US: 0.8936
- Temperature (Minimum):
  - China: -1.7583
  - EA: -2.4578
  - Japan: -1.7947
  - Korea: -3.1941
  - UK: -3.7659
  - US: -1.8784
- Temperature (Maximum):
  - China: 3.9747
  - EA: 4.0698
  - Japan: 3.9183
  - Korea: 4.3788
  - UK: 3.2510
  - US: 3.7734
- Supply chain pressures (Mean):
  - China: -0.0216
  - EA: -0.0098
  - Japan: -0.0224
  - Korea: -0.0177
  - UK: -0.0122
  - US: -0.0192
- Supply chain pressures (Variance):
  - China: 0.9381
  - EA: 0.9931
  - Japan: 0.9330
  - Korea: 0.9605
  - UK: 0.9827
  - US: 0.9523
- Supply chain pressures (Minimum):
  - China: -1.7932
  - EA: -3.2800
  - Japan: -2.7560
  - Korea: -7.1594
  - UK: -4.1183
  - US: -1.8996
- Supply chain pressures (Maximum):
  - China: 4.3582
  - EA: 3.5000
  - Japan: 3.6264
  - Korea: 3.1049
  - UK: 3.3160
  - US: 4.0856
- Headline Inflation (Mean):
  - China: 1.0826
  - EA: 1.1642
  - Japan: 0.0618
  - Korea: 1.4928
  - UK: 1.3602
  - US: 1.4024
- Headline Inflation (Variance):
  - China: 1.1059
  - EA: 0.3301
  - Japan: 0.5586
  - Korea: 0.3863
  - UK: 0.1640
  - US: 0.4095
- Headline Inflation (Minimum):
  - China: -1.5297
  - EA: -0.5688
  - Japan: -1.6528
  - Korea: -0.4141
  - UK: 0.1985
  - US: -1.4863
- Headline Inflation (Maximum):
  - China: 2.8716
  - EA: 2.3124
  - Japan: 2.0192
  - Korea: 2.9522
  - UK: 2.2814
  - US: 2.6493
- Core Inflation (compact format as presented):
  - Mean row: 0.9568 1.1459 -0.1151 1.3880 1.2145 1.4493
  - Variance row: 0.3686 0.0947 0.4221 0.3173 0.1284 0.0611
  - Minimum row: -1.2995 0.3900 -1.2507 -0.6502 -0.1288 0.6184
  - Maximum row: 1.6187 1.7191 1.6746 2.5116 2.0479 2.4044

### Econometric strategy and identification
- Three-variable SVAR model with Cholesky decomposition and causal ordering: weather anomalies → supply chain pressures → inflation.
- Temperature anomalies assumed strictly exogenous for short-run effects.
- AB-model SVAR specification estimated for each country separately, with matrices A (upper triangular) and B (diagonal).
- Reduced-form representation with Φ_j coefficients upper triangular.
- Optimal lag length determined by SIC criteria: 1 to 2 lags depending on country.
- Robustness of inference: 2,000 bootstrapped error bands for impulse response coefficients and 68 percent confidence intervals.

### Empirical results (baseline)
- Weather shocks → supply chain pressures:
  - Larger temperature deviations tend to exert upward pressure on supply chain pressures, though not statistically significant for all countries.
  - Verbatim quantitative finding: "a 1°C increase in temperature with respect to historical average increases supply chain pressures by roughly 0.3 and 0.6 standard deviation after 2 years in China, Korea and the United States, respectively."
  - Weather shocks appear negligible for supply chain pressures in the Euro Area, Japan, and the United Kingdom.
  - SCPI components (shipping rates, air freight prices, PMI sub-components) can be influenced by oil prices and port congestion, possibly reducing the share of supply chain pressures explained by weather shocks.
- Supply chain pressures → inflation:
  - Clear upward pressure on headline inflation across all countries except Korea.
  - Quantitative estimates over a 24-month period (cumulative responses to a one-standard deviation SCPI shock):
    - China: about 1.3 percent
    - United Kingdom: about 0.7 percent
    - United States: about 1.5 percent
  - Smaller effects in the Euro Area and Japan; no adverse effect in Korea in baseline (attributed in part to the Asian financial crisis 1997-1999).
  - Core inflation responses follow a similar pattern (except for Japan) with considerably lower estimated coefficients (except for the United States).
- Weather shocks → inflation (direct):
  - Results statistically insignificant and ambiguous across countries.
  - Possible offsetting channels: higher temperatures may increase agricultural output and reduce energy demand (warmer winters reduce energy demand more than warmer summers raise it), potentially lowering inflation; but increased volatility in weather patterns can disrupt supply chains and raise inflation.
- Visuals and inference:
  - Impulse responses presented over 24-month horizons with 68 percent confidence intervals.
  - Estimates for Euro Area and China begin from December 2001 and January 2006, respectively, to December 2021 (figures noted).

### Robustness checks and subsample analysis
- Panel LP method (Jordà, 2005):
  - Panel LP results confirm supply chain pressures exert upward pressure on headline inflation; the headline inflation response rises sharply and declines rapidly after one year.
  - Temperature shocks in panel setting reduce inflation, aligning with SVAR baseline results.
  - Results robust after dropping COVID-19 period (2020-2021).
- Country SVAR excluding COVID-19 (2020-2021):
  - Weather shocks generally do not affect supply chain pressures when COVID-19 period is excluded.
  - Supply chain pressures raise inflation in most countries; the negative relationship in Korea is driven by the Asian financial crisis (1997-1999).
  - Weather shocks and inflation estimates similar to baseline; COVID-19 did not considerably alter baseline estimates.
- Pre- and post-crisis subsamples:
  - Two subsamples used: January 2000–December 2007 and January 2010–December 2019.
  - Key patterns:
    - Weather anomalies could raise supply chain pressures before 2008, particularly in the US; no impact after (excluding COVID-19).
    - Effects of supply chain pressures on inflation are statistically more significant after 2008 in China, Korea, the UK, and the US than before.
    - Weather anomalies could raise inflation before 2008 (particularly US), but generally have no impact or reduce inflation after 2008.
- Cyclical vs trend inflation (Hodrick-Prescott filter):
  - Decomposing inflation into cyclical component yields broadly similar results to baseline.
  - Difference noted: weather shocks could increase the cyclical component of inflation in the US for about three months.
  - Negative relationship between supply chain pressures and cyclical inflation in Korea is driven by the Asian financial crisis; positive relationship emerges when that period is removed.

### Appendix Figure A1 — Korea (description and relevance)
- Figure title: Supply Chain Pressures and Inflation: Korea.
- Presents the cumulative response of core inflation to supply chain pressures in dark blue line and 68 percent confidence intervals in light blue.
- The Asian financial crisis (1997-1999) and the COVID-19 pandemic (2020-2021) periods are eliminated.
- Relevant findings reiterated:
  - Weather anomalies could disrupt supply chains and subsequently lead to inflationary pressures.
  - Significant heterogeneity across countries.
  - Analysis uses monthly data for China, the Euro area, Japan, Korea, the United Kingdom, and the United States over 1997-2021 and implements a SVAR model.

### Unit root test (Appendix Table A1) — Augmented Dickey-Fuller p-values (C = constant, C+T = constant and trend; ***, **, * denote 1%, 5% and 10% significance levels; lag 1)
- Temperature:
  - China C 0.00***, C+T 0.00***
  - EA C 0.00***, C+T 0.00***
  - Japan C 0.00***, C+T 0.00***
  - Korea C 0.00***, C+T 0.00***
  - UK C 0.00***, C+T 0.00***
  - US C 0.00***, C+T 0.00***
- SCPI:
  - China C 0.00***, C+T 0.00***
  - EA C 0.00***, C+T 0.00***
  - Japan C 0.00***, C+T 0.00***
  - Korea C 0.00***, C+T 0.00***
  - UK C 0.00***, C+T 0.00***
  - US C 0.00***, C+T 0.00***
- Headline CPI:
  - China C 0.00***, C+T 0.09*
  - EA C 0.00***, C+T 0.36
  - Japan C 0.00***, C+T 0.00***
  - Korea C 0.01***, C+T 0.02**
  - UK C 0.01***, C+T 0.53
  - US C 0.00***, C+T 0.00***
- Core CPI:
  - China C 0.00***, C+T 0.27
  - EA C 0.01***, C+T 0.68
  - Japan C 0.00***, C+T 0.09*
  - Korea C 0.07*, C+T 0.22
  - UK C 0.06*, C+T 0.37
  - US C 0.01***, C+T 0.49

### Policy implications and recommendations
- Central banks:
  - Consider the persistent impact of weather anomalies on supply chains and inflation dynamics to prevent entrenching second-round effects and de-anchoring inflation expectations.
- Governments:
  - Invest in climate change adaptation to strengthen critical infrastructure and thereby minimize supply chain disruptions.

*IMF Working Paper — This Is Going to Hurt: Climate Change, Supply Chain Pressures and Inflation — Section IV and Executive Summary (content as provided).*

### Executive Summary ......................................................................................................

### Executive Summary

### Context and objective
- The world economy experienced a series of unprecedented shocks over the past three years, disrupting supply chains, causing a deep recession, and pushing inflation to the highest level since the 1970s.
- Inflation declined from a peak of 11.6 percent in 2022 to 5.3 percent in 2023, but long-term vulnerabilities persist.
- Climate change is identified as one of the most significant risks to the global economy and financial markets due to greater frequency and severity of extreme weather conditions.
- Objective: Investigate the impact of weather anomalies—measured by unseasonal deviation in temperature from the historical average—on global supply chain pressures and inflation dynamics.

### Data and methods
- Monthly data for six large and well-diversified economies: China, the Euro area, Japan, Korea, the United Kingdom, and the United States.
- Sample period: 1997-2021.
- Primary econometric approach: structural vector autoregressive (SVAR) model estimated for each country to trace contemporaneous effects of weather anomalies on supply chains and inflation.
- Robustness check: local projection (LP) method to trace dynamic effects of temperature anomalies over time.

### Mechanisms linking temperature shocks to inflation
- Three principal transmission channels highlighted:
  - Higher energy demand and prices leading to second-round effects on non-energy prices.
  - Lower individual productivity in climate-sensitive sectors altering inflation dynamics.
  - Supply chain disruptions causing productivity shortfalls, higher transportation and production costs, and consequently higher inflation.
- Conceptual emphasis that opposing effects on demand and supply can operate simultaneously.

### Key empirical findings
- Weather anomalies can contribute to supply chain disruptions and subsequently lead to inflationary pressures.
- Results are based on high-frequency data and are robust to alternative estimation methodologies.
- Significant heterogeneity across countries in the sample, attributed to differences in the severity of weather shocks and vulnerability to supply chain disruptions.
- The impact of weather shocks on supply chains and inflation dynamics is likely to become more pronounced with accelerating climate change that can have non-linear effects.
- The empirical results do not always show a strong positive link between weather shocks and supply chain disruptions; partial explanation is the use of aggregated supply-side disruption measures.
- Example of real-world disruption: a severe drought reduced water level considerably in the Panama Canal, disrupting the trade route connecting Asia and North America.
- Supporting empirical context from literature and data:
  - Global land temperature increase reached more than 1.1 degrees Celsius (°C) compared with the preindustrial average.
  - Projections: mean temperature may rise above the 1.5°C threshold in the near term and by as much as 4°C over the next century absent a global green transition.
  - Supply chain disruptions persisting a month or longer already occur every 3.7 years on average.

### Policy implications and recommendations
- Central banks:
  - Consider the persistent impact of weather anomalies on supply chains and inflation dynamics to prevent entrenching second-round effects and de-anchoring inflation expectations.
- Governments:
  - Invest in climate change adaptation to strengthen critical infrastructure and thereby minimize supply chain disruptions.

### Structure of the paper (as presented)
- Following the Executive Summary: I. Introduction; II. Literature Review; III. Data Overview; IV. Econometric Strategy; V. Empirical Results; VI. Conclusion.

*IMF Working Paper — This Is Going to Hurt: Climate Change, Supply Chain Pressures and Inflation — Executive Summary.*

### Section IV offers concluding remarks with policy implications.

### wpiea2024079-print-pdf - Section IV offers concluding remarks with policy implications

### Literature Review
- The paper situates itself within literature on climate change impacts on economic activity and financial markets, citing seminal and recent studies that find:
  - Higher temperatures reduce economic growth in developing countries (Gallup, Sachs, and Mellinger (1999); Nordhaus (2006); Dell, Jones, and Olken (2012); Burke, Hsiang, and Miguel (2015)).
  - Long-term impacts of weather anomalies are heterogeneous and nonlinear across countries (Acevedo et al., 2018; Burke and Tanutama, 2019; Kahn et al., 2021; Akyapi, Bellon, and Massetti, 2022).
  - Climate-related natural disasters reduce economic development, human capital accumulation, and worsen external balances (Loyaza et al., 2012; Noy, 2009; Raddatz, 2009; Skidmore and Toya, 2002; Cuaresma, 2010; Gassebner, Kesk, and The, 2006).
  - Climate vulnerability raises government borrowing costs and sovereign default probability (Cevik and Jalles, 2020; 2021; 2022).
  - Climate risks depress asset valuations and real estate prices exposed to sea level rise (Bansal, Kiku, and Ochoa, 2016; IMF, 2020; Bernstein, Gustasson, and Lewis, 2019).
  - Limited but growing evidence links climate change and extreme weather to consumer price inflation (Faccia, Parker, and Stracca, 2021; Kabundi, Mlachila, and Yao, 2022; Cevik and Jalles, 2023).
- Literature on supply chain pressures and inflation has expanded since COVID-19, showing supply bottlenecks raise inflation (Benigno et al., 2022; Di Giovanni et al., 2022; Finck and Tillmann, 2022; LaBelle and Santacreu, 2022; Kabaca and Tuzcuoglu, 2023; Andriantomanga, Bolhuis, and Hakobyan, 2023).

### Data Overview
- Sample and variables:
  - Monthly observations for six economies: China, the Euro area, Japan, Korea, the United Kingdom, and the United States, over the period 1997–2021.
  - Endogenous variables: weather anomalies, Supply Chain Pressures Index (SCPI), and consumer price inflation (headline or core CPI).
  - Inflation definition: 휋_{i,t} = (CPI_{i,t} / CPI_{i,t−12}) * 100 (year-on-year percent change); transformed via inverse hyperbolic sine: ln(휋_{i,t} + √(휋_{i,t}^2 + 1)).
  - SCPI: constructed by the Federal Reserve Bank of New York, measured in standard deviations from the sample average and normalized to have an average of zero.
  - Weather anomalies: monthly mean temperature deviation from historical average for that month during the reference period 1901-1996; for Euro area a GDP-weighted average of 19 countries is used.
  - Extreme weather events defined as anomalies large relative to historical variation.
- Stationarity:
  - Augmented Dickey-Fuller (1981) tests indicate all series are stationary after logarithmic transformation (Appendix Table A1).
- Descriptive statistics (Table 1: summary statistics reproduced exactly as presented):
  - Temperature (Mean):
    - China: 0.9592
    - EA: 1.2417
    - Japan: 1.2687
    - Korea: 1.1846
    - UK: 0.8603
    - US: 0.9203
  - Temperature (Variance):
    - China: 0.7228
    - EA: 1.3944
    - Japan: 0.7898
    - Korea: 1.3405
    - UK: 0.9623
    - US: 0.8936
  - Temperature (Minimum):
    - China: -1.7583
    - EA: -2.4578
    - Japan: -1.7947
    - Korea: -3.1941
    - UK: -3.7659
    - US: -1.8784
  - Temperature (Maximum):
    - China: 3.9747
    - EA: 4.0698
    - Japan: 3.9183
    - Korea: 4.3788
    - UK: 3.2510
    - US: 3.7734
  - Supply chain pressures (Mean):
    - China: -0.0216
    - EA: -0.0098
    - Japan: -0.0224
    - Korea: -0.0177
    - UK: -0.0122
    - US: -0.0192
  - Supply chain pressures (Variance):
    - China: 0.9381
    - EA: 0.9931
    - Japan: 0.9330
    - Korea: 0.9605
    - UK: 0.9827
    - US: 0.9523
  - Supply chain pressures (Minimum):
    - China: -1.7932
    - EA: -3.2800
    - Japan: -2.7560
    - Korea: -7.1594
    - UK: -4.1183
    - US: -1.8996
  - Supply chain pressures (Maximum):
    - China: 4.3582
    - EA: 3.5000
    - Japan: 3.6264
    - Korea: 3.1049
    - UK: 3.3160
    - US: 4.0856
  - Headline Inflation (Mean):
    - China: 1.0826
    - EA: 1.1642
    - Japan: 0.0618
    - Korea: 1.4928
    - UK: 1.3602
    - US: 1.4024
  - Headline Inflation (Variance):
    - China: 1.1059
    - EA: 0.3301
    - Japan: 0.5586
    - Korea: 0.3863
    - UK: 0.1640
    - US: 0.4095
  - Headline Inflation (Minimum):
    - China: -1.5297
    - EA: -0.5688
    - Japan: -1.6528
    - Korea: -0.4141
    - UK: 0.1985
    - US: -1.4863
  - Headline Inflation (Maximum):
    - China: 2.8716
    - EA: 2.3124
    - Japan: 2.0192
    - Korea: 2.9522
    - UK: 2.2814
    - US: 2.6493
  - Core Inflation (as presented in table):
    - Mean row (compact format in source): 0.9568 1.1459 -0.1151 1.3880 1.2145 1.4493
    - Variance row: 0.3686 0.0947 0.4221 0.3173 0.1284 0.0611
    - Minimum row: -1.2995 0.3900 -1.2507 -0.6502 -0.1288 0.6184
    - Maximum row: 1.6187 1.7191 1.6746 2.5116 2.0479 2.4044

### Econometric Strategy
- Objective:
  - Estimate the impact of temperature deviations (weather anomalies) on supply chain disruptions and inflation using monthly data for six economies (1997–2021).
- Model and identification:
  - Three-variable SVAR model with Cholesky decomposition to orthogonalize reduced-form residuals and impose causal ordering: weather anomalies → supply chain pressures → inflation.
  - Temperature anomalies assumed strictly exogenous for short-run effects.
  - AB-model SVAR specification estimated for each country separately, with matrices A (upper triangular) and B (diagonal), following Lütkepohl (2005).
  - Reduced-form representation and Φ_j coefficients are upper triangular, allowing direct restrictions on reduced-form coefficients.
  - Optimal lag length determined by SIC criteria: 1 to 2 lags depending on country.
  - Robustness of inference: 2,000 bootstrapped error bands for impulse response coefficients and 68 percent confidence intervals.

### Empirical Results (baseline)
- Weather shocks → supply chain pressures:
  - Larger temperature deviations tend to exert upward pressure on supply chain pressures, though not statistically significant for all countries.
  - Quantitative finding (verbatim): "a 1°C increase in temperature with respect to historical average increases supply chain pressures by roughly 0.3 and 0.6 standard deviation after 2 years in China, Korea and the United States, respectively."
  - Weather shocks appear negligible for supply chain pressures in the Euro Area, Japan, and the United Kingdom.
  - Authors note SCPI components (shipping rates, air freight prices, PMI sub-components) can be influenced by oil prices and port congestion, possibly reducing the share of supply chain pressures explained by weather shocks.
- Supply chain pressures → inflation:
  - Clear upward pressure on headline inflation across all countries except Korea.
  - Quantitative estimates over a 24-month period (cumulative responses to a one-standard deviation SCPI shock):
    - China: about 1.3 percent
    - United Kingdom: about 0.7 percent
    - United States: about 1.5 percent
  - Smaller effects in the Euro Area and Japan; no adverse effect in Korea in baseline (attributed in part to the Asian financial crisis 1997-1999).
  - Core inflation responses follow a similar pattern (except for Japan) with considerably lower estimated coefficients (except for the United States).
- Weather shocks → inflation (direct):
  - Results statistically insignificant and ambiguous across countries.
  - Possible offsetting channels: higher temperatures may increase agricultural output and reduce energy demand (warmer winters reduce energy demand more than warmer summers raise it), potentially lowering inflation; but increased volatility in weather patterns can disrupt supply chains and raise inflation.
- Visuals and inference:
  - Impulse responses presented over 24-month horizons with 68 percent confidence intervals.
  - Estimates for Euro Area and China begin from December 2001 and January 2006, respectively, to December 2021 (figures noted).

### Robustness Checks
- Panel LP method (Jordà, 2005):
  - Panel LP results confirm supply chain pressures exert upward pressure on headline inflation; the headline inflation response rises sharply and declines rapidly after one year.
  - Temperature shocks in panel setting reduce inflation, aligning with SVAR baseline results.
  - Results robust after dropping COVID-19 period (2020-2021).
- Country SVAR excluding COVID-19 (2020-2021):
  - Weather shocks generally do not affect supply chain pressures when COVID-19 period is excluded.
  - Supply chain pressures raise inflation in most countries; the negative relationship in Korea is driven by the Asian financial crisis (1997-1999).
  - Weather shocks and inflation estimates similar to baseline; COVID-19 did not considerably alter baseline estimates.
- Pre- and post-crisis subsamples:
  - Two subsamples used: January 2000–December 2007 and January 2010–December 2019 (to avoid Asian/global financial crises and COVID-19).
  - Key patterns across periods:
    - Weather anomalies could raise supply chain pressures before 2008, particularly in the US; no impact after (excluding COVID-19).
    - Effects of supply chain pressures on inflation are statistically more significant after 2008 in China, Korea, the UK, and the US than before.
    - Weather anomalies could raise inflation before 2008 (particularly US), but generally have no impact or reduce inflation after 2008.
- Cyclical vs trend inflation (Hodrick-Prescott filter):
  - Decomposing inflation into cyclical component yields broadly similar results to baseline.
  - Difference noted: weather shocks could increase the cyclical component of inflation in the US for about three months.
  - Negative relationship between supply chain pressures and cyclical inflation in Korea is driven by the Asian financial crisis; positive relationship emerges when that period is removed.

*Source: IMF Working Paper "This Is Going to Hurt: Climate Change, Supply Chain Pressures and Inflation" (section contents as provided).*

### Appendix Figure A1.

### Appendix Figure A1.

### Figure description
- Title: Supply Chain Pressures and Inflation: Korea
- The figure presents the cumulative response of core inflation to supply chain pressures in dark blue line and 68 percent confidence intervals in light blue.
- The Asian financial crisis (1997-1999) and the COVID-19 pandemic (2020-2021) periods are eliminated.

### Findings from the paper relevant to this figure
- Weather anomalies could disrupt supply chains and subsequently lead to inflationary pressures.
- Results show significant heterogeneity across countries, attributed to differences in the severity of weather shocks and vulnerability to supply chain disruptions.
- The impact of weather shocks on supply chains and inflation dynamics is likely to become more pronounced with accelerating climate change that can have non-linear effects.
- The paper’s analysis uses monthly data covering China, the Euro area, Japan, Korea, the United Kingdom, and the United States over 1997-2021 and implements a SVAR model to trace contemporaneous effects of weather anomalies on supply chains and inflation.

### Policy implications (as stated in the source)
- Central bankers should consider the persistent impact of weather anomalies on supply chains and inflation dynamics to prevent entrenching second-round effects and de-anchoring inflation expectations.
- Governments can invest more for climate change adaptation to strengthen critical infrastructure and thereby minimize supply chain disruptions.

### Unit root test (Appendix Table A1) — p-values reported (Augmented Dickey-Fuller, C = constant, C+T = constant and trend; ***, **, * denote 1%, 5% and 10% significance levels; lag 1)
- Temperature:
  - China C 0.00***, C+T 0.00***
  - EA C 0.00***, C+T 0.00***
  - Japan C 0.00***, C+T 0.00***
  - Korea C 0.00***, C+T 0.00***
  - UK C 0.00***, C+T 0.00***
  - US C 0.00***, C+T 0.00***
- SCPI:
  - China C 0.00***, C+T 0.00***
  - EA C 0.00***, C+T 0.00***
  - Japan C 0.00***, C+T 0.00***
  - Korea C 0.00***, C+T 0.00***
  - UK C 0.00***, C+T 0.00***
  - US C 0.00***, C+T 0.00***
- Headline CPI:
  - China C 0.00***, C+T 0.09*
  - EA C 0.00***, C+T 0.36
  - Japan C 0.00***, C+T 0.00***
  - Korea C 0.01***, C+T 0.02**
  - UK C 0.01***, C+T 0.53
  - US C 0.00***, C+T 0.00***
- Core CPI:
  - China C 0.00***, C+T 0.27
  - EA C 0.01***, C+T 0.68
  - Japan C 0.00***, C+T 0.09*
  - Korea C 0.07*, C+T 0.22
  - UK C 0.06*, C+T 0.37
  - US C 0.01***, C+T 0.49

*Italic: Source — Appendix Figure A1 and accompanying text from the IMF Working Paper "This Is Going to Hurt: Climate Change, Supply Chain Pressures and Inflation."*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024079-print-pdf.pdf_