## 1.1 NETWORK EXPOSURES: DESCRIPTIVE EVIDENCE

## Source details

**Canonical URL:** [1.1 NETWORK EXPOSURES: DESCRIPTIVE EVIDENCE](https://www.imf.org/-/media/files/publications/weo/2024/october/english/ch1onlineannexes.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/weo/2024/october/english/ch1onlineannexes.pdf.md)
- [Structured JSON version](/-/media/files/publications/weo/2024/october/english/ch1onlineannexes.pdf.json)

---

### Role of production networks and Leontief exposures
- Sector i’s exposure to a metals (M) productivity shock is given by the element (I−Ω)−1_{iM} of the Leontief inverse.
- The matrix Ω element iM is the ratio between expenditure on input M by sector i and total gross output of sector i.
- (I−Ω)−1_{iM} captures direct and indirect effects of a metals productivity decline on sector i’s total intermediate input costs (including inputs of suppliers and suppliers’ suppliers).
- Given large import shares for metals and oil in many countries, Figures 1.SF.4 and 1.SF.5 in the source also consider intermediate input imports of metals and oil by domestic sectors.
- Empirical example: results in Figure 1.SF.4 use BEA input-output matrices for the US in 2018; sectoral exposures are calculated using (I−Ω)−1_{iM}.

### Mapping metals (or oil) shocks to aggregate inflation measures
- GDP deflator (DGDP) and consumer price index (CPI) definitions used:
  - log DGDP_t = ∑ v̄_k · log p_{k,t}
  - log CPI_t = ∑ b̄_k · log p_{k,t}
  - v̄_k and b̄_k are value-added and final consumption expenditure shares of sector k, respectively.
- Price of sector k depends on metals productivity through Leontief inverse element (I−Ω)^{-1}_{kM}.
- If the only shock is to metals productivity (log Z_M), changes are:
  - d log DGDP_t = ∑ v̄_k · ∂ log p_{k,t} / ∂ log Z_M = ∑ v̄_k (I−Ω)^{-1}_{kM}
  - d log CPI_t = ∑ b̄_k · ∂ log p_{k,t} / ∂ log Z_M = ∑ b̄_k (I−Ω)^{-1}_{kM}
- These expressions capture inflationary pressure from an exogenous metals price shock; the overall CPI effect depends on the monetary policy rule (nominal variables pinned down by money supply targeting a given nominal GDP level in the described exercise).

### Intuition and illustrative country scenarios (spending vs exposure channels)
- Two mechanisms for how a metals price shock affects country inflation:
  1. Spending-share channel: different households spend different shares of income on goods.
     - Example figures: Germans spend 3.29 percent on cars vs French 0.84 percent on cars.
     - With similar metals exposure of cars, higher consumption share increases inflationary impact (Germany > France in this example).
  2. Exposure channel: the same consumed good can have different metals exposure across countries.
     - Example exposures: car metals exposure = 9.65 percent for Germany vs 28.06 percent for China.
     - If consumption shares similar (Germans 3.29 percent vs Chinese 2.83 percent), higher metals exposure (China) produces greater inflationary impact.
- Specific illustrative figures from source:
  - Germans: 3.29 percent consumption share on cars; car metals exposure 9.65 percent.
  - French: 0.84 percent consumption share on cars; car metals exposure 9.94 percent.
  - Chinese: 2.83 percent consumption share on cars; car metals exposure 28.06 percent.
  - Observation: German car production has more than three times spending share going to labor costs and R&D compared to a car produced in China (cited Liu, 2018).

### Data sources referenced for network exposure analysis
- Commodity prices and indices: monthly January 1996 to December 2020.
- CPI (headline and core): global inflation database assembled by Ha et al. (2023).
- Controls:
  - Global economic activity index from Baumeister et al. (2022).
  - US 1 year treasury bill yield from FRED.
  - Bilateral exchange rates from BIS.
  - Excess bond premium (EBP) from Gilchrist and Zakrajšek (2012) for the USA.
- Instrumental shock series: Baumeister, Ohnsorge, and Verduzco-Bustos (2024) for copper; Baumeister and Hamilton (2019) for oil.

---

### Theoretical framework for cross-country regressions (model summary)

### Model environment and agents
- Model basis: Silva (2023) — a small open economy model with production networks.
- Agents and sets:
  - N domestically produced goods (i ∈ N)
  - M imported goods (m ∈ M)
  - F factors of production (f ∈ F)
- Representative household:
  - Utility U(C_D, C_M) with C_D={C_i}_{i∈N}, C_M={C_m}_{m∈M}.
  - Price index P = P(P_D, P_M); first-order approximation: P̂ = b̄_D^T P̂_D + b̄_M^T P̂_M, where b̄_i = P_i C_i / E and E = P_D^T C_D + P_M^T C_M = PC.
  - Budget constraint: PC + T = ∑_{f∈F} W_f L_f + ∑_{i∈N} Π_i (T exogenous net transfer).
- Firms:
  - Representative firm in sector i: Q_i = Z_i F_i({L_if}_{f∈F},{M_ij}_{j∈N},{M_im}_{m∈M}).
  - Marginal cost MC_i = MC_i(Z_i, P_D, P_M, W). With CRS and perfect competition, P_i Q_i = ∑ W_f L_if + ∑ P_j M_ij + ∑ P_m M_im.
- Market clearing: Q_i = C_i + X_i + ∑_{j∈N} M_ji for each i ∈ N; X_i exogenous.
- Nominal anchor / cash-in-advance assumption: PC ≤ M_u = E (money supply M_u dictates nominal spending E).
- Equilibrium characterized by households and firms optimizing given (W, P_D, P_M) and market clearing with binding cash-in-advance (PC = M_u = E).

### Log-linear changes in aggregate price index — key proposition
- Proposition summary (Silva, 2023): for perturbations (Ẑ, Ŵ, P̂_M), up to first order:
  - P̂ = −(λ_T̄ − λ_T̃) Ẑ + (Γ_T̃ − Γ_T̄) Ŵ + (b̄_M^T + b̄_M^T̃) P̂_M   (Equation (1) in source)
  - where λ_T̃ = x̄^T ⊙ D̃ ; Γ_T̃ = x̄^T ⊙ D̃ A ; b_M^T̃ = b̄_D^T ⊙ D̃ ⊙ ; D̃ = (I − Ω_D)^{-1}
- In the absence of productivity and wage shocks (Ŵ_F = 0_N and Ẑ_N = 0_N), import price shocks imply:
  - P̂ = (b̄_M^T + b̄_D^T ⊙ D̃ ⊙) P̂_M   (Equation (2) in source)
  - Interpretation: inflation depends on direct import consumption share (b̄_M) and indirect consumption via imports embedded in domestic production (b̄_D^T ⊙ D̃ ⊙), emphasizing metals/oil.
- Country-level metals/oil exposure measure:
  - (b_M^T + b_D^T ⊙ D̃ ⊙) — where b_M is consumption expenditure share of imported metals (oil), b_D the vector of domestic consumption expenditure shares, D̃ the domestic Leontief inverse, and ⊙ the vector of sectoral shares of imported metals (oil) in gross output.

---

### Panel local projections — empirical specification, identification, and impulse responses

### Empirical local projection specification
- Regression (Jordà, 2005):
  - ln CPI_{i,t+h} − ln CPI_{i,t−1} = α_i^h + β_1^h p_{t}^c + β_2^h p_t^c · (z_i − z̄) + ∑_{l=0}^L φ_{xl}^h X_{t−l}^i + ε_{i,t+h}
  - CPI_{i,t+h} is headline or core CPI for country i at t+h; α_i are country fixed effects.
  - p_t^c is log of copper (or oil) price at time t, deflated using the trend of US CPI derived from the Hodrick-Prescott filter.
  - Controls X_{t−l}^i include global economic activity index, US treasury bill rates, bilateral exchange rates, excess bond premium, L=12 lags of log change of inflation, and lags of log copper (oil) prices. Contemporaneous and lagged log food prices and oil (copper) prices included as appropriate.
  - Interaction z_i is country i’s network exposure to metals (oil): z_i = b_m + b_D^T ⊙ D̃ ⊙_m (as defined in model).
  - The term β_1^h p_t^c + β_2^h p_t^c · (z_i − z̄) captures the inflationary impact of copper (oil) price changes, conditional on network exposure.

### Identification via instrumental variables
- Use commodity price shocks identified in literature as instruments for copper (oil) prices.
  - Copper instrument: Baumeister, Ohnsorge, and Verduzco-Bustos (2024).
  - Oil instrument: Baumeister and Hamilton (2019).
- Online Annex Figure1.SF.1 shows impulse responses of real copper and real oil prices to a one standard deviation supply shock (estimated with local projection, 12 lags; sample: 1996:M2 to 2019:M12; shaded 90 percent confidence bands using Newey-West standard errors with h lags).

### Empirical sample and illustrative impulse responses
- Regression estimated on monthly data from 1996:m2 to 2019:m12 for a balanced panel with 39 countries.
- Online Annex Figure1.SF.2 displays impulse responses to a 1 percent increase in copper prices (top) and oil prices (bottom), evaluated at metal (oil) exposure at the 90th and 10th percentiles of the sample.
  - “Average” is the average inflationary effect from running the LP without the interaction term.
  - Figures present responses for Headline CPI and Core CPI with low exposure, high exposure, and average lines; shaded 90 percent confidence bands. Sample: 1996:M2 to 2019:M12.

---

### 1. Headline CPI

### Methodology and data sources (summary)
- To correct for potential endogeneity in copper prices, the analysis uses the estimated copper supply shock from Baumeister, Ohnsorge, and Verduzco-Bustos (2024).
- For oil supply shocks, the analysis uses Baumeister and Hamilton (2019).
- Input-output data:
  - BEA for the US with 71 sectors.
  - OECD for cross-country comparisons with 45 sectors.
- Sectoral classifications correspond to the lists used in the empirical analysis and reflect the year 2018.
- Sample period for the impulse response analysis: 1996:M2 to 2019:M12.

### Network exposure and impulse response analysis (highlight)
- Network exposure to metals and oil measured for 39 economies in the sample and depicted for the year 2018 (data labels use ISO country codes).
- Online Annex Figure 1.SF.3 shows countries’ network exposure to metals and oil (percent).
- Online Annex Figure 1.SF.4 reports impulse responses to 1 percent increases in commodity prices:
  - Copper = impulse responses to copper supply shock.
  - Oil = impulse responses to oil supply shock.
  - “High exposure” and “Low exposure” indicate the 90th and 10th percentiles of exposure to metals (for copper shock) and oil (for oil shock).
  - Shaded areas in the figure are 90 percent confidence bands.
  - Figures present additional impulse responses examining the importance of production networks in the transmission of metals (oil) shocks, including versions using metals (oil) consumption share as a non-network-adjusted exposure measure.

*Source: 1.1 NETWORK EXPOSURES: DESCRIPTIVE EVIDENCE, ch1onlineannexes (ONLINE ANNEX 1.1) — World Economic Outlook, International Monetary Fund | October 2024.*

### 1.1 NETWORK EXPOSURES: DESCRIPTIVE EVIDENCE

### 1.1 NETWORK EXPOSURES: DESCRIPTIVE EVIDENCE

### Role of production networks and Leontief exposures
- Sector i’s exposure to a metals (M) productivity shock is given by the element (I−Ω)−1_{iM} of the Leontief inverse.
- The matrix Ω element iM is the ratio between expenditure on input M by sector i and total gross output of sector i.
- (I−Ω)−1_{iM} captures direct and indirect effects of a metals productivity decline on sector i’s total intermediate input costs (including inputs of suppliers and suppliers’ suppliers).
- Given large import shares for metals and oil in many countries, Figures 1.SF.4 and 1.SF.5 in the source also consider intermediate input imports of metals and oil by domestic sectors.
- Empirical example: results in Figure 1.SF.4 use BEA input-output matrices for the US in 2018; sectoral exposures are calculated using (I−Ω)−1_{iM}.

### Mapping metals (or oil) shocks to aggregate inflation measures
- GDP deflator (DGDP) and consumer price index (CPI) in the model are defined as:
  - log DGDP_t = ∑ v̄_k · log p_{k,t}
  - log CPI_t = ∑ b̄_k · log p_{k,t}
  - where v̄_k and b̄_k are value-added and final consumption expenditure shares of sector k, respectively.
- Price of sector k depends on metals productivity through Leontief inverse element (I−Ω)^{-1}_{kM}.
- If the only shock is to metals productivity (log Z_M), changes are:
  - d log DGDP_t = ∑ v̄_k · ∂ log p_{k,t} / ∂ log Z_M = ∑ v̄_k (I−Ω)^{-1}_{kM}
  - d log CPI_t = ∑ b̄_k · ∂ log p_{k,t} / ∂ log Z_M = ∑ b̄_k (I−Ω)^{-1}_{kM}
- These expressions capture inflationary pressure from an exogenous metals price shock; the overall CPI effect depends on the monetary policy rule (nominal variables pinned down by money supply targeting a given nominal GDP level in the described exercise).

### Theoretical framework for cross-country regressions (model summary)
- Model basis: Silva (2023) — a small open economy model with production networks.
- Agents and sets:
  - N domestically produced goods (i ∈ N)
  - M imported goods (m ∈ M)
  - F factors of production (f ∈ F)
- Representative household:
  - Utility U(C_D, C_M) with C_D={C_i}_{i∈N}, C_M={C_m}_{m∈M}.
  - Price index P = P(P_D, P_M); first-order approximation: P̂ = b̄_D^T P̂_D + b̄_M^T P̂_M, where b̄_i = P_i C_i / E and E = P_D^T C_D + P_M^T C_M = PC.
  - Budget constraint: PC + T = ∑_{f∈F} W_f L_f + ∑_{i∈N} Π_i (T exogenous net transfer).
- Firms:
  - Representative firm in sector i: Q_i = Z_i F_i({L_if}_{f∈F},{M_ij}_{j∈N},{M_im}_{m∈M}).
  - Marginal cost MC_i = MC_i(Z_i, P_D, P_M, W). With CRS and perfect competition, P_i Q_i = ∑ W_f L_if + ∑ P_j M_ij + ∑ P_m M_im.
- Market clearing: Q_i = C_i + X_i + ∑_{j∈N} M_ji for each i ∈ N; X_i exogenous.
- Nominal anchor / cash-in-advance assumption: PC ≤ M_u = E (money supply M_u dictates nominal spending E).
- Equilibrium characterized by households and firms optimizing given (W, P_D, P_M) and market clearing with binding cash-in-advance (PC = M_u = E).

### Log-linear changes in aggregate price index — key proposition
- Proposition summary (Silva, 2023): for perturbations (Ẑ, Ŵ, P̂_M), up to first order:
  - P̂ = −(λ_T̄ − λ_T̃) Ẑ + (Γ_T̃ − Γ_T̄) Ŵ + (b̄_M^T + b̄_M^T̃) P̂_M   (Equation (1) in source)
  - where λ_T̃ = x̄^T ⊙ D̃ ; Γ_T̃ = x̄^T ⊙ D̃ A ; b_M^T̃ = b̄_D^T ⊙ D̃ ⊙ ; D̃ = (I − Ω_D)^{-1}
- In the absence of productivity and wage shocks (Ŵ_F = 0_N and Ẑ_N = 0_N), import price shocks imply:
  - P̂ = (b̄_M^T + b̄_D^T ⊙ D̃ ⊙) P̂_M   (Equation (2) in source)
  - Interpretation: inflation depends on direct import consumption share (b̄_M) and indirect consumption via imports embedded in domestic production (b̄_D^T ⊙ D̃ ⊙), emphasizing metals/oil.
- Country-level metals/oil exposure measure: (b_M^T + b_D^T ⊙ D̃ ⊙) — where b_M is consumption expenditure share of imported metals (oil), b_D the vector of domestic consumption expenditure shares, D̃ the domestic Leontief inverse, and ⊙ the vector of sectoral shares of imported metals (oil) in gross output.

### Intuition and illustrative country scenarios (spending vs exposure channels)
- Two mechanisms for how a metals price shock affects country inflation:
  1. Spending-share channel: different households spend different shares of income on goods (example numbers provided).
     - Example: Germans spend 3.29% on cars vs French 0.84% on cars.
     - Consequence: with similar metals exposure of cars, higher consumption share increases inflationary impact (Germany > France in this example).
  2. Exposure channel: the same consumed good can have different metals exposure across countries.
     - Example exposures: car metals exposure = 9.65% for Germany vs 28.06% for China.
     - If consumption shares similar (Germans 3.29% vs Chinese 2.83%), higher metals exposure (China) produces greater inflationary impact.
- Specific illustrative figures from source:
  - Germans: 3.29% consumption share on cars; car metals exposure 9.65%.
  - French: 0.84% consumption share on cars; car metals exposure 9.94%.
  - Chinese: 2.83% consumption share on cars; car metals exposure 28.06%.
  - Observations: German car production has more than three times spending share going to labor costs and R&D compared to a car produced in China (cited Liu, 2018).

### Panel local projections — empirical specification and identification
- Empirical LP regression (Jordà, 2005) used to gauge inflation effects of primary metals prices:
  - ln CPI_{i,t+h} − ln CPI_{i,t−1} = α_i^h + β_1^h p_{t}^c + β_2^h p_t^c · (z_i − z̄) + ∑_{l=0}^L φ_{xl}^h X_{t−l}^i + ε_{i,t+h}
  - CPI_{i,t+h} is headline or core CPI for country i at t+h; α_i are country fixed effects.
  - p_t^c is log of copper (or oil) price at time t, deflated using the trend of US CPI derived from the Hodrick-Prescott filter.
  - Controls X_{t−l}^i include global economic activity index, US treasury bill rates, bilateral exchange rates, excess bond premium, L=12 lags of log change of inflation, and lags of log copper (oil) prices. Contemporaneous and lagged log food prices and oil (copper) prices included as appropriate.
  - Interaction z_i is country i’s network exposure to metals (oil): z_i = b_m + b_D^T ⊙ D̃ ⊙_m (as defined in model).
  - The term β_1^h p_t^c + β_2^h p_t^c · (z_i − z̄) captures the inflationary impact of copper (oil) price changes, conditional on network exposure.
- Identification via instrumental variables:
  - Use commodity price shocks identified in literature as instruments for copper (oil) prices.
  - For copper: Baumeister, Ohnsorge, and Verduzco-Bustos (2024).
  - For oil: Baumeister and Hamilton (2019).
  - [Online Annex Figure1.SF.1] shows impulse responses of real copper and real oil prices to a one standard deviation supply shock (estimated with local projection, 12 lags; sample: 1996:M2 to 2019:M12; shaded 90 percent confidence bands using Newey-West standard errors with h lags).

### Empirical sample, estimation and illustrative impulse responses
- Regression estimated on monthly data from 1996:m2 to 2019:m12 for a balanced panel with 39 countries.
- [Online Annex Figure1.SF.2] displays impulse responses to a 1 percent increase in copper prices (top) and oil prices (bottom), evaluated at metal (oil) exposure at the 90th and 10th percentiles of the sample.
  - “Average” is the average inflationary effect from running the LP without the interaction term.
  - The figures present responses for Headline CPI and Core CPI with low exposure, high exposure, and average lines; shaded 90 percent confidence bands. Sample: 1996:M2 to 2019:M12.

### Data sources
- Commodity prices and indices: monthly January 1996 to December 2020.
- CPI (headline and core): global inflation database assembled by Ha et al. (2023).
- Controls:
  - Global economic activity index from Baumeister et al. (2022).
  - US 1 year treasury bill yield from FRED.
  - Bilateral exchange rates from BIS.
  - Excess bond premium (EBP) from Gilchrist and Zakrajšek (2012) for the USA.
- Instrumental shock series: Baumeister, Ohnsorge, and Verduzco-Bustos (2024) for copper; Baumeister and Hamilton (2019) for oil.

*Source: 1.1 NETWORK EXPOSURES: DESCRIPTIVE EVIDENCE, ch1onlineannexes (ONLINE ANNEX 1.1) — World Economic Outlook, International Monetary Fund | October 2024.*

### 1. Headline CPI

### 1. Headline CPI

### Methodology and Data Sources
- To correct for potential endogeneity in copper prices, the analysis uses the estimated copper supply shock from Baumeister, Ohnsorge, and Verduzco-Bustos (2024).
- For oil supply shocks, the analysis uses Baumeister and Hamilton (2019).
- Input-output data are collected from:
  - BEA for the US with 71 sectors.
  - OECD for cross-country comparisons with 45 sectors.
- Sectoral classifications correspond to the lists used in the empirical analysis and reflect the year 2018.
- Sample period for the impulse response analysis: 1996:M2 to 2019:M12.

### BEA and OECD Sectoral Classifications (inventory)
- BEA sectoral classification highlights (selected sectors as listed):
  - Farms; Forestry, fishing, and related activities; Oil and gas extraction; Mining, except oil and gas; Support activities for mining; Utilities; Construction; Wood products; Nonmetallic mineral products; Primary metals; Fabricated metal products; Machinery; Computer and electronic products; Electrical equipment, appliances, and components; Motor vehicles, bodies and trailers, and parts; Other transportation equipment; Furniture and related products; Miscellaneous manufacturing; Food and beverage and tobacco products; Textile mills and textile product mills; Apparel and leather and allied products; Paper products; Printing and related support activities; Petroleum and coal products; Chemical products; Plastics and rubber products; Wholesale trade; Motor vehicle and parts dealers; Food and beverage stores; General merchandise stores; Other retail; Air transportation; Rail transportation; Truck transportation; Transit and ground passenger transportation; Pipeline transportation; Other transportation and support activities; Warehousing and storage; Publishing industries, except internet (includes software); Motion picture and sound recording industries; Broadcasting and telecommunications; Data processing, internet publishing, and other information services; Federal Reserve banks, credit intermediation, and related activities; Securities, commodity contracts, and investments; Insurance carriers and related activities; Funds, trusts, and other financial vehicles; Housing; Other real estate; Rental and leasing services and lessors of intangible assets; Legal services; Computer systems design and related services; Miscellaneous professional, scientific, and technical services; Management of companies and enterprises; Administrative and support services; Waste management and remediation services; Educational services; Ambulatory health care services; Hospitals; Nursing and residential care facilities; Social assistance; Performing arts, spectator sports, museums, and related activities; Amusements, gambling, and recreation industries; Accommodation; Food services and drinking places; Other services, except government.
- OECD sectoral classification highlights (selected sectors as listed):
  - Agriculture, hunting, forestry; Fishing and aquaculture; Mining and quarrying, energy producing products; Mining and quarrying, non-energy producing products; Mining support service activities; Food products, beverages and tobacco; Textiles, textile products, leather and footwear; Wood and products of wood and cork; Paper products and printing; Coke and refined petroleum products; Chemical and chemical products; Rubber and plastics products; Other non-metallic mineral products; Basic metals; Fabricated metal products; Computer, electronic and optical equipment; Electrical equipment; Machinery and equipment, nec; Motor vehicles, trailers, and semi-trailers; Other transport equipment; Manufacturing nec; repair and installation of machinery and equipment; Electricity, gas, steam and air conditioning supply; Construction; Wholesale and retail trade; repair of motor vehicles; Land transport and transport via pipelines; Water transport; Air transport; Warehousing and support activities for transportation; Postal and courier activities; Accommodation and food service activities; Publishing, audiovisual, and broadcasting activities; Telecommunications; IT and other information services; Financial and insurance activities; Real estate activities; Professional, scientific, and technical activities; Administrative and support services; Education; Human health and social work activities; Arts, entertainment and recreation; Other service activities; Water supply; sewerage, waste management and remediation activities; Pharmaceuticals, medicinal chemical and botanical products; Public administration and defence; compulsory social security.

### Network Exposure and Impulse Response Analysis
- Network exposure to metals and oil is measured for 39 economies in the sample and depicted for the year 2018 (data labels use ISO country codes).
- Online Annex Figure 1.SF.3 shows countries’ network exposure to metals and oil (percent).
- Online Annex Figure 1.SF.4 reports impulse responses to 1 percent increases in commodity prices:
  - Copper = impulse responses to copper supply shock.
  - Oil = impulse responses to oil supply shock.
  - “High exposure” and “Low exposure” indicate the 90th and 10th percentiles of exposure to metals (for copper shock) and oil (for oil shock).
  - Shaded areas in the figure are 90 percent confidence bands.
  - The figures present additional impulse responses examining the importance of production networks in the transmission of metals (oil) shocks, including versions using metals (oil) consumption share as a non-network-adjusted exposure measure.

### References Cited in the Annex (selected)
- Baumeister, C., and J.D. Hamilton. 2019. “Structural Interpretation of Vector Autoregressions with Incomplete Identification: Revisiting the Role of Oil Supply and Demand Shocks.” American Economic Review, 109(5): 1873-1910.
- Baumeister, C., F. Ohnsorge, and G. Verduzco-Bustos. 2024. “Evaluating the Role of Demand and Supply Shocks in Copper and Aluminum Price Fluctuations.” Unpublished Manuscript.
- Silva, Alvaro. 2023. “Inflation in disaggregated small open economies.” Mimeo, The University of Maryland.
- Miranda-Pinto, Jorge, Andrea Pescatori, Martin Stuermer, and Xueliang Wang. 2024. “Beyond Energy: The Inflationary Effects of Metal Price Shocks in Production Networks.” IMF Working Paper 24/215, International Monetary Fund, Washington DC.

*Source: ch1onlineannexes - 1. Headline CPI (PDF chapter/section).*

---


_Source: https://www.imf.org/-/media/files/publications/weo/2024/october/english/ch1onlineannexes.pdf_