## _wp12221

## Source details

**Canonical URL:** [_wp12221](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12221.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12221.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12221.pdf.json)

---

### Introduction: research questions, framework, and scope
- Economic integration as experienced by EMU countries can potentially result in significant changes in specialization patterns.
- Trade-theory motivations:
  - Classical theory: elimination of obstacles to trade leads to greater divergence in productive structures and deepening of comparative advantages.
  - New trade theory: integration encourages economies of scale and is likely to reduce cross-country sectoral specialization.
- Paper objectives:
  - What is the relative competitiveness position of individual euro area countries and China in key economic sectors?
  - How does specialization shape the adjustment to technological progress in Chinese textiles and machinery?
- Modeling approach:
  - Multi-sector Ricardian-gravity trade model (Eaton and Kortum, 2002; Shikher, 2004) with incomplete specialization.
  - Analysis of adjustment under a flex-price-full-mobility model and an immobile labor calibration.
  - Focus sectors: textiles and machinery (China’s largest export sectors).

### Trade and sectoral specialization: China and the euro area (empirical patterns)
- Chinese exports to the euro area outpaced intra-EMU trade since the start of the single currency (Figure 1).
- Evolution of imports from China relative to intra-EMU trade:
  - Textiles: from 20 percent in 1999 to around 80 percent in 2009.
  - Machinery: from 10 percent in 1999 to around 35 percent in 2009.
- By 2007:
  - Textiles and machinery accounted for more than half of all Chinese exports to EMU.
  - Machinery was the largest export sector to the euro area (and to the world) in absolute terms.
- Sectoral specialization remained stable and distinct between 2000 and 2007 (Figure 2). Examples:
  - Textiles concentrated in Portugal, Italy, and Greece.
  - Food concentrated in Ireland, Greece and the Netherlands.
  - Machinery concentrated in Austria, Italy and Germany (transport equipment equally predominant in Germany).
  - Transportation and communication services concentrated in Greece.
  - Non-tradables concentrated in Greece, Portugal, and Spain.

### Ricardian-Gravity Trade Model — framework and key expressions
- Model features and primitives:
  - World economy with I countries and J sectors.
  - Each sector j has a continuum of goods m in (0,1) aggregated by CES with elasticity σ_j.
  - Inputs: labor and capital (capital fully mobile; labor mobile across sectors but immobile across countries).
- Technology distribution (Frechet):
  - F(z) = exp(-T_i,j^(-θ_j) z^(-θ_j)).
  - Parameters T_i,j and θ_j govern scale and dispersion.
- Price and cost formation:
  - Cost of producing good m in sector j from country i delivered to n: c = p q d.
  - Sector price index under Frechet: p_nj = [∑_i T_i,j (c_i,j d_nij)^(-θ_j)]^(-1/θ_j).
- Trade shares:
  - Probabilistic supply (i supplies n in sector j) yields gravity-like expressions augmented with technology.
  - When technological dispersion is limited (high θ_j), trade shares are more elastic to costs and trade barriers.
- Input bundle production: Cobb-Douglas with notation including:
  - β_nj (value added),
  - α_nj (sector j’s share in total final goods spending),
  - σ_n,j,k (sector k’s share of intermediates),
  - ϕ_nj (convolution coefficient).

### Estimating relative competitiveness — estimation strategy and calibration
- Empirical gravity specification (log-linear):
  - Trade barrier proxies in ln(d_nij) include: dist, border, language, RTA, CU, plus exporter fixed effect.
  - Fixed effects A_i,j and A_n,j capture ln(T_i,j^(1/θ_j) / c_i,j) and ln(T_n,j^(1/θ_j) / c_n,j).
- Data and sample (2005):
  - Nine euro area countries: Austria, Belgium, Finland, France, Germany, Greece, Italy, Portugal, Spain.
  - Six additional EU: Czech Republic, Hungary, Poland, Slovenia, Sweden, United Kingdom.
  - Japan, United States, China.
  - Sectors: 11 tradable sectors (9 manufacturing, 2 services) and one non-tradable sector (numeraire).
- Calibration inputs:
  - OECD input-output tables for σ_n,j,k, β_nj, and α_nj (with α_nj set to cross-country sample average).
  - Sector-specific technology dispersion θ_j from Caliendo and Parro (2011); robustness to alternatives tested.
  - Wages w_i from UN INDSTAT compensation per employee.
  - Full capital mobility implies equalized capital costs across countries (r_i = r).
- Competitiveness extraction:
  - Compute competitiveness measure T_i,j c_i,j^(-θ_j) relative to Germany (Germany = 1).
  - Countries placed on “iso-competitiveness curves” (loci of technology and factor cost combinations giving same competitiveness).

### Key empirical findings (2005)
- China:
  - Most competitive in textile industry, followed by nonmetals, food and chemicals.
  - Performs poorly in service sectors.
  - Strong performance primarily stems from low production costs (position on “iso-unemployment curves” leans towards the Y-axis — cost advantage).
- Euro area:
  - Periphery and France fare well in sectors where China is competitive (textiles) and in metals and tradable services.
  - Core (Germany, Finland) most competitive in machinery and transport equipment.
  - Portugal and Greece: competitiveness in textiles mainly driven by low costs.
  - Germany and Finland: machinery competitiveness mainly from superior technology.
- Export barriers (2005):
  - Estimated sector- and country-specific export barriers are low in China and Germany and high in Greece, Portugal and Finland.
  - Color scheme in Figure 6 (as reported): dark green < -1 std. dev.; yellow green -1 std. < -½ std.; yellow - ½ std. < +½ std.; orange ½ std. < 1 std.; red > 1 std. dev.

### Illustrative scenarios — calibration and equilibrium environment
- General equilibrium system solved numerically:
  - Equations include cost-minimizing functions, price indexes, gravity trade shares, goods and factor market clearing (equations (5)–(19)).
  - World GDP assumed constant in simulations.
  - Equilibrium unknowns: J·I prices (p_nj), costs (c_i,j), I wages (w_i), J·I trade shares (π_nij), and J·I goods spending quantities (X_nj).
- Scenario shocks:
  - Sectoral productivity increases by 20 percent in China relative to all other countries.
  - Outcomes reported: GDP, prices, real wages, labor reallocation.
  - Two model variants: Flex-Price-Full-Mobility and Immobile Labor.

### Adjustment under the Flex-Price-Full-Mobility Model (results)
- Mechanism: Countries adjust via nominal wage reductions and labor reallocation to keep trade balances unchanged.
- Textile shock (Chinese textiles productivity +20 percent):
  - Economy-wide nominal wage reductions needed: 0.9 percent in the euro area.
  - Wage declines in textile-exporting countries: 1 ½ percent on average in Portugal, Italy, Spain and Greece.
  - Wage decline in Germany: 0.4 percent.
  - Euro area job losses in manufacturing (textiles): 2 ½ percent of manufacturing employment.
  - By country: textiles job losses range from 7½ percent of manufacturing employment in Portugal to 1.2 percent in Germany.
  - In most countries, textiles’ share in total manufacturing is almost half of its pre-shock level.
- Machinery shock (Chinese machinery productivity +20 percent):
  - Euro area nominal wages decline by 1.3 percent relative to baseline.
  - Countries bearing the adjustment: Germany (1.7 percent wage loss), Austria (1.5 percent), France (1.3 percent).
  - Wage losses less than 1 percent in Portugal, Greece and Belgium.
  - Euro area job losses in manufacturing (machinery): 2.9 percent of manufacturing employment.
  - By country: machinery job losses range from 4.7 percent of manufacturing employment in Finland to 1 ½ percent in Greece.
  - Post-shock machinery output represents around 85–90 percent of its ex ante level.
- Prices and real wages:
  - Nominal wage reductions are largely offset by declines in prices.
  - Real wage gains and losses are within ¼ percent both for textiles and machinery and slightly positive for the euro area as a whole.

### Adjustment with Immobile Labor (results)
- Overall nominal wage and price reductions are similar to the mobile labor case, but:
  - More marked in countries specialized in the shocked industry.
  - Distributional impacts across industries are large:
    - Real wages in the textile industry decline by about 15 percent.
    - Real wages in machinery decline by about 6–8 percent in response to sector-specific productivity shocks.
- Explanation: shocks are absorbed by sectoral wages when labor cannot reallocate, to preserve trade balances.

### Adjustment via higher domestic demand (widening trade deficits)
- If countries do not undergo internal devaluation and labor reallocation, domestic demand expansion can sustain employment and wages temporarily at the cost of deteriorating trade balances.
- Estimated trade deficit changes if domestic demand expands to keep GDP unchanged after the shock:
  - Textiles shock: trade deficits would increase by around 1½ to 3½ percent in Portugal, Italy and Greece, against 1 percent in Germany.
  - Machinery shock: trade deficits deteriorate by more than 2 percent of GDP in Germany and Finland, against less than 1 percent in Greece, Belgium, Portugal, and Spain.

### Conclusion: asymmetry of exposure and policy implications
- Asymmetric exposure:
  - Euro area periphery fares well in sectors where China is most competitive (e.g., textiles).
  - Many core euro area economies are strong performers in industries where China is weaker (e.g., machinery and transport equipment).
- Policy options for member countries absent nominal exchange rate adjustments:
  - Restore competitiveness through internal devaluation and factor reallocation to new sectors.
  - Or face worsening trade deficits if such adjustments do not occur.
- Broader implications:
  - Distribution of gains and losses from trade shocks is asymmetric across countries and sectors, with implications for distributional outcomes over time and fiscal union debates.

*Source: _wp12221*

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

### _wp12221 - References

### Figures
- 1. China’s Exports to the Euro Area ..........................................................................................4
- 2. Sectoral Specialization in Ten Euro Area Countries, 2000 and 2007 ....................................5
- 3. Density Function for Technology ..........................................................................................6
- 4. Production Structure with 2 Countries and 2 Sectors ............................................................9
- 5a. Competitiveness in Nine Euro Area Countries and China (I) ............................................12
- 5b. Competitiveness in Nine Euro Area Countries and China (II) ..........................................13
- 6. Barriers to Export .................................................................................................................14
- 7a. Adjustment to a Productivity Shock in Chinese Textiles ...................................................19
- 7b. Adjustment to a Productivity Shock in Chinese Machinery ..............................................20
- 8. Adjustment via Higher Trade Deficits .................................................................................21

### Introduction: research questions, framework, and scope
- Economic integration as experienced by EMU countries can potentially result in significant changes in specialization patterns.
- Trade theory offers different predictions as to how integration affects specialization:
  - Classical theory suggests that the elimination of obstacles to trade leads to greater divergence in the productive structures and the deepening of comparative advantages.
  - The new trade theory holds that integration encourages the use of economies of scale and is likely to reduce cross-country sectoral specialization.
- One decade later production structures remain diverse.
- The paper investigates how different specialization patterns expose countries asymmetrically to external competitiveness shocks, in particular sector-specific technological progress in China.
- The paper adopts a multi-sector Ricardian-gravity trade model (Eaton and Kortum, 2002; Shikher, 2004) where specialization and trade flows are shaped by technology, factor costs and trade barriers.
- Unlike conventional gravity models (starting with Tinbergen, 1962), which assume complete specialization arising from either differences in factor endowments (Evenett and Keller, 2002) or economies of scale (Krugman, 1991), the framework postulates incomplete specialization.
- The paper uses the model to answer two distinct questions:
  - What is the relative competitiveness position of individual euro area countries and China in key economic sectors?
  - How does specialization shape the adjustment to technological progress in Chinese textiles and machinery?
- The analysis addresses adjustment both in a flex-price-full-mobility model and where these adjustment mechanisms are defective.
- The focus on textiles and machinery is illustrative and justified by the fact that they represent China’s largest export sectors.
- The rest of the paper is organized as follows:
  - Section II describes trade and specialization patterns in euro area countries and China.
  - Section III presents the model that is used to estimate relative competitiveness positions in Section IV.
  - Section V gauges the general equilibrium impact on nine euro area countries of sector-specific technology shocks in China.

*Source: _wp12221 - References*

### Section VI concludes with a number of policy implications.

### _wp12221 - Section VI concludes with a number of policy implications.

### Trade and Sectoral Specialization: China and the euro area
- Chinese exports to the euro area have largely outpaced intra-EMU trade since the start of the single currency (Figure 1).
- By sectors, imports from China (relative to intra-EMU trade) evolved:
  - Textiles: from 20 percent in 1999 to around 80 percent in 2009 (one decade later as reported).
  - Machinery: from 10 percent in 1999 to around 35 percent in 2009 (one decade later as reported).
- By 2007, textiles and machinery accounted for more than half of all Chinese exports to EMU; machinery was the largest export sector to the euro area (and to the world) in absolute terms.
- Sectoral specialization across euro area countries remained remarkably stable and distinct between 2000 and 2007 (Figure 2). Examples:
  - Textiles concentrated in Portugal, Italy, and Greece.
  - Food concentrated in Ireland, Greece and the Netherlands.
  - Machinery concentrated in Austria, Italy and Germany (transport equipment equally predominant in Germany).
  - Transportation and communication services concentrated in Greece.
  - Non-tradables (such as construction) concentrated in Greece, Portugal, and Spain (not reported).

### Ricardian-Gravity Trade Model — framework and key expressions
- Model features:
  - World economy with ܰ countries and ܬ sectors.
  - Each sector j has a continuum of goods ݍ in (0,1).
  - CES aggregation of goods with elasticity ߪ௝.
  - Two inputs: labor and capital (capital fully mobile; labor mobile across sectors but immobile across countries).
- Technology: country i’s technology governed by Frechet cumulative distribution
  - F(z) = exp(-T_i^(-θ_j) z^(-θ_j))  (equation (1) as presented with parameters ܶ௜,௝ and ߠ௝).
  - Parameters ܶ௜,௝ and ߠ௝ shape the distribution’s scale and dispersion.
- Price and production cost formation:
  - Cost of producing good m in sector j from country i delivered to n: c = p q d (equation (2)).
  - Expected foreign vs domestic price ratio linked to relative technology, factor costs and trade barriers (equation (3)).
- Input bundle production: Cobb-Douglas technology (equation (4)).
  - Notation includes shares ߚ௜,௝ (value added), ߜ௜,௝ (labor’s share in value added), ߛ௜,௝௞ (sector k’s share of intermediates), and convolution coefficient ߢ௜,௝.
- Sector price index under Frechet: p_nj = [∑_i T_ij (c_ij d_nij)^(-θ_j)]^(-1/θ_j) (equation (6)).
- Trade shares (probability i supplies n in sector j) follow from Frechet structure and yield a gravity-like expression augmented with technology (equations (7) and (8)).
  - When technological dispersion is limited (high ߠ௝), trade shares become more elastic to costs and trade barriers.

### Estimating relative competitiveness — estimation strategy and calibration
- Empirical gravity specification (log-linear form) linking import shares to relative technology, costs, and trade barriers (equations (9)–(11)):
  - Trade barrier proxies in ln(d_ni,j) include: dist, border, language, RTA, CU, plus exporter fixed effect (equation (10)).
  - Fixed effects ܵ௜,௝ and ܵ௡,௝ capture ln(T_i,j^(1/θ) / c_i,j) and ln(T_n,j^(1/θ) / c_n,j) respectively.
- Data and sample (2005):
  - Countries: nine euro area countries (Austria, Belgium, Finland, France, Germany, Greece, Italy, Portugal, Spain); six additional EU (Czech Republic, Hungary, Poland, Slovenia, Sweden, United Kingdom); Japan, US, China.
  - Sectors: 11 tradable sectors (9 manufacturing, 2 services) and one non-tradable sector (non-tradable acts as numeraire).
- Calibration inputs:
  - Use OECD input-output tables for ߛ௜,௝௞, ߚ௜,௝, and ߜ௜,௝ (with ߜ௜,௝ set to cross-country sample average).
  - Sector-specific technology dispersion ߠ௝ from Caliendo and Parro (2011) (findings robust to alternative ߠ௝).
  - Wages ݓ_i from UN INDSTAT (compensation per employee).
  - Full capital mobility assumption implies equalized capital costs across countries (ݎ_i = ݎ).
- Extraction of technology and cost components:
  - Compute competitiveness measure ܶ௡,௝ ܿ௡,௝ ِ relative to Germany (Germany = 1).
  - Place countries on “iso-competitiveness curves” defined as loci of technology and factor cost combinations giving the same competitiveness.

Key empirical findings (2005, reported in Figures 5a–5b and text):
- China:
  - Most competitive in textile industry, followed by nonmetals, food and chemicals.
  - Performs poorly in service sectors.
  - Strong performance primarily stems from low production costs (position on “iso-unemployment curves” leans towards the Y-axis — i.e., cost advantage).
- Euro area:
  - Periphery and France fare well in sectors where China is competitive (textiles) and in metals and tradable services.
  - Core (Germany, Finland) most competitive in machinery and transport equipment.
  - Portugal and Greece: competitiveness in textiles mainly driven by low costs.
  - Germany and Finland: machinery competitiveness mainly from superior technology.
- Export barriers (2005, Figure 6):
  - Estimated sector- and country-specific export barriers are low in China and Germany and high in Greece, Portugal and Finland (color scheme: dark green < -1 std. dev; yellow green -1 std. < -½ std.; yellow - ½ std. < +½ std.; orange ½ std. < 1 std.; red > 1 std. dev.).

### Illustrative scenarios — calibration and equilibrium environment
- General equilibrium setup:
  - Solve system comprising cost-minimizing functions (5), price indexes (6), gravity trade shares (8), goods and factor market clearing (equations (13)–(19)).
  - World GDP is assumed constant in simulations.
  - Equilibrium unknowns: ܬ·ܰ prices (p_nj), costs (c_i,j), ܰ wages (w_i), ܬܰܰ trade shares (π_nij), and ܬܰ goods spending quantities (X_nj). The system determines ܬܰܰ·ܰ·ܬܰ3 unknowns with same number of equations.
- Scenario focus:
  - Use calibrated model to gauge impacts on individual euro area countries of technological progress in China’s two largest export sectors: textiles and machinery.
  - The general equilibrium is characterized by market clearing in all countries and sectors and solved numerically based on the model structure presented.

*Source: _wp12221 - Section VI concludes with a number of policy implications.*

### Box 1: Definition of Variables Characterizing the General Equilibrium

### Box 1: Definition of Variables Characterizing the General Equilibrium

### Definitions of core variables
- X_n: Country n’s total spending on final goods (all sectors).
- X_nj: Country n’s total expenditure of sector j goods. This consists of two terms:
  - α_nj X_n: Country n’s requirements of sector j goods for final demand, with α_nj sector j’s share in total final goods spending.
  - p_nj ∑_k a_njk Y_nk: Country n’s requirements of sector j goods for intermediate consumption (production of input bundles in sectors, k = 1,2,...,S), with p_nj the price index for sector j in country n.
- Y_nj: Country n’s gross production (for final goods and intermediate inputs) of sector j goods.
- B_nj ≡ X_nj − Y_nj: Country n’s trade deficit in sector j.
- Y_n ≡ ∑_j β_nj Y_nj: Country n’s GDP, with β_nj the value added share in sector j.
- B_n ≡ ∑_j B_nj: Overall trade deficit of country n, and B_n ≡ X_n − Y_n.

### Key algebraic relationships referenced
- Equation (20) and (21) (as presented in the source) express how sectoral demand and gross production relate to changes in expenditures, prices, and trade shares; these equations show that pre-shock production can be sustained provided domestic absorption increases sufficiently to offset initial deterioration in demand associated with a technology shock and a worsening trade deficit B_n.

### Notation remark
- A counterfactual notation is used where x^ denotes the counterfactual value, x̂, relative to the original value, x.

---

### Impact of Trade Shocks on Euro Area Countries: Illustrative Scenarios

### Scenario setup
- Shocks assume sectoral productivity increases by 20 percent relative to all other countries.
- Outcomes quantified: GDP, prices, real wages, labor reallocation.
- Two model variants:
  - Flex-Price-Full-Mobility model (labor fully mobile, wages flexible).
  - Immobile Labor calibration (labor does not move across sectors).

### Adjustment under the Flex-Price-Full-Mobility Model
- Mechanism:
  - Countries adjust to increased competition by reducing wages and reallocating labor to ensure trade balances remain unchanged at pre-shock levels.
- Textile shock (Chinese textiles productivity +20 percent):
  - Economy-wide nominal wage reductions needed: 0.9 percent in the euro area.
  - Wage declines in textile-exporting countries: 1 ½ percent on average in Portugal, Italy, Spain and Greece.
  - Wage decline in Germany: 0.4 percent.
  - Euro area job losses in manufacturing (textiles): 2 ½ percent of manufacturing employment.
  - By country: textiles job losses range from 7½ percent of manufacturing employment in Portugal to 1.2 percent in Germany.
  - In most countries, textiles’ share in total manufacturing is almost half of its pre-shock level.
- Machinery shock (Chinese machinery productivity +20 percent):
  - Euro area nominal wages decline by 1.3 percent relative to baseline.
  - Countries bearing the adjustment: Germany (1.7 percent wage loss), Austria (1.5 percent), France (1.3 percent).
  - Wage losses less than 1 percent in Portugal, Greece and Belgium.
  - Euro area job losses in manufacturing (machinery): 2.9 percent of manufacturing employment.
  - By country: machinery job losses range from 4.7 percent of manufacturing employment in Finland to 1 ½ percent in Greece.
  - Post-shock machinery output represents around 85–90 percent of its ex ante level.
- Prices and real wages:
  - In the flex-price-full-mobility model, nominal wage reductions are largely offset by declines in prices.
  - Real wage gains and losses are within ¼ percent both for textiles and machinery and slightly positive for the euro area as a whole.

### Immobile Labor
- When labor is immobile, overall nominal wage and price reductions are similar to the mobile labor case, more marked in countries specialized in the shocked industry.
- Distributional impacts between industries are large:
  - Real wages in the textile industry decline by about 15 percent.
  - Real wages in machinery decline by about 6–8 percent in response to sector-specific productivity shocks.
- Explanation: shocks are fully absorbed by sectoral wages to preserve trade balances when labor cannot reallocate.

### Widening Trade Deficits (Adjustment via higher domestic demand)
- Sector-specific productivity shocks increase Chinese trade shares since countries purchase goods from the lowest cost provider.
- If countries do not undergo internal devaluation and labor reallocation, a boom in domestic demand could temporarily sustain employment and wage levels at the cost of deteriorating trade balances.
- Estimated trade deficit changes if domestic demand expands to keep GDP unchanged after the shock:
  - Textiles shock: trade deficits would increase by around 1½ to 3½ percent in Portugal, Italy and Greece, against 1 percent in Germany.
  - Machinery shock: trade deficits deteriorate by more than 2 percent of GDP in Germany and Finland, against less than 1 percent in Greece, Belgium, Portugal, and Spain.

---

### Conclusion: Asymmetry of Exposure and Policy Implications
- The Ricardian-gravity trade model indicates asymmetric exposure of euro area countries to external competition due to different specialization patterns.
- Findings:
  - Euro area periphery fares well in sectors where China is most competitive (e.g., textiles).
  - Many core euro area economies are strong performers in industries where China is weaker (e.g., machinery and transport equipment).
- Policy options for member countries absent nominal exchange rate adjustments:
  - Restore competitiveness through internal devaluation and factor reallocation to new sectors.
  - Or face worsening trade deficits if such adjustments do not occur.
- The distribution of gains and losses from trade shocks is asymmetric across countries and sectors, raising questions about the distributional implications over time and relevance to broader fiscal union debates.

---

### Appendix: Sample, Data, and Calibration Notes

### Sample
- Nine euro area members: Austria, Belgium, Finland, France, Germany, Greece, Italy, Portugal, Spain.
- Six EU countries: Czech Republic, Hungary, Poland, Slovenia, Sweden, United Kingdom.
- Additional countries: United States, Japan, China.
- Exclusions: Cyprus, Latvia, Lithuania, Malta (lack of GTAP data for two service sectors); Denmark, Estonia, Ireland, Luxembourg, Netherlands, Slovakia (measured export larger than gross production in some sectors leading to negative home expenditure values X_nn,j).

### Data sources and mapping
- Trade data: UN COMTRADE database for manufacturing industries at 4-digit SITC; concordance to 2-digit ISIC from WITS.
- Wages, value added, gross production: UNIDO’s INDSTAT2 database at 2-digit ISIC.
- Service sectors: GTAP 2004-database for bilateral trade flows and production.
- Bilateral trade shares for two service sectors: ISIC classification combined with actual gross production in 2005 from OECD.
- Input-output shares (αs, βs, γs, σs): OECD input-output tables for 2005.
- Bilateral distances, common language, shared border: CEPII database.
- Trade elasticities (θ):
  - Benchmark relies on estimates from Caliendo and Parro (2011) where available.
  - For “Chemicals, rubber, plastics and fuel products” elasticity for Chemicals used.
  - For machinery and transport equipment: overall manufacturing estimate used.
  - Goods in two service sectors assumed relatively inelastic; elasticity chosen as 2.
  - Simulations experimented with alternative values for θ (lower and median estimates of Eaton and Kortum (2002)); ranking of countries’ exposure is robust to these alternatives.

### Table A1: Sectors (short names, ISIC codes, and a numeric column as presented)
- Food: ISIC 15-16 — 2.62
- Textiles: ISIC 17-19 — 8.10
- Wood: ISIC 20 — 11.50
- Paper: ISIC 21 — 16.52
- Chemicals: ISIC 23-25 — 3.13
- Non-metals: ISIC 26 — 2.41
- Metals: ISIC 27-28 — 6.99
- Machinery: ISIC 29-33 — 9.29
- Transport equipment: ISIC 34-35 — 9.29
- Transp. & Comm. Serv.: ISIC 50-52, 55, 60-64 — 2.00
- Financial and other serv.: ISIC 65-67, 70-74, 90-93 — 2.00
- Non-tradables: All other — NA

*Source: IMF Working Paper content (Box 1 and associated sections).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp12221.pdf_
