## Appendix B: Structural Gravity Estimates

## Source details

**Canonical URL:** [Appendix B: Structural Gravity Estimates](https://www.imf.org/-/media/files/publications/wp/2026/english/wpiea2026125-source-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2026/english/wpiea2026125-source-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2026/english/wpiea2026125-source-pdf.pdf.json)

---

### I. Purpose and model overview
- Examines potential long-term economic effects of recent US tariff increases (and associated retaliatory measures) on CAPDR (Costa Rica, Dominican Republic, Guatemala, Honduras, Nicaragua, Panama and El Salvador) using a quantitative multi-region multi-industry model.
- Model features:
  - Eaton and Kortum (2002) static Ricardian comparative advantage framework with input-output linkages and trade in intermediate and final goods.
  - Global aggregation: 78 regions, each with 13 industries.
  - Production distinguishes gross output and value added and includes capital and labor as variable primary factors.
  - Representative household utility: u(C_s^R, N_s) = C_s^R − ν/(1+ν) N_s^(1+ν)/ν with labor supply elasticity ν.

### II. Recent US tariff events (chronology and headline rates)
- February 2025: United States announced tariffs of 25 percent on most goods imported from Canada and Mexico.
- Through 2025: United States widened tariff coverage to include most countries with higher rates for varying periods on Brazil, China, the European Union and India, among others.
- By end-2025: United States imposed additional statutory tariff rates of 18 percent on Nicaragua, 15 percent on Costa Rica, and 10 percent on the remaining members of CAPDR.
- November 13, 2025: Framework agreements with Guatemala and El Salvador to remove tariffs on selected textiles & apparel and some agricultural and mining products upon ratification in 2026.
- February 20, 2026: United States Supreme Court overturned many 2025 tariffs; President invoked alternative legal instruments to introduce a uniform 10 percent tariff (with exemptions for CAFTA-DR‑compliant textiles & apparel and USMCA‑compliant products).
- May 7, 2026: United States Court of International Trade declared these tariffs unlawful but left them in place while appeals are pending.

### III. Stylized CAPDR trade facts and GVC exposure
- Trade openness and exposure:
  - Honduras: total trade nearly 100 percent of GDP.
  - Guatemala: total trade around 40 percent of GDP.
  - United States: ~25 percent of GDP (total trade) for US; Canada ~65 percent, Mexico ~90 percent (comparative totals cited).
  - On average for CAPDR members: United States accounts for around 40 percent of exports and around 37 percent of imports; Canada accounts for less than 2 percent of exports and less than 1 percent of imports.
- Export and import composition (2023, EORA-based):
  - Principal exports: manufactured goods and primary agricultural products; services increasingly important.
  - Main manufacturing exports: textiles & apparel, machinery & equipment, food & beverages.
  - Textiles & apparel significance:
    - Honduras: textiles & apparel account for around 11 percent of GDP.
    - El Salvador: textiles & apparel account for around 8 percent of GDP.
  - Mining accounts for, on average, less than 1 percent of exports for CAPDR countries.
  - Petrol and chemical products imports range from around 3 percent of GDP in the Dominican Republic to nearly 10 percent of GDP in Nicaragua.
- GVC integration metrics:
  - Textiles & apparel industry: on average around 90 percent of gross output is exported, with around two thirds of those exports going to the United States (Panama exception: just below 2 percent of textiles & apparel exports to the United States).
  - For most manufacturing industries, on average around one third of gross output is exported.
  - Mining, utilities, construction and services: on average around 95 percent of their gross output purchased domestically.
  - Implication: Increased GVC participation raises exposure to overseas trade policies; tariffs applied upstream accumulate downstream via intermediate-input linkages.

### IV. Model structure, calibration and solution
- Key structural features:
  - Industries: i = {1,...,I}, with I = 13 (agriculture, mining, utilities, construction, services and eight manufacturing industries).
  - Trade frictions: ad-valorem tariffs (τ_jsd) and iceberg trade costs (κ_jsd).
  - Exact-hat formulation used for comparative statics (relative changes X̂ = X′/X).
- Data and calibration:
  - Main data source: EORA multi-region input-output database aggregated into 78 regions and 13 industries; base period for calibrations: 2023.
  - Labor supply elasticity ν = 0.5.
  - Trade elasticities θ_i estimated by structural gravity; for mining, utilities, construction and services τ_jsd set to 0 and θ̃_i calibrated to 8.
  - Table 1: Trade Elasticities (by industry):
    - Agriculture: 7.8
    - Mining*: 8.0
    - Food & beverages: 5.6
    - Textiles & apparel: 5.7
    - Wood & paper: 10.0
    - Petrol & chemicals: 7.4
    - Metal products: 9.1
    - Electrical & machinery: 15.1
    - Transport equipment: 13.4
    - Other manufacturing: 5.9
    - Utilities*: 8.0
    - Construction*: 8.0
    - Services*: 8.0
    - Median: 7.8
    - Note: * Calibrated elasticity
- Numerical solution:
  - Implemented in MATLAB using a variant of the Alvarez and Lucas (2007) algorithm.
  - Aggregate trade balance for each region held fixed at initial level: T̂B_d = T B_d′ T B_d = 1.
  - Iterative algorithm guesses expenditure prices and real wages, solves equilibrium conditions, updates guesses and iterates until error ε within tolerance.

### V. Scenario construction
- Scenarios evaluated (all results relative to baseline tariff rates on January 1, 2025):
  - Scenario 10pc: change in US tariff rates set at 10 percentage points on imports of all ‘traded goods’.
  - Scenario PreSC: change in tariff rates levied by the United States up to February 6, 2025.
  - Scenario PostSC: change in tariff rates levied by the United States up to February 24, 2025 (implicitly assumes Congress makes Trade Act tariffs permanent).
- Scenario notes:
  - Product-level tariff rates from WTO-IMF Tariff Tracker (2026) aggregated into model industries.
  - All scenarios include retaliatory tariffs enacted by Canada and China and exclude tariffs announced but not yet implemented.
  - Traded goods definition excludes utilities, construction and services.
  - Textiles & apparel exemptions in PostSC reduce effective tariffs for GTM, HND, SLV, NIC below the 10pc benchmark; CRI and DOM face higher effective tariffs more aligned with rest of world; PAN largely unchanged across PreSC and PostSC.

### VI. Main quantitative results (percent deviation from baseline)
- Real GDP (percent deviation):
  - Scenario 10pc: CRI -0.9, DOM -1.1, GTM -0.4, HND -0.8, NIC -0.7, PAN -0.5, SLV -0.5
  - Scenario PreSC: CRI -1.3, DOM -1.2, GTM -0.4, HND -0.8, NIC -1.1, PAN -0.6, SLV -0.5
  - Scenario PostSC: CRI -1.0, DOM -1.0, GTM -0.2, HND -0.4, NIC -0.3, PAN -0.5, SLV -0.1
- Real exports (percent deviation):
  - Scenario 10pc: CRI -1.4, DOM -1.7, GTM -0.6, HND -1.5, NIC -1.0, PAN -0.9, SLV -0.9
  - Scenario PreSC: CRI -2.0, DOM -1.9, GTM -0.6, HND -1.4, NIC -1.6, PAN -1.1, SLV -0.8
  - Scenario PostSC: CRI -1.4, DOM -1.5, GTM -0.4, HND -0.7, NIC -0.5, PAN -0.9, SLV -0.2
- Consumption (percent deviation):
  - Scenario 10pc: CRI 0.8, DOM 1.1, GTM 0.7, HND 1.2, NIC 0.4, PAN 0.6, SLV 0.7
  - Scenario PreSC: CRI 1.1, DOM 1.3, GTM 0.7, HND 1.2, NIC 0.5, PAN 0.7, SLV 0.7
  - Scenario PostSC: CRI 0.8, DOM 1.0, GTM 0.5, HND 0.6, NIC 0.2, PAN 0.6, SLV 0.3
- Domestic sales (percent deviation):
  - Scenario 10pc: CRI -0.5, DOM -0.6, GTM -0.2, HND -0.5, NIC -0.3, PAN -0.3, SLV -0.3
  - Scenario PreSC: CRI -0.7, DOM -0.6, GTM -0.2, HND -0.5, NIC -0.5, PAN -0.4, SLV -0.3
  - Scenario PostSC: CRI -0.5, DOM -0.5, GTM -0.1, HND -0.2, NIC -0.2, PAN -0.3, SLV -0.1
- Employment (percent deviation):
  - Scenario 10pc: CRI -0.9, DOM -1.2, GTM -0.4, HND -1.0, NIC -0.7, PAN -0.6, SLV -0.6
  - Scenario PreSC: CRI -1.3, DOM -1.3, GTM -0.4, HND -1.0, NIC -1.1, PAN -0.7, SLV -0.5
  - Scenario PostSC: CRI -1.0, DOM -1.0, GTM -0.2, HND -0.5, NIC -0.3, PAN -0.6, SLV -0.1
- Real wage (percent deviation, in terms of consumer prices):
  - Scenario 10pc: CRI -1.7, DOM -2.3, GTM -1.1, HND -1.4, NIC -0.8, PAN -1.0, SLV -0.8
  - Scenario PreSC: CRI -2.5, DOM -2.8, GTM -1.2, HND -1.5, NIC -1.3, PAN -1.4, SLV -0.9
  - Scenario PostSC: CRI -1.7, DOM -2.1, GTM -0.7, HND -0.8, NIC -0.5, PAN -1.1, SLV -0.3
- Terms of trade** (percent deviation):
  - Scenario 10pc: CRI -3.9, DOM -4.3, GTM -3.5, HND -3.3, NIC -3.5, PAN -3.2, SLV -3.3
  - Scenario PreSC: CRI -5.5, DOM -5.6, GTM -4.4, HND -4.1, NIC -4.9, PAN -4.4, SLV -4.3
  - Scenario PostSC: CRI -4.0, DOM -4.2, GTM -3.2, HND -2.7, NIC -3.2, PAN -3.4, SLV -2.9
- Key interpretive points:
  - Tariff increases act as an adverse foreign demand shock lowering exports, domestic prices and terms of trade for CAPDR members.
  - Cross-country heterogeneity driven primarily by direct export exposure to the United States, industry composition, input-output linkages, and trade elasticities.
  - Simple regressions suggest around 60 percent to 80 percent of cross-country variation in estimated real GDP change explained by variation in exports to the United States as a share of GDP (scenario-dependent).
  - Textiles & apparel exemptions in Scenario PostSC materially mitigate export and GDP losses for exempt countries (notably El Salvador).

### VII. Sensitivity analyses and model limitations
- Sensitivity to labor supply elasticity (Scenario PostSC, ν ∈ [0,2]):
  - Baseline ν = 0.5; ν = 0 corresponds to inelastic labor supply.
  - Higher ν → larger employment responses for given real wage changes; limited real wage response implies limited changes to terms of trade and real exchange rates.
- Sensitivity to trade elasticities (Scenario PostSC, scaling parameter ω ∈ (0,2.1], ω = 1 baseline):
  - Higher trade elasticities → larger import falls and greater reallocation to domestic sales; above thresholds, domestic prices and real wages may rise.
- Model limitations explicitly noted:
  - Static model: computes long-run steady-state changes following a policy change but does not trace transitional dynamics.
  - Omitted mechanisms: intertemporal optimization, monopolistic competition and firm entry, factor mobility frictions, labor market frictions and informal employment dynamics, trade policy uncertainty effects on investment and households.
  - Parameter uncertainty: trade elasticities and labor supply elasticity vary by industry, aggregation and horizon; sensitivity analyses used to assess robustness.

### VIII. Policy implications and suggested extensions
- Findings imply US tariff increases adversely affect CAPDR members, with severity correlated strongly with direct export exposure to the United States.
- Exemptions (especially for textiles & apparel) can offset real GDP losses to varying degrees across CAPDR members.
- Suggested model extensions and policy analysis:
  - Incorporate dynamics via intertemporal saving decisions to endogenize trade imbalances and analyze transitional adjustment.
  - Investigate policy levers to offset adverse effects, including interactions between exchange rate flexibility and adjustment channels.
  - Assess how deeper regional integration and reductions in non-tariff barriers could mitigate tariff impacts.

*Italic: IMF Working Papers — Appendix B: Structural Gravity Estimates (content extracted from the supplied source PDF).*

### References .............................................................................................................

### wpiea2026125-source-pdf - References .............................................................................................................

### References
- References ......................................................................................................................................................... 26

### Appendix
- Appendix A: Model Aggregation ..................................................................................................................... 29

*Source: https://www.imf.org/-/media/files/publications/wp/2026/english/wpiea2026125-source-pdf.pdf*

### Appendix B: Structural Gravity Estimates ...............................................................................

### Appendix B: Structural Gravity Estimates

### I. Introduction
- Context and purpose
  - Examines potential long-term economic effects of recent US tariff increases (and associated retaliatory measures) on CAPDR (Costa Rica, Dominican Republic, Guatemala, Honduras, Nicaragua, Panama and El Salvador) using a quantitative multi-region multi-industry model of international trade.
  - Model is an Eaton and Kortum (2002) static Ricardian comparative advantage trade model comparable to Caliendo and Parro (2015) and Shikher (2012).
  - Global economy in the model: 78 separate regions, each with 13 industries; each CAPDR member modelled as a separate region; production divided into agriculture, mining, utilities, construction, services and eight manufacturing industries.
  - Model includes detailed global input-output linkages and international trade in intermediate and final goods; distinguishes gross output and value added by including capital and labor as variable primary factors of production.

- Recent tariff events summarized in source text
  - February 2025: United States announced tariffs of 25 percent on most goods imported from Canada and Mexico.
  - Through 2025: United States widened tariff coverage to include most countries; higher rates for varying periods on Brazil, China, the European Union and India, among others.
  - By end-2025: United States imposed additional statutory tariff rates of 18 percent on Nicaragua, 15 percent on Costa Rica, and 10 percent on the remaining members of CAPDR.
  - Guatemala and El Salvador negotiated exemptions for most textiles & apparel (The White House, 2025a and 2025b).
  - February 2026: United States Supreme Court overturned many 2025 tariffs; President invoked alternative legal instruments to introduce a uniform 10 percent tariff, with exemptions for CAFTA-DR‑compliant textiles & apparel and wide-ranging exemptions for USMCA‑compliant products (The White House, 2026).
  - May 7, 2026: United States Court of International Trade declared these tariffs unlawful but left them in place while appeals are pending.

- Key high-level model insights reported
  - Tariff increases act as an adverse foreign demand shock by lowering exports, domestic prices and the terms of trade for all CAPDR members.
  - All CAPDR members experience falls in real GDP to varying degrees, severity linked closely to direct export exposure to the United States.
  - Under the scenario most representative of the current external environment facing CAPDR, estimated effects on real GDP range between a fall of around 1 percent in Costa Rica and the Dominican Republic, to a fall of around 0.1 percent in El Salvador.
  - Manufacturing industries that are more integrated into Global Value Chains (GVCs) and heavily reliant on imported intermediate inputs are the most vulnerable.
  - Exemptions—most importantly for textiles & apparel—open additional channels for international relative price adjustments and may provide scope to offset some real GDP losses.

### II. Stylized Facts (CAPDR trade and GVC exposure)
- Trade openness and exposure
  - Honduras: most trade exposed with total trade nearly 100 percent of GDP.
  - Guatemala: least trade exposed, with total trade of around 40 percent of GDP.
  - Comparative totals: United States ~25 percent of GDP, Canada ~65 percent of GDP, Mexico ~90 percent of GDP (total trade).
  - CAFTA timeline: 2006 initial effect for El Salvador, Honduras, Nicaragua and Guatemala; Dominican Republic joined in 2007; Costa Rica joined in 2009. Panama not a CAFTA-DR member but has a separate US agreement effective 2012.
  - On average, the United States accounts for around 40 percent of exports and around 37 percent of imports for CAPDR members.
  - Canada accounts for, on average, less than 2 percent of exports and less than 1 percent of imports for CAPDR members.

- Export and import composition (2023, EORA-based)
  - Principal exports: manufactured goods and primary agricultural products; services increasingly important.
  - Main manufacturing exports: textiles & apparel, machinery & equipment, food & beverages.
  - Textiles & apparel exports especially significant:
    - Honduras: textiles & apparel account for around 11 percent of GDP.
    - El Salvador: textiles & apparel account for around 8 percent of GDP.
  - CAPDR increasingly exports services and higher value-added manufactured goods (electrical components, pharmaceuticals, medical devices).
  - Primary products: coffee, sugar, bananas and other tropical fruits.
  - Mining accounts for, on average, less than 1 percent of exports for all CAPDR countries.
  - CAPDR imports: services, manufactured goods and fuel.
    - Petrol and chemical products imports range from around 3 percent of GDP in the Dominican Republic to nearly 10 percent of GDP in Nicaragua.

- GVC integration metrics
  - Textiles & apparel industry: on average around 90 percent of the gross output of this industry is exported, with around two thirds of those exports going to the United States (Panama exception: just below 2 percent of textiles & apparel exports to the United States).
  - For most manufacturing industries, on average around one third of gross output is exported.
  - Mining, utilities, construction and services are the least export-exposed, with an average of around 95 percent of their gross output purchased domestically.
  - VAX/gross exports ratio trends:
    - Domestic value-added component of exports has become less important for most CAPDR countries as they integrated into GVCs.
    - Panama: very low domestic value-added content in exports due to re-exports through Colon Free Trade Zone and export processing zones.
    - Industries closer to non-traded sectors (mining, utilities, construction, services) show higher domestic value-added content in exports; manufacturing industries show higher foreign content.
  - Implication: Increased GVC participation raises exposure to overseas trade policies; tariffs applied upstream accumulate downstream through intermediate-input linkages.

### III. Model Description (structure, agents, and equilibrium)
- Model overview and agents
  - Static Ricardian multi-industry trade model with input-output linkages and trade in intermediate and final goods; follows Eaton and Kortum (2002) with multicountry/multiindustry extensions.
  - Regions: s = {1,2,...,S} source regions and d = {1,2,...,D} destination regions with S = D.
  - Industries: i = {1,2,...,I} supply industries and j = {1,2,...,J} demand industries with I = J.
  - Representative household in each region maximizes u(C_s^R, N_s) = C_s^R − ν/(1+ν) N_s^(1+ν)/ν, where ν is the labor supply elasticity.
  - Aggregate real consumption is a Cobb-Douglas bundle across industries with shares α_js^C summing to 1.
  - Household income components: capital income R_s^K K_is, labor income w_s N_is, tariff revenues rebated TR_s, and net overseas transfers (trade balance TB_jd).

- Production technology and factor inputs
  - Each industry i in region s comprises a continuum of firms producing unique varieties; real gross production assembled via a CES aggregator with elasticity η across varieties.
  - Firm-level production combines capital, labor and intermediate inputs via a Cobb-Douglas technology; firm productivity z_is(ω) drawn from a Fréchet distribution with parameters T_is and θ_i.
  - Unit cost P_is^Y depends on real capital rental rate R_s^K, real wage w_s, composite intermediate input price P_s^J, capital share α_is^K and value-added share α_is^V:
    - P_is^Y = ((R_s^K)^{α_is^K} (w_s)^{1−α_is^K})^{α_is^V} (P_s^J)^{1−α_is^V}.
  - Composite intermediate input price P_s^J is Cobb-Douglas across industry input prices P_js^Q with shares α_js^J summing to 1.

- Trade frictions and pricing
  - Productivity draws follow F_is(z) = exp(−T_is z^{−θ_i}); θ_i is the trade elasticity.
  - International trade subject to ad-valorem tariffs (τ_jsd) and iceberg trade costs (κ_jsd); κ_jsd > 1 units must be shipped for one unit to arrive.
  - Price of output from industry j shipped from s to d equals unit production cost adjusted for (1+τ_jsd), κ_jsd and firm productivity; purchasers buy lowest-cost supplier.
  - Aggregate price and trade share formulas (Eaton and Kortum derivations) reported in model: P_jd^Q and π_jsd given in closed form involving T_is, (1+τ_jsd), κ_jsd, P_js^Y and θ_i.

- Equilibrium and clearing conditions
  - Investment, consumption and intermediate demand sums satisfy budget and market-clearing conditions:
    - Total expenditure Q_jd = J_jd + C_jd + I_jd = D_jd + Σ_s M_jsd.
    - Gross output Y_is = D_jsd + Σ_d X_jsd = Q_jd + TB_jd.
    - Value added VA_s equals sum of gross output less intermediate usage and equals sum of factor incomes plus transfers and trade balance components.
  - Tariff revenue: TR_s = Σ_{j,s} τ_jsd π_jsd Q_jd / (1+τ_jsd).

- Exact-hat formulation for comparative statics
  - Model re-expressed in exact hat form (relative changes X̂ = X′/X) to simplify calibration and numerical solution.
  - Exact-hat system reduces parameters to trade shares π_jsd, α shares, trade elasticities θ_i, iceberg trade costs κ_jsd, and changes in gross tariff rates (1+τ_jsd)̂.
  - Aggregate reporting variables defined via share-weighted sums; real GDP (real value added) change formula provided as VÂ_s^R = Σ α_is^{VA} [Ŷ_is / P̂_is^Y − (1−α_is^V) Ĵ_is / P̂_s^J].

### IV. Model Parameterization and Solution
- Data and aggregation
  - Main data source: EORA multi-region input-output database.
  - EORA aggregated into 78 regions (77 countries + Rest of World) and 13 industries to calibrate the model.
  - Main focus: United States and individual CAPDR members modelled as separate regions.
  - 13 industries: agriculture, mining, utilities, construction, services and eight manufacturing industries.

- Calibration and computational approach
  - Use of exact-hat formulation reduces calibration to observable shares and elasticities; productivity level parameters T_is substituted out.
  - Numerical solution solves the exact-hat system once for comparative static outcomes under alternative tariff scenarios.
  - Model closure includes steady-state relationships for capital rental rate R_s^K = (r+δ) P_s^I and nominal investment proportionality Ω_is = (δ/(r+δ)) R_s^K K_is.

### Key quantitative figures and parameters reported in text (preserved exactly)
- US tariff policy chronology and rates:
  - 25 percent: tariffs announced February 2025 on most goods imported from Canada and Mexico.
  - 18 percent: additional statutory tariff rate on Nicaragua by end-2025.
  - 15 percent: additional statutory tariff rate on Costa Rica by end-2025.
  - 10 percent: additional statutory tariff rate on remaining CAPDR members by end-2025.
  - 10 percent: uniform tariff introduced February 2026 via alternative legal instruments (with exemptions for CAFTA-DR textiles & apparel and USMCA‑compliant products).
- Estimated real GDP effects under representative scenario:
  - Fall of around 1 percent in Costa Rica and the Dominican Republic.
  - Fall of around 0.1 percent in El Salvador.
- Trade exposure and composition:
  - United States accounts for around 40 percent of exports and around 37 percent of imports on average for CAPDR members.
  - Honduras: total trade nearly 100 percent of GDP.
  - Guatemala: total trade around 40 percent of GDP.
  - Textiles & apparel: around 90 percent of gross output exported on average in CAPDR, with around two thirds to the United States (Panama exception: just below 2 percent of textiles & apparel exports to the United States).
  - Textiles & apparel share of GDP: Honduras around 11 percent, El Salvador around 8 percent.
  - Petrol and chemical products imports: around 3 percent of GDP in the Dominican Republic to nearly 10 percent of GDP in Nicaragua.
- Model dimensionality and structure:
  - 78 regions, 13 industries, inclusion of capital and labor as variable primary factors; production nesting distinguishing gross output and value added.

*Italic: IMF Working Papers — Appendix B: Structural Gravity Estimates (content extracted from the supplied source PDF).*

### Appendix A

### Appendix A

### Calibrated Parameters
- Base period for calibrations: 2023 (the most recent year for which data was available).
- Production, expenditure and trade shares calibrated using EORA.
- Capital income share of value added (훼푖푠퐾) = (net operating surplus + consumption of fixed capital) / (net operating surplus + consumption of fixed capital + compensation of employees + gross mixed income) for each industry in each country.
- Share of value added in gross output (훼푖푠푉) = 1 − (total intermediate input usage (summed across source industries) / gross output) for each industry in each country.
- 훼푖푠퐽 = share of total expenditure on intermediate inputs (summing across destination industries) allocated to each industry.
- 훼푖푠퐶 and 훼푖푠퐼 calculated similarly:
  - Total consumption = consumption by households + government + non-profit institutions serving households.
  - Total investment = gross fixed capital formation + acquisitions less disposals.
  - Note: Change in inventories is excluded because its inclusion does not guarantee positive investment for all industries at the level of industry aggregation in this model.
- Expenditure on exports, imports and domestic sales calculated from EORA using equations (15), (17) and (18).
- Resulting 78 × 78 matrices for each of the 13 industries represent 푄푗푠푑:
  - Domestic sales = diagonal elements.
  - Imports = off-diagonal elements down the columns.
  - Exports = off-diagonal elements across the rows (퐗 = 퐌′).
- 휋푗푠푑 calculated as matrix of imports from all sources (including domestic) as a share of total expenditure in the destination region for each industry.
- Labor supply elasticity (휐) = 0.5 (Calibrated to Chetty et. al (2011) preferred estimate; assumed same for all regions following Ignatenko et. al (2025)).

### Estimated Parameters
- Trade elasticities (휃푖) estimated using a structural gravity approach with variation in bilateral tariff rates.
- Estimation approach:
  - Start from equation (12), divide each import expenditure share by corresponding domestic expenditure share:
    - 휋푗푠푑 / 휋푗푑푑 = (푇푖푠 / 푇푖푑) × ((1+휏푗푠푑)휅푗푠푑 × 푃푗푠푌 / 푃푗푑푌)^(−휃푖)
  - Take logs and define exporter fixed effect 퐹퐸푗푠 = log(푇푖푠 푃푗푠 푌^(−휃푖)) and importer fixed effect 퐹퐸푗푑 = −log(푇푖푑 푃푗푑 푌^(−휃푖)) to obtain:
    - log(휋푗푠푑 / 휋푗푑푑) = 퐹퐸푗푠 + 퐹퐸푗푑 − 휃푖 log(1+휏푗푠푑) − 휃푖 log(휅푗푠푑)
  - Iceberg trade costs proxied by k gravity variables: log(distance), dummies for domestic trade, common legal origin, common land border, common language, common free trade agreement membership, etc.
  - Estimation by Poisson Pseudo Maximum Likelihood (Santos Silva and Tenreyo, 2006) using gravity variables (Gurevich and Herman, 2018) and tariff rates (Teti, 2024).
  - Econometric specification:
    - 휋푗푠푑 / 휋푗푑푑 = exp[퐹퐸̃푗푠 + 퐹퐸̃푗푑 − 휃̃푖 log(1+휏푗푠푑) + ∑(훽̃푗푘 × gravityk)_{K_k=1}] × 휀푗푠푑
  - Fixed effects normalized: 퐹퐸̃USA,s = 퐹퐸̃USA,d = 0.
  - 휃̃푖 is the estimated trade elasticity for industry i; 휀푗푠푑 is multiplicative error term varying by industry and bilateral pair.
  - The 78×78×13 ‘estibrated’ iceberg trade costs are recovered using estimated fixed effects, trade elasticities and error terms.
- Treatment of industries with unreliable tariff data:
  - For mining, utilities, construction and services, 휏푗푠푑 set to 0 and 휃̃푖 calibrated to 8 (approximately the median of estimates for the remaining 9 industries).

### Table 1: Trade Elasticities (by industry)
- Agriculture: 7.8
- Mining*: 8.0
- Food & beverages: 5.6
- Textiles & apparel: 5.7
- Wood & paper: 10.0
- Petrol & chemicals: 7.4
- Metal products: 9.1
- Electrical & machinery: 15.1
- Transport equipment: 13.4
- Other manufacturing: 5.9
- Utilities*: 8.0
- Construction*: 8.0
- Services*: 8.0
- Median: 7.8
- Note: * Calibrated elasticity

### Solution Method
- Numerical solution implemented in MATLAB using a variant of the Alvarez and Lucas (2007) algorithm (similar approaches in Alvarez (2017), Ravikumar et. al (2019, 2024)).
- Aggregate trade balance for each region held fixed at initial level: 푇퐵̂푑 = 푇퐵푑′ 푇퐵푑 = 1.
- Algorithm steps:
  1. Guess a 13×78 matrix of expenditure prices (푃̂푗푑푄) and a 1×78 vector of real wages (푤̂푠).
  2. Use these guesses to solve equations (20) to (37).
  3. Update the guesses according to:
     - 풢_{푛푒푤} = 풢 × [1 + 0.01( ∑_{푗=1}^{풱}(훼푗푑_푋 𝑋̂푗푑 − 훼푗푑_푀 𝑀̂푖푑 − 훼푗푑_푇퐵 푇퐵̂푖푑) / ∑_{푖=1}^{ℐ} 훼푖푠_푉퐴 𝑉퐴̂푖푠 ) ]
     - Where 풢 = 푃̂푗푑푄 or 푤̂푠.
     - ∑_{푗=1}^{풱} 훼푗푑_푇퐵 푇퐵̂푖푑 = 훼푑_푇퐵 and the 훼푗푑_푘’s denote the share of k in world value added.
  4. Compute the error (휀) as:
     - 휀 = max( max|푃̂푗푑,푛푒푤푄 / 푃̂푗푑푄 − 1| ) + max|푤̂푠,푛푒푤 / 푤̂푠 − 1|
- Appendix B reports results for the structural gravity regressions.

*Source: IMF staff calculations.*

### 5. Use the updated guesses to repeat steps 1 to 4.  Iterate until 휀 is within a given tolerance level.

### 5. Use the updated guesses to repeat steps 1 to 4.  Iterate until 휀 is within a given tolerance level.

### Model Limitations
- The model is a static model that numerically computes the change in steady states following a policy change and does not trace transitional dynamics between short and long term.
- Static approach rationale:
  - Theoretical and empirical studies generally conclude that permanent tariffs have little to no effect on external imbalances over the long term (i.e. Furceri et. al, 2022 and Gourinchas et. al, 2026) because external balances are determined by intertemporal choices affecting savings and investment; permanent tariffs generally lead to offsetting real exchange rate appreciation.
- Important omitted dynamics and mechanisms:
  - Intertemporal optimization and transitional dynamics (would allow trade balances to be determined endogenously and capture amplified welfare effects—Ravikumar et. al, 2024; Hoang and Mix, 2026; Ravikumar et. al, 2019).
  - Monopolistic competition, firm-level entry, and market-size effects (Melitz, 2003).
  - Adjustment frictions preventing factor mobility across industries (Ahn and Tan, 2025).
  - Labor market frictions and informal employment dynamics (Arias et. al, 2018).
  - Trade policy uncertainty effects on investment and households (Caldara et. al, 2020; Jakubik and Wei, 2026).
- Parameter uncertainty:
  - Trade elasticities vary by industry, aggregation, and horizon; commonly used aggregate long-term value ~4 (Simonovska and Waugh, 2014); cross-industry median in Table 1 ~8 (Caliendo and Parro, 2015; Teti, 2024).
  - Short-term trade elasticities are generally smaller (Boehm et. al, 2023 find short-term elasticities ~ one third of long-term counterparts; convergence ~7 to 10 years).
  - Labor supply elasticity estimates vary widely (Chetty et. al, 2011).
- Sensitivity analyses are used to assess robustness to trade and labor supply elasticity choices.

### Scenario Construction
- Key dates and policy actions (as described in the source):
  - February 2025: United States announced tariffs of 25 percent on most goods from Canada and Mexico.
  - April 2025: United States announced a minimum 10 percent tariff rate on imports from most countries.
  - August 2025: Additional tariffs ranging from 5 percent to 40 percent on imports from many source countries.
  - End-2025: United States statutory tariff rates of 18 and 15 percent on Nicaragua and Costa Rica respectively, and 10 percent on the Dominican Republic, Guatemala, Honduras, Panama and El Salvador.
  - November 13, 2025: ‘Framework agreements’ announced with Guatemala and El Salvador to take effect upon ratification in 2026, including US removal of tariffs on selected textiles & apparel and some agricultural and mining products not produced domestically.
  - December 10, 2025: Additional incremental tariffs on Nicaragua to be phased in between 2027 and 2028.
  - February 20, 2026: US Supreme Court overturned many tariffs enacted under IEEPA; President issued Trade Act proclamation introducing a 10 percent ad-valorem tariff on all imports (maximum period 150 days before congressional action), with exemptions for textiles & apparel from Costa Rica, Dominican Republic, Guatemala, Honduras, Nicaragua and El Salvador compliant with CAFTA-DR, and products compliant with USMCA.
  - May 7, 2026: US Court of International Trade declared Trade Act tariffs unlawful but left them in place during appeals.
- Scenarios evaluated:
  - Scenario 10pc: The change in tariff rates levied by the United States on imports of all ‘traded goods’ from all trading partners is set at 10 percentage points.
  - Scenario PreSC: The change in tariff rates levied by the United States up to February 6, 2025.
  - Scenario PostSC: The change in tariff rates levied by the United States up to February 24, 2025 (implicitly assumes Congress makes tariffs imposed under the Trade Act permanent).
- Notes on scenario construction:
  - Product-level tariff rates obtained from the WTO-IMF Tariff Tracker (2026) and aggregated into model industries.
  - All results reported relative to baseline tariff rates prevailing on January 1, 2025.
  - All scenarios include retaliatory tariffs enacted by Canada and China and exclude tariffs announced but not yet implemented.
  - Scenario PreSC excludes exemptions for Guatemala and El Salvador under framework agreements; Scenario PostSC excludes planned additional incremental tariffs on Nicaragua.
  - ‘Traded goods’ definition: agriculture, mining, food & beverages, textiles & apparel, wood & paper, petrol & chemicals, metal products, electrical & machinery, transport equipment and other manufacturing industries; excludes utilities, construction and services.
  - Panama does not receive textiles & apparel exemptions and faces similar effective tariff rates in PreSC and PostSC.
- Aggregate effective tariff changes summarized in Figure 7 (described): exemptions for textiles & apparel in PostSC reduce effective tariffs for GTM, HND, SLV, NIC below the 10pc benchmark; CRI and DOM face higher effective tariffs more aligned with rest of world; PAN largely unchanged across PreSC and PostSC.

### Main Results
- General equilibrium channels:
  - Higher US tariffs widen the wedge between source producer prices and US importer prices → US import volumes fall → expenditure switches toward domestic sales in US.
  - Higher tariffs raise prices of imported intermediate inputs → through input-output linkages raise export prices → affect global production costs.
  - Even uniform tariffs (Scenario 10pc) change composition of US imports: industries with higher trade elasticities see larger import falls; non-traded-good industries (utilities, construction, services) rise or fall by less.
  - For rest of world (excluding CAPDR): lower US demand reduces exports, partially offset by lower prices, increased domestic production, or increased exports to other destinations; net effect mostly negative for real GDP and leads to real exchange rate depreciation.
  - Limited retaliatory measures by most countries so far; Canada (steel, aluminum, some auto) and China (broad) are notable exceptions—retaliation shifts demand toward third countries but has limited bearing on CAPDR results given current scope.
- Mechanism for CAPDR effects:
  - Tariffs act as adverse foreign demand shock → lower real GDP, domestic prices in CAPDR.
  - Tariffs raise import prices in CAPDR → increase domestic production costs via imported intermediate inputs → fall in terms of trade → negative income effect reducing consumption, investment, intermediate input demand.
  - Lower productivity → lower real wages → households reduce labor supply → reinforces employment declines but dampens real wage fall.
  - Input-output linkages disperse tariff effects across countries and industries and dampen trade diversion benefits for CAPDR.
  - Exemptions (notably textiles & apparel) provide scope for substitution across industries and trade diversion that can offset some negative effects depending on country exposure.
- Summary of main quantitative results (Table 2: Percent deviation from baseline)
  - Real GDP:
    - Scenario 10pc: CRI -0.9, DOM -1.1, GTM -0.4, HND -0.8, NIC -0.7, PAN -0.5, SLV -0.5
    - Scenario PreSC: CRI -1.3, DOM -1.2, GTM -0.4, HND -0.8, NIC -1.1, PAN -0.6, SLV -0.5
    - Scenario PostSC: CRI -1.0, DOM -1.0, GTM -0.2, HND -0.4, NIC -0.3, PAN -0.5, SLV -0.1
  - Real exports:
    - Scenario 10pc: CRI -1.4, DOM -1.7, GTM -0.6, HND -1.5, NIC -1.0, PAN -0.9, SLV -0.9
    - Scenario PreSC: CRI -2.0, DOM -1.9, GTM -0.6, HND -1.4, NIC -1.6, PAN -1.1, SLV -0.8
    - Scenario PostSC: CRI -1.4, DOM -1.5, GTM -0.4, HND -0.7, NIC -0.5, PAN -0.9, SLV -0.2
  - Consumption:
    - Scenario 10pc: CRI 0.8, DOM 1.1, GTM 0.7, HND 1.2, NIC 0.4, PAN 0.6, SLV 0.7
    - Scenario PreSC: CRI 1.1, DOM 1.3, GTM 0.7, HND 1.2, NIC 0.5, PAN 0.7, SLV 0.7
    - Scenario PostSC: CRI 0.8, DOM 1.0, GTM 0.5, HND 0.6, NIC 0.2, PAN 0.6, SLV 0.3
  - Domestic sales:
    - Scenario 10pc: CRI -0.5, DOM -0.6, GTM -0.2, HND -0.5, NIC -0.3, PAN -0.3, SLV -0.3
    - Scenario PreSC: CRI -0.7, DOM -0.6, GTM -0.2, HND -0.5, NIC -0.5, PAN -0.4, SLV -0.3
    - Scenario PostSC: CRI -0.5, DOM -0.5, GTM -0.1, HND -0.2, NIC -0.2, PAN -0.3, SLV -0.1
  - Employment:
    - Scenario 10pc: CRI -0.9, DOM -1.2, GTM -0.4, HND -1.0, NIC -0.7, PAN -0.6, SLV -0.6
    - Scenario PreSC: CRI -1.3, DOM -1.3, GTM -0.4, HND -1.0, NIC -1.1, PAN -0.7, SLV -0.5
    - Scenario PostSC: CRI -1.0, DOM -1.0, GTM -0.2, HND -0.5, NIC -0.3, PAN -0.6, SLV -0.1
  - Real wage* (expressed in terms of consumer prices):
    - Scenario 10pc: CRI -1.7, DOM -2.3, GTM -1.1, HND -1.4, NIC -0.8, PAN -1.0, SLV -0.8
    - Scenario PreSC: CRI -2.5, DOM -2.8, GTM -1.2, HND -1.5, NIC -1.3, PAN -1.4, SLV -0.9
    - Scenario PostSC: CRI -1.7, DOM -2.1, GTM -0.7, HND -0.8, NIC -0.5, PAN -1.1, SLV -0.3
  - Terms of trade** (export price index relative to import price index):
    - Scenario 10pc: CRI -3.9, DOM -4.3, GTM -3.5, HND -3.3, NIC -3.5, PAN -3.2, SLV -3.3
    - Scenario PreSC: CRI -5.5, DOM -5.6, GTM -4.4, HND -4.1, NIC -4.9, PAN -4.4, SLV -4.3
    - Scenario PostSC: CRI -4.0, DOM -4.2, GTM -3.2, HND -2.7, NIC -3.2, PAN -3.4, SLV -2.9
  - Notes:
    - Exports reductions are driven largely by reduced exports to the United States, with manufacturing and agriculture most affected.
    - Services exports often increase since services are not subject to tariffs in the model and become relatively cheaper to foreigners.
    - Textiles & apparel exemptions in Scenario PostSC materially mitigate export and GDP losses for exempt countries (notably El Salvador).
- Cross-country heterogeneity and drivers:
  - Differences driven by composition of output/expenditure, global input-output linkages, trade elasticities, and changes in tariff rates.
  - Higher industry trade elasticities → larger export changes for those industries.
  - Countries with higher direct export exposure to the United States experience larger adverse GDP effects.
  - Simple linear regressions suggest around 60 percent to 80 percent of cross-country variation in estimated change in real GDP can be explained by cross-country variation in exports to the United States as a share of GDP before the tariff change (scenario-dependent).
  - Relationship between GDP changes and aggregate effective tariff rate changes is less clear.

### Sensitivity Analysis
- Sensitivity to labor supply elasticity (Scenario PostSC, labor supply elasticity parameter 휐 ∈ [0,2]):
  - Baseline calibration uses 휐 = 0.5; 휐 = 0 corresponds to inelastic labor supply.
  - Higher labor supply elasticity → larger employment responses for given real wage changes → reflected in output through production structure.
  - Labor demand curves appear relatively elastic: real wages remain stable while labor supply curves flatten as 휐 increases.
  - Limited real wage response implies limited changes to terms of trade and real exchange rates; domestic sales adjust more to match output changes.
- Sensitivity to trade elasticities (Scenario PostSC, scaling parameter 휔 ∈ (0,2.1], where 휔 = 1 is baseline):
  - Higher trade elasticities → domestic sales and imports more responsive to tariff changes (imports fall and domestic sales increase more when tariffs rise).
  - Lower imports by destination countries → lower exports by source countries; production redirected to domestic sales where feasible.
  - Above country-specific threshold trade elasticities, higher domestic sales give way to higher domestic prices and real wages.
  - Figures 12 and 13 report detailed percent deviations under varying 휐 and 휔 for Real GDP, Real exports, Real wage, Real domestic sales, and Terms of trade (Scenario PostSC).

### Conclusion
- The multi-region multi-industry quantitative model suggests US tariff increases act as an adverse foreign demand shock that adversely affects CAPDR members over the long term.
- Under the scenario most representative of the current external environment facing CAPDR (Scenario PostSC), estimated long-term effects on real GDP range between:
  - a fall of around 1 percent in Costa Rica and the Dominican Republic, to
  - a fall of around 0.1 percent in El Salvador.
- Severity correlates strongly with direct export exposure to the United States; exemptions (especially textiles & apparel) can offset real GDP losses to varying degrees across CAPDR members.
- Model extensions that could provide additional insights:
  - Incorporating dynamics via intertemporal saving decisions to endogenize trade imbalances and analyze transitional adjustment.
  - Investigating policy levers to offset adverse effects, including exploring interactions between exchange rate flexibility and adjustment channels.
  - Assessing how deeper regional integration and reductions in non-tariff barriers could mitigate tariff impacts.

*IMF Working Papers — Long-term Effects of US Tariff Increases on CAPDR*

### References

### References

### Scholarly references and working papers
- Ahn, J. and Tan, B., 2025, Supply Chain Diversification and Resilience, IMF Working Paper No. WP/25/102.
- Alvarez, F., and Lucas, R., 2007, ‘General Equilibrium Analysis of the Eaton-Kortum Model of International Trade’, Journal of Monetary Economics, vol. 54: 1726-1768.
- Alvarez, F., 2017, ‘Capital Accumulation and International Trade’, Journal of Monetary Economics, vol. 91: 1-18.
- Anderson, J., Larch, M., and Yotov, Y., 2018, ‘GEPPML: General Equilibrium Analysis with PPML’, The World Economy, vol. 41: 2750–2782.
- Arias, J., Artuc, E., Lederman, D., and Rojas, D., 2018, ‘Trade, Informal Employment and Labor Adjustment Costs’, Journal of Development Economics, vol. 133: 396-414.
- Aslam, A., Novta, N., and Rodrigues-Bastos, F., 2017, Calculating Trade in Value Added, IMF Working Paper No. WP/17/178.
- Boehm, C., Levchenko, A., and Pandalai-Nayar, N., 2023, ‘The Long and Short (Run) of Trade Elasticities’, American Economic Review, vol. 113: 861–905.
- Borchert, I., Larch, M., Shikher, S., and Yotov, Y., 2022, ‘Disaggregated Gravity: Benchmark Estimates and Stylized Facts from a New Database’, Review of International Economics, vol. 30: 113-136.
- Brotto, A., Exton, O., Gonclarz, T., Jakubik, A., Ruta, M., and Verbeet, T., 2026, Methodology Note for the WTO-IMF Tariff Tracker, updated 13 May 2026, available at: https://ttd.wto.org/en/reports/tariff-actions.
- Caldara, D., Iacoviello, M., Molligo, P., Prestipino, A., and Raffo, A., 2020, ‘The Economic Effects of Trade Policy Uncertainty’, Journal of Monetary Economics, vol. 109: 38-59.
- Caliendo, L., and Parro, F., 2015, ‘Estimates of the Trade and Welfare Effects of NAFTA’, Review of Economic Studies, vol. 82: 1-44.
- Chetty, R., Guren, A., Manoli, D., and Weber, A., 2011, ‘Are Micro and Macro Labor Supply Elasticities Consistent?  A Review of the Evidence on the Intensive and Extensive Margins’, American Economic Review: Papers and Proceedings, vol. 101: 471-475.
- Dekle, R., Eaton, J., and Kortum, S., 2008, ‘Global Rebalancing with Gravity: Measuring the Burden of Adjustment’, IMF Staff Papers, vol. 55: 511-540.
- Eaton, J., and Kortum, S., 2002, ‘Technology, Geography, and Trade’, Econometrica, vol. 70: 1741-1779.
- Eugster, J., Jaumotte, F., MacDonald, M., and Piazza, R., 2022, The Effect of Tariffs on Global Value Chains, IMF Working Paper No. WP/22/40.
- Fontagne, L., Guimbard, H., and Orefice, G., 2022, ‘Tariff-based Product-level Trade Elasticities’, Journal of International Economics, vol. 137: 103593.
- Furceri, D., Hannan, S., Ostry, J., and Rose, A., 2022, The Macroeconomy After Tariffs, The World Bank Economic Review, vol. 36: 361–381.
- Gourinchas, P-O., Kindberg-Hanlon, G., Patnam, M., Rotunno, L., and Ruta, M., 2026, Global Imbalances, Industrial Policy and Tariffs, IMF Working Paper No. WP/26/67.
- Gurevich, T., and Herman, P., 2018, The Dynamic Gravity Dataset: 1948-2016. USITC Working Paper 2018-02-A.
- Head, K., and Mayer, T., in Gopinath, G., and Rogoff, K. (eds.), 2014, ‘Gravity Equations: Workhorse, Toolkit and Cookbook’, in Handbook of International Economics, vol. 4: 131-194.
- Hoang, T., and Mix, C., 2026, ‘Trade Wars and Rumors of Trade Wars: The Dynamic Effects of US-China Tariff Hikes’, Journal of International Economics, vol. 160: 104229.
- Ignatenko, A., Lashkaripour, A., Macedoni, L., and Simonovska, I., 2025, ‘Making America Great Again? The Economic Impacts of Liberation Day Tariffs’, Journal of International Economics, vol. 157: 104138.
- Jakubik, A. and Wei, Y., 2026, US Trade Policy Uncertainty and the Current Account: Unpacking Trade and Financial Channels, IMF Working Paper No. WP/26/17.
- Johnson, R., 2018, ‘Measuring Global Value Chains’, Annual Review of Economics, vol. 10: 207–236.
- Koopman, R., Wang, Z., and Wei, S., 2014, ‘Tracing Value Added and Double Counting in Gross Exports’, American Economic Review, vol. 104: 459–494.
- Lenzen, M., Kanemoto, K., Moran, D., and Geschke, A., 2012, ‘Mapping the Structure of the World Economy’. Environmental Science & Technology, vol. 46: 8374–8381.
- Lenzen, M., Kanemoto, K., Moran, D., and Geschke, A., 2013, ‘Building EORA: A Global Multi-Regional Input-Output Database at High Country and Sector Resolution’. Economic Systems Research, vol. 25: 20-49.
- Levchenko, A., and Zhang, J., 2016, ‘The Evolution of Comparative Advantage: Measurement and Welfare Implications’, Journal of Monetary Economics, vol. 78: 96-111.
- Manak, I., 2026, How Trump’s Tariffs Could Survive a Supreme Court Ruling, Council on Foreign Relations, updated 20 February 2026, available at: https://www.cfr.org/articles/how-trumps-tariffs-could-survive-the-supreme-court-ruling
- Melitz, M., 2003, ‘The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity’, Econometrica, vol. 71: 1695-1725.
- Mutreja, P., Ravikumar, R., and Sposi, M., 2018, ‘Capital Goods Trade, Relative Prices and Economic Development’, Review of Economic Dynamics, vol. 27: 101-122.
- Ravikumar, R., Santacreu, A.M., and Sposi, M., 2019, ‘Capital Accumulation and the Dynamic Gains from Trade’, Journal of International Economics, vol. 119: 93-110.
- Ravikumar, R., Santacreu, A.M., and Sposi, M., 2024, ‘Trade Liberalization versus Protectionism: Dynamic Welfare Asymmetries’, European Economic Review, vol. 163: 104692.
- Rotunno, L., and Ruta, M., 2026, ‘Trade Partners’ Response to US Tariffs’, IMF Economic Review.
- Santos Silva, J., and Tenreyo, S., 2006, ‘The Log of Gravity’, Review of Economics and Statistics, vol. 88: 641-658.
- Shikher, S., 2012, ‘Putting Industries in the Eaton-Kortum Model’, Journal of International Trade and Economic Development, vol. 21: 807-837.
- Simonovska, I., and Waugh, M., 2014, ‘The Elasticity of Trade: Estimates and Evidence’, Journal of International Economics, vol. 92: 34-50.
- Teti, F., 2024, Missing Tariffs, CESifo Working Paper No. 11590.

### Policy, institutional, and data sources
- International Monetary Fund, 2024, Annual Report on Exchange Arrangements and Exchange Restrictions 2023: Overview, available at: https://www.imf.org/en/publications/annual-report-on-exchange-arrangements-and-exchange-restrictions/issues/2024/12/19/annual-report-on-exchange-arrangements-and-exchange-restrictions-2023-541890.
- Office of the United States Trade Representative, 2025, USTR Section 301 Action on Nicaragua’s Acts, Policies, and Practices Relating to Labor Rights, Human Rights and Fundamental Freedoms, and the Rule of Law, updated 10 December 2025, available at: https://ustr.gov/about/policy-offices/press-office/press-releases/2025/december/ustr-section-301-action-nicaraguas-acts-policies-and-practices-relating-labor-rights-human-rights
- The White House, 2025a, Joint Statement on Framework for United States – Guatemala Agreement on Reciprocal Trade, updated 13 November 2025, available at: https://www.whitehouse.gov/briefings-statements/2025/11/joint-statement-on-framework-for-united-states-guatemala-agreement-on-reciprocal-trade/
- The White House, 2025b, Joint Statement on Framework for United States – El Salvador Agreement on Reciprocal Trade, updated 13 November 2025, available at: https://www.whitehouse.gov/briefings-statements/2025/11/joint-statement-on-framework-for-united-states-el-salvador-agreement-on-reciprocal-trade/
- The White House, 2026, Fact Sheet: President Donald J. Trump Imposes a Temporary Import Duty to Address Fundamental International Payment Problems, updated 20 February 2026, available at: https://www.whitehouse.gov/fact-sheets/2026/02/fact-sheet-president-donald-j-trump-imposes-a-temporary-import-duty-to-address-fundamental-international-payment-problems/
- World Trade Organization, 2026, WTO-IMF Tariff Tracker, updated 5 May 2026, available at: https://ttd.wto.org/en/analysis/tariff-actions

### Appendix A: Model aggregation (region and country mapping)
- Rest of World is calculated as the sum of the remaining 112 countries in EORA.
- Source: EORA database
- Country list with ISO3 codes (selected entries as in source):
  - United Arab Emirates — ARE
  - Argentina — ARG
  - Australia — AUS
  - Austria — AUT
  - Belgium — BEL
  - Bangladesh — BGD
  - Bahamas, The — BHS
  - Bosnia and Herzegovina — BIH
  - Bolivia — BOL
  - Brazil — BRA
  - Canada — CAN
  - Chile — CHL
  - China — CHN
  - Cote d'Ivoire — CIV
  - Cameroon — CMR
  - Colombia — COL
  - Costa Rica — CRI
  - Czech Republic — CZE
  - Denmark — DNK
  - Dominican Republic — DOM
  - Algeria — DZA
  - Ecuador — ECU
  - Egypt, Arab Rep. — EGY
  - Finland — FIN
  - France — FRA
  - Gabon — GAB
  - Georgia — GEO
  - Guatemala — GTM
  - Honduras — HND
  - Croatia — HRV
  - Hungary — HUN
  - Indonesia — IDN
  - India — IND
  - Ireland — IRL
  - Iran — IRN
  - Iraq — IRQ
  - Israel — ISR
  - Italy — ITA
  - Japan — JPN
  - Germany — DEU
  - Korea, South — KOR
  - Kuwait — KWT
  - Mexico — MEX
  - Malaysia — MYS
  - Netherlands — NLD
  - New Zealand — NZL
  - Norway — NOR
  - Nicaragua — NIC
  - Netherlands — NLD
  - Nicaragua — NIC
  - Nicaragua appears in the list as NIC
  - Pakistan — PAK
  - Panama — PAN
  - Paraguay — PRY
  - Peru — PER
  - Philippines — PHL
  - Poland — POL
  - Portugal — PRT
  - Romania — ROU
  - Russia — RUS
  - Saudi Arabia — SAU
  - El Salvador — SLV
  - Serbia — SRB
  - Slovenia — SVN
  - Spain — ESP
  - Sweden — SWE
  - Switzerland is not listed explicitly in the excerpt
  - Tanzania — TZA
  - Thailand — THA
  - Turkey — TUR
  - Trinidad and Tobago is not listed explicitly in the excerpt
  - United Kingdom — GBR
  - United States — USA
  - Uruguay — URY
  - Vietnam — VNM
  - South Africa — ZAF
  - Nigeria — NGA
  - Niger — NER
  - Jamaica — JAM
  - Jordan — JOR
  - Sri Lanka — LKA
  - Tanzania — TZA
  - Rest of World — ROW

### Appendix A: Industry aggregation (EORA model mapping)
- Source: EORA database
- Industry mappings (selected lines preserved as in source):
  - 1 Agriculture
  - 2 Fishing
  - 3 Mining and Quarrying
  - 4 Food & Beverages  (source label: 3 Food & beverages)
  - 5 Textiles and Wearing Apparel  (source label: 4 Textiles & apparel)
  - 6 Wood and Paper  (source label: 5 Wood & paper)
  - 7 Petroleum, Chemical and Non-Metallic Mineral Products  (source label: 6 Petrol & chemicals)
  - 8 Metal Products  (source label: 7 Metal products)
  - 9 Electrical and Machinery  (source label: 8 Electrical & machinery)
  - 10 Transport Equipment  (source label: 9 Transport equipment)
  - 11 Other Manufacturing  (source label: 10 Other manufacturing)
  - 12 Recycling
  - 13 Electricity, Gas and Water
  - 14 Construction  (source label: 12 Construction)
  - 15 Maintenance and Repair
  - 16 Wholesale Trade
  - 17 Retail Trade
  - 18 Hotels and Restraurants
  - 19 Transport
  - 20 Post and Telecommunications
  - 21 Finacial Intermediation and Business Activities
  - 22 Public Administration
  - 23 Education, Health and Other Services
  - 24 Private Households
  - 25 Others
  - 26 Re-export & Re-import
- EORA / Model mapping lines in source:
  - Agriculture — 1
  - Utilities — 11
  - Services — 13

### Appendix B: Structural gravity estimates (table excerpt and notes)
- Table header: Structural Gravity Estimates (p-values in parentheses)
- Coefficient rows and significance indicators (excerpting layout and values as in source):
  - Row of coefficients (first row): -0.328 *** -0.618 ** -0.385 *** -0.633752 *** -0.368 *** -0.442 *** -0.403 *** -0.400 *** -0.629 *** -0.392 *** -0.298 *** -0.360 *** -0.305 ***
    - Corresponding p-values row: (0.000)(0.010)(0.000)(0.001)(0.000)(0.000)(0.000)(0.000)(0.007)(0.000)(0.000)(0.000)(0.000)
  - Second coefficient row: -4.841 *** -1.916 *** -4.390 *** -1.924744 *** -3.946 *** -3.405 *** -3.604 *** -2.638 *** -3.126 *** -3.337 *** -6.899 *** -7.618 *** -5.828 ***
    - p-values: (0.000)(0.006)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)
  - Third coefficient row: 0.967 *** 1.190 *** 0.860 *** 0.951 *** 0.911 *** 0.937 *** 0.962 *** 1.067 *** 1.072 *** 0.908 *** 0.693 *** 1.814 *** 1.102 ***
    - p-values: (0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)
  - Fourth coefficient row: 0.956 *** 0.705 * 1.017 *** 0.373 0.858 *** 0.904 *** 0.843 *** 0.591 *** 0.623 ** 0.763 *** 1.682 *** 0.226 0.385 **
    - p-values: (0.000)(0.094)(0.000)(0.168)(0.000)(0.000)(0.000)(0.001)(0.038)(0.000)(0.000)(0.442)(0.011)
  - Fifth coefficient row: 0.806 *** 0.643 ** 1.164 *** 0.786 *** 0.928 *** 0.610 *** 0.740 *** 0.428 *** 0.193 * 0.898 *** 1.458 *** 0.812 *** 0.820 ***
    - p-values: (0.000)(0.017)(0.000)(0.000)(0.000)(0.000)(0.000)(0.000)(0.085)(0.000)(0.000)(0.000)(0.000)
  - Sixth coefficient row: 0.512 *** -0.804 *** 0.547 *** 0.376 ** 0.556 *** 0.464 *** 0.336 *** 0.186 0.722 *** 0.119 0.114 0.217 0.671 ***
    - p-values: (0.000)(0.001)(0.000)(0.018)(0.000)(0.000)(0.001)(0.117)(0.000)(0.454)(0.498)(0.162)(0.000)
  - Seventh coefficient row: -7.769 ** -5.626 *** -5.676 *** -10.036 *** -7.352 ** -9.138 *** -15.147 *** -13.378 *** -5.913 ***
    - p-values: (0.012)(0.000)(0.000)(0.000)(0.021)(0.003)(0.000)(0.000)(0.000)
- Regression setup and diagnostics (as in source):
  - Exporter FE — YYYYYYYYYYYYY
  - Importer FE — YYYYYYYYYYYYY
  - R-sqr — 0.9463 0.5555 0.8926 0.7718 0.9576 0.8565 0.9113 0.9295 0.9230 0.8310 0.9989 0.9997 0.9934
  - N. obs — 6084 6084 6084 6084 6084 6084 6084 6084 6084 6084 6084 6084 6084
- Regressors column labels in source (abbreviated): FTA, log(DIST), BRDR, CLEG, CNTG, CLNG, log(1+TARIFF)
- Note from source:
  - DIST is the geographic distance between the source and destination regions, while BRDR, CLEG, CNTG, CLNG and FTA are dummy variables that equal one if the source and destination regions are the same (domestic trade), share a common legal origin, share a common land border, share a common official language, or are members of the same free trade agreement respectively, and zero otherwise.
- Significance legend: * indicates significance at the 10% level, ** indicates significance at the 5% level, *** indicates significance at the 1% level.

*Source: wpiea2026125-source-pdf - References (PDF chapter/section).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2026/english/wpiea2026125-source-pdf.pdf_