## ttgvcsieea

## Source details

**Canonical URL:** [ttgvcsieea](https://www.imf.org/-/media/files/publications/dp/2019/english/ttgvcsieea.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/dp/2019/english/ttgvcsieea.pdf.md)
- [Structured JSON version](/-/media/files/publications/dp/2019/english/ttgvcsieea.pdf.json)

---

### Executive Summary — Overview
- Europe is deeply integrated into global value chains (GVCs), both within and outside the region.
- About 70 percent of total European exports are linked to forward and backward supply chains.
- Shocks affecting trade flows between the major trade hubs—the United States, China, and Germany—could affect European economies through those supply chain linkages.
- For Europe, the difference between value-added and gross trade statistics is larger than for Asia or the Americas.

### Methodology (databases and analytical frameworks)
- Databases used:
  - EORA Multi-Regional Input-Output (MRIO) (latest vintages).
  - World Input-Output (WIOD) (latest vintages).
- Three complementary analytical frameworks:
  - Input-output accounting exercise to trace short-term tariff-related shocks through supply-chain networks (Chapter 3).
  - Demand-system framework linking changes in demand for value added to changes in value-added prices and final expenditure levels, using a value-added REER index (Chapter 4).
  - Two-step macroeconometric approach:
    - Step 1: estimate country-specific growth shocks as residual growth after purging common shocks (PCA).
    - Step 2: trace spillovers to other economies using local projections, exploring nonlinearities associated with trade exposure in value-added terms (Chapter 5).

### Key findings — Tariff scenarios and value-added vs gross trade
- Distribution of export losses differs significantly when measured by value added versus gross trade.
  - Scenario: a 25 percent tariff by the United States on imports of cars and car parts.
    - About half of the impact is transmitted directly to the affected sector-country and the rest via supply chains.
    - Negative impact would spread across more countries when measured using value-added indicators than when using gross exports.
    - The order of export losses does not necessarily reflect the order of gross exports.
  - Short-term losses in the 25 percent car tariff scenario average 0.1 of EU GDP.
- Hub importance differs by measurement:
  - Germany’s importance for its European trading partners in the REER is about 10 percent lower in value-added terms than in terms of gross exports.
  - The importance of the United States and China in European countries’ REER is almost 50 percent larger in value-added terms.
- Price and tariff simulations:
  - Simulation: a 5 percent general tariff on all US imports (with retaliation from the rest of the countries).
    - The average reduction in the demand for Europe’s value-added output would be 50 percent higher when measured in value-added terms than if estimated in gross terms.

### Growth spillovers (macro-econometric findings)
- Two-step estimation setup:
  - PCA on GDP (quarter-over-quarter) growth for 68 economies during 1995:Q1–2018:Q3; three principal components explain about 28 percent of variance; first PC about 21 percent.
  - Local projections trace effects of orthogonalized country-specific growth shocks, allowing interaction with value-added export exposure.
- Exact empirical spillover estimates (1 percent negative country-specific growth shock):
  - United States:
    - Median European economy → peak cumulative growth impact over a two-year horizon of about –0.6 percentage point.
    - High-exposure (90th percentile) European economy → about –1.4 percentage points.
    - High-exposure advanced Europe → about –1.6 percentage points.
  - China:
    - Spillovers sizable but smaller than those from the United States (numerical peaks not restated).
  - Germany:
    - For the median European economy, a 1 percent negative German growth shock → effects not statistically significant in the baseline (orthogonalized shocks).
    - For a high-exposure European economy, peak cumulative impact around –0.1 percentage point (baseline).
    - Using actual German growth (without purging common shocks), a 1 percent negative German growth → peak cumulative impact on the median European economy about –0.7 percentage point.
- Advanced vs emerging Europe:
  - Spillovers from the United States are smaller for emerging Europe than for advanced Europe, reflecting smaller exposure in value-added terms.
  - For a high-exposure economy, the peak cumulative impact due to a 1 percent negative US growth shock is about –1.6 percentage points in advanced Europe and about half that for a high-exposure emerging market European economy.

### Stylized facts and sectoral patterns
- GVC participation and regional shares:
  - Between 2000 and 2013, share of European exports involved in GVC trade rose from about 60 to 70 percent of total exports of goods and services.
  - Over the same period, increases were 2 percentage points in the Americas and 6 percentage points in Asia.
  - Almost half of world trade occurs in Europe, about 30 percent in Asia, and 20 percent in the Americas.
- Gross vs value‑added measures:
  - In Europe, gross exports are 18 percentage points of GDP higher than value-added exports; the difference is 9 percentage points for Asia and 3 percentage points for the Americas.
  - About two-thirds of Europe’s value-added exports go to regional partners and one-third go outside the region.
  - When measured in value-added terms, exports destined outside of Europe are almost as large as those of Asia and larger than those of the Americas.
- Sector concentration:
  - GVC participation concentrated in manufacturing, transport and communications, and wholesale and retail.
  - Forward and backward linkages of Europe’s manufacturing sector represent almost 30 percent of total exports.
  - Manufacture of motor vehicles represents about 4 percent of total value-added exports.
- Employment:
  - Wholesale and retail trade employed nearly 50 million people in 2014 (about 20 percent of total employment).
  - Transport and communications employed more than 20 million workers.

### Box 2 highlights — Robotization, automation, and GVCs
- Rising presence of industrial robots since the mid-1990s boosted productivity and increased domestic value-added content in exports.
- Positive relationship between number of robots per thousand workers and value-added exports strengthened in Europe between 2000 and 2013.
- Relationship stronger in CESEE than in advanced economies; in advanced economies relationship weakened at end of 2000s as robot penetration growth slowed.
- Implication: Intensifying robotics, AI, and automation may shorten and localize GVCs if automation reduces offshoring incentives and intermediate goods are increasingly sourced domestically.

### Illustrative network analysis — US 25 percent tariff on cars and car parts (Box 2 empirical case)
- Aggregate context:
  - Total exports of cars and car parts to the United States in 2017: Total: 228 billion USD (top 10 exporters account for 99 percent of exports).
- Domestic value‑added shares in car exports:
  - Typical domestic value-added share across exporters about 40–70 percent; Slovak Republic and Hungary at 13 percent.
  - Czech Republic about 50 percent.
  - About a third of the value of German car exports attributable to other countries in its supply chain.
- Impacts in the 25 percent tariff scenario:
  - Gross exports of cars and car parts from the EU to the US: roughly 0.3 percent of EU GDP.
  - Mapping export losses through supply chains yields estimated impact on the EU of 0.1 percent of GDP.
  - Estimated impact on Germany (most affected): about 0.15 percent of GDP; about half of this impact related to supply chains.
  - Distributional result: Czech Republic appears as the 4th country potentially most affected when supply chains are fully considered, despite low gross exports to the US; Slovak Republic exports $1,600 million gross but domestic value-added share only 13 percent; Czech Republic exports $156 million gross but domestic value-added share about 50 percent.
- Caveats:
  - Short-term estimation abstracts from policy responses, confidence effects, nonlinearities, trade diversion, and disaggregated elasticities.

### Tariff simulation — 5 percent US tariff on all imports (with global retaliation)
- Model, data, and assumptions:
  - Uses Matlab code from Bems and Johnson (2017) and 2016 vintage of WIOD for 43 countries (2000–2014).
  - Assumes full pass-through of a country’s own tariff to its own prices.
  - Omits trade diversion and holds supply fixed (short-run partial equilibrium).
- Aggregate impacts (Europe):
  - Average decrease in demand for European value added: about 0.2 percent.
  - Impact in terms of demand for gross output: 0.1 percent.
  - Decline in demand for Europe’s output is 50 percent higher when measured in value-added terms than in gross terms.
- Heterogeneity:
  - Difference between value-added and gross impacts three times larger for emerging Europe than for advanced Europe.
  - Around German value chain:
    - Decline twice as large in value-added terms as gross terms for Germany.
    - Difference three times larger for Poland.
    - Difference four times larger for Czech Republic, Slovenia, and Slovak Republic.
  - Luxembourg (financial services exporter) among least affected; short-run effect estimated positive for Luxembourg.

### VA-REER vs conventional REER and effective EoS insights
- Determinants of effective elasticity of substitution (EoS):
  - Relative size of EoS for final and intermediate goods.
  - Relative importance of final and intermediate trade in total trade flows.
  - Size of market shares in value-added terms.
- Assumptions in analysis:
  - Leontief production (production elasticities zero).
  - Nested CES demand with final demand EoS of 3 and production elasticities zero.
- Empirical ranges and patterns:
  - Effective EoS varies between 0.3 and 1.2 across a sample of 32 European countries.
  - Effective EoS lower for countries trading relatively more in intermediate goods.
  - Large and more distant economies (China, United States, Japan, UK) display larger EoS.
- VA-REER vs GO-REER:
  - VA-REER and GO-REER comove, but cumulative gaps emerge; VA-REER shows more real appreciation for most countries.
  - Gaps peaked in 2008 at almost 10 percentage points on average.
  - In 2014, European countries remained on average 6 percent less competitive in value-added terms relative to 2000.
  - Standard deviation of VA-REER index was 45 percent larger in 2014 than GO-REER.

### Country cases — Slovak Republic (selected insights)
- Short-term impact of 25 percent US tariff on cars and parts:
  - Participation in integrated European supply chains can reduce direct competitive exposure measured in gross-trade terms.
- Normalization and VA-REER approach:
  - Gross exports divided by gross turnover (GO); value-added exports divided by total value added (GDP).
  - VA-REER uses GDP deflator rather than CPI.
- Key patterns:
  - Large-but-remote economies (United States and China) become more important in value-added terms than conventional terms.
  - Conventional weights of the largest trading partner were 16 percent larger than value-added weights in 2000 and 32 percent larger in 2014.
  - Germany’s weight in VA-REERs about 10 percent lower than in gross trade terms for Slovakia, Poland, and Hungary.
- Policy implications:
  - Use value-added measures to assess exposure and spillovers.
  - Recognize intermediate-input linkages can make some partners complements rather than substitutes.
  - Account for heterogeneous elasticities of substitution across countries.

### Annex methods — Network accounting and demand modeling
- Network accounting for tariff τ on country c, sector s exports (exp_cs):
  - Loss in exports assumed α × τ × exp_cs; α calibrated (10 percent tariff → decline in demand of 8 percent nominally; calibration 8 percent).
  - Loss to country c’s domestic value added: (DVA_exp_cs) × α × τ × exp_cs.
  - Remaining loss: (1 − DVA_exp_cs) × α × τ × exp_cs, traced to foreign country‑sectors.
  - Exercise repeated to capture direct and indirect losses across countries and sectors.
- Demand-model specification highlights:
  - Two good types (value added V_i and CES composite intermediate X_i), three elasticities γ, ρ, σ.
  - Final goods follow Armington CES with elasticity σ; empirical support suggests σ > γ, ρ.
  - Value-added price change characterization: p^_VA = [1 − Ω] [diag(s_i^VA)]^−1 p^_GO.
  - Retaliation modeled; retaliation impact about 10 times smaller than initial US tariff impact on average (given US size and openness).
  - Sensitivity: different pass-through rescales effects linearly.

### Annex methods — Two-step spillover estimation (exact specification)
- Step 1: Principal components analysis purging common shocks from quarter-over-quarter GDP growth for 68 economies (1995:Q1–2018:Q3); country-specific shocks = residuals after removing three principal components.
- Step 2: Local projections:
  - y_{i,t+h} = u_{i,h} + β_h U_t + θ_h (U_t × GVC_{i,t−1}) + ψ_h(L) y_{i,t−1} + X_{i,t} Γ_h + ε_{i,t+h}.
  - GVC_{i,t} normalized as GVC_{i,t} = (gvc_{i,t} − gvc_{50}) / (gvc_{90} − gvc_{50}).
  - Marginal impact for median economy = β_h; for high-exposure economy (90th percentile) = β_h + θ_h.
  - Controls include own lagged growth, oil prices, global uncertainty (stock market volatility index), and country fixed effects; Driscoll and Kraay standard errors for inference; two standard deviation confidence bands reported.

### Key statistics — Value‑Added Exports (Percent of GDP) (Annex Table 3.1)
- Median: United States 2.50; China 0.7; Germany 2.3
- High-exposure (90th percentile): United States 4.1; China 1.4; Germany 5.4
- Select country examples (United States | China | Germany):
  - Austria: 3.3 | 1.4 | 5.4
  - Belgium: 4.4 | 1.6 | 5.0
  - Czech Republic: 4.0 | 1.3 | 11.7
  - Ireland: 5.2 | 0.7 | 2.3
  - Netherlands: 3.7 | 1.2 | 2.6
  - Switzerland: 5.7 | 2.3 | 5.3
  - Israel: 6.4 | 0.7 | 2.4
  - Moldova: 0.0 | 0.0 | 0.0
- Source for table: EORA database and IMF staff calculations.

### Policy implications and recommendations
- Measuring trade and spillovers in value-added terms gives a more accurate assessment of distribution of costs across countries and sectors.
- Understanding propagation through GVCs helps policymakers:
  - Quantify more precisely the impact of growing trade tensions.
  - Calibrate countercyclical measures as needed.
  - Target social policies to citizens most likely to be affected.
- Practical guidance:
  - Use VA-REER and value-added exposure metrics for competitiveness and vulnerability analysis, particularly for economies participating in deep supply chains.
  - Account for heterogeneous elasticities of substitution and the possibility that some partners act as complements due to intermediate-input linkages.
  - Consider short-run network accounting as a first-round estimate but complement with models that capture policy responses, nonlinearities, and trade diversion for larger disruptions.

*Source: Executive Summary, ttgvcsieea - Executive Summary*

### Executive Summary ������������������������������������������������������������������������������������������������������

### Executive Summary

### Overview
- Europe is deeply integrated into global value chains, both within and outside the region.
- About 70 percent of total European exports are linked to forward and backward supply chains.
- Shocks affecting trade flows between the major trade hubs—the United States, China, and Germany—could affect European economies through those supply chain linkages.
- For Europe, the difference between value-added and gross trade statistics is larger than for Asia or the Americas.

### Methodology
- Databases used: the latest vintages of the EORA Multi-Regional Input-Output (MRIO) and World Input-Output (WIOD) databases.
- Three complementary analytical frameworks:
  - A simple accounting exercise based on input-output tables to trace short-term tariff-related shocks through supply chain networks (Chapter 3).
  - A demand-system framework linking changes in the demand for value added to changes in value-added prices and final expenditure levels, using a value-added REER index (Chapter 4).
  - A two-step macroeconometric approach: (1) estimate country-specific growth shocks as residual growth after purging common shocks; (2) trace spillovers to other economies using local projections, exploring nonlinearities associated with trade exposure in value-added terms (Chapter 5).

### Key Findings
- Distribution of export losses differs significantly when measured by value added versus gross trade.
  - Scenario: a 25 percent tariff by the United States on imports of cars and car parts.
    - About half of the impact is transmitted directly to the affected sector-country and the rest via supply chains.
    - The negative impact would spread across more countries when measured using value-added indicators than when using gross exports.
    - The order of export losses does not necessarily reflect the order of gross exports.
  - Short-term losses in the 25 percent car tariff scenario average 0.1 of EU GDP.
- Hub importance differs by measurement:
  - Germany’s importance for its European trading partners in the REER is about 10 percent lower in value-added terms than in terms of gross exports.
  - The importance of the United States and China in European countries’ REER is almost 50 percent larger in value-added terms.
- Price and tariff simulations:
  - Simulation: a 5 percent general tariff on all US imports (with retaliation from the rest of the countries).
    - The average reduction in the demand for Europe’s value-added output would be 50 percent higher when measured in value-added terms than if estimated in gross terms.
- Growth spillovers:
  - Spillover effects from United States and China to European economies are sizable and larger for economies more exposed to them in terms of value-added exports.
  - Spillovers from German growth shocks on European countries are estimated to be much smaller, likely reflecting Germany’s smaller size relative to the United States and China and Germany’s openness.
  - The analysis does not rule out Germany as a transmitter of shocks originating elsewhere or its potentially larger spillovers if its growth becomes more driven by domestic demand.

### Additional Stylized Facts
- About 70 percent of total European exports are linked to forward and backward supply chains, compared with roughly 40 and 45 percent of total exports from the Americas and Asia, respectively.
- Europe’s global value chain participation is concentrated in a few sectors: manufacturing, transport and communications, and wholesale and retail.
- Conventional trade indicators can be misleading:
  - Example: Gross exports would suggest the Slovak Republic’s exposure to the United States is much larger than the Czech Republic’s for cars and car parts; value-added exports show the opposite because Czech value added is embedded in other countries’ exports to the United States.
  - Germany’s share of trade in value added is significantly smaller than gross measures indicate.

### Policy Implications
- Measuring trade and spillovers in value-added terms gives a more accurate assessment of the distribution of costs across countries and sectors.
- Understanding how trade shocks propagate through global value chains can help policymakers:
  - Quantify more precisely the impact of growing trade tensions.
  - Calibrate countercyclical measures as needed.
  - Target social policies aimed at the citizens most likely to be affected.

*Source: Executive Summary, ttgvcsieea - Executive Summary*

### Introduction

### ttgvcsieea - Introduction

### Trade shocks and growth spillovers
- A 5 percent US tariff on all its imports (with retaliation from the rest) yields an average reduction in the demand for Europe’s value-added output of 0.2 percent.
- Growth spillovers from the United States and China to European economies are sizable:
  - For the median European economy, a 1 percent negative US growth shock would have a peak cumulative growth impact during a two-year horizon of about –0.6 percentage point.
  - For countries most exposed to the United States, the estimated impact would be –1.4 percentage points.
- Within Europe, growth spillovers originating in the United States would be smaller for emerging Europe than for advanced Europe, reflecting their smaller exposure in value-added terms.
- The effect of German growth shocks on European countries is estimated to be much smaller.

### Global value chains (GVCs): concepts and measurement
- Since the early 1990s, global value chains have become the center of global production and trade.
- The value chain encompasses activities from conception to end use, including design, production, marketing, distribution, and support to the final consumer.
- Traditional gross trade statistics mask embodied value added created in other countries and sectors; measuring value-added flows is crucial to analyze trade shocks.
- Decomposition of gross exports follows Koopman, Wang, and Wei (2014) and Aslam, Novta, and Rodrigues-Bastos (2017), splitting gross exports into:
  - Foreign value added embedded in gross exports (backward linkages).
  - Domestic value added in exports, further decomposed into:
    - Exported as final goods.
    - Exported as intermediates absorbed in destination (value-added exports).
    - Exported as intermediates further re-exported to third countries (forward linkages).
    - Exported as intermediates that return to home country.

### Indicators used
- Value-added exports indicator = share of domestic value added exported in final goods plus share of value added in intermediate goods absorbed in the country of destination (reflects value added consumed abroad).
- Global value chain participation indicator = sum of forward and backward linkages (measure of openness and integration into GVCs).

### Data sources and choices
- Databases referenced: Trade in Value-Added Statistics (TiVA) covering 64 countries and 36 sectors; World Input-Output Database (WIOD) covering 43 countries and 56 sectors; EORA MRIO covering 190 countries and 26 sectors.
- This paper uses EORA MRIO for broad country coverage (especially for European countries) and WIOD when more sectors provide more accurate input-output linkage estimates or for consistency with existing studies (EORA used in Chapter 3, WIOD in Chapter 4).

### Stylized facts: Europe and global value chains
- Europe is more deeply integrated into global value chains than the Americas and Asia:
  - Between 2000 and 2013, the share of European exports involved in GVC trade increased 10 percentage points, from about 60 to 70 percent of total exports of goods and services.
  - Over the same period, the ratio increased 2 percentage points in the Americas and 6 percentage points in Asia.
- Regional clustering and differences:
  - Almost half of world trade occurs in Europe, about 30 percent in Asia, and 20 percent in the Americas.
  - In Europe, gross exports are 18 percentage points of GDP higher than value-added exports; the difference is 9 percentage points for Asia and 3 percentage points for the Americas.
  - About two-thirds of Europe’s value-added exports go to regional partners and one-third go outside the region.
  - When measured in value-added terms, exports destined outside of Europe are almost as large as those of Asia and larger than those of the Americas.
- Exposure to trade hub countries:
  - Exports of other European countries to Germany are 8.3 percent of GDP in gross terms, but only 2.9 percent of GDP in value-added terms.
  - European exports to the United States are 2.7 percent of GDP in both gross and value-added terms.
  - Advanced European economies have larger value-added exports to the United States, China, and Germany than emerging European economies.

### GVC participation, income, and productivity (Box 1 summary)
- Global value chain participation is positively correlated with income per capita.
- Econometric findings (Ignatenko, Raei, and Mircheva (2019)):
  - Share of intermediate trade in GDP is positively and significantly associated with log(Income per capita) (coefficient 1.610*** with standard error 0.339 in column (1)).
  - Share of GVC-trade in GDP shows positive associations with log(Income per capita) and with human capital and productivity in various specifications (e.g., coefficient 2.738** with standard error 1.148 in column (2); coefficient 0.367*** with standard error 0.103 in another specification).
  - Results differ when restricting the sample to smaller country sets available in other databases, underscoring the importance of larger EORA coverage.
- Channels: GVC-related trade flows positively affect investment, human capital, and productivity measures.

### Sectoral concentration and labor (Box 2 summary)
- Most of Europe’s GVC participation and jobs are concentrated in a few sectors:
  - Manufacturing, transport and communications, and wholesale and retail trade dominate GVC integration.
  - Forward and backward linkages of Europe’s manufacturing sector represent almost 30 percent of total exports.
  - Within manufacturing, machinery and equipment and motor vehicles are the subsectors with the highest percentage of value-added exports; manufacture of motor vehicles represents about 4 percent of total value-added exports.
  - Transport and communications account for about 10 percent and wholesale and retail trade about 5 percent of total exports (by sector shares noted).
- Employment:
  - Wholesale and retail trade employed nearly 50 million people in 2014 (about 20 percent of total employment).
  - Manufacturing employed a similar large share.
  - Transport and communications employed more than 20 million workers.
- Services vs. manufacturing:
  - “Servification” of manufacturing means a portion of manufacturing exports is value added generated by the service sector.
  - Services (business and financial) exhibit more forward linkages; manufacturing exhibits more backward linkages.

*Source: https://www.imf.org/-/media/files/publications/dp/2019/english/ttgvcsieea.pdf*

### Box 2. Europe’s Global Value Chain Participation

### Box 2. Europe’s Global Value Chain Participation

### Europe’s GVC participation and employment (stylized facts)
- Sectoral GVC participation and employment are presented for 2014 (figures in source).
- Manufacturing is a major sector in both GVC linkages and employment (figures in source).

### Automation and participation in Global Value Chains
- Observation: The increasing presence of industrial robots in Europe’s factories since the mid-1990s has boosted productivity and raised the amount of domestic value-added content in final and intermediate exports going to other countries.
- Empirical pattern:
  - The positive relationship between value-added exports and the number of robots per thousand workers strengthened in Europe between 2000 and 2013 (Box Figure 3.1).
  - The positive relationship between robotization and value-added exports is more apparent in central, eastern, and southeastern European countries (CESEE) than in advanced economies, with the relationship in advanced economies weakening at the end of the 2000s as robot penetration growth slowed (Box Figure 3.2).
- Projection/implication: As robotics, artificial intelligence, and automation intensify, global value chains may become less extensive and shorter if automation reduces incentives to offshore and intermediate goods are increasingly sourced domestically in advanced economies.

### Short-term impact of tariff shocks in a supply-chain network (methodology and calibration)
- Analytical approach: Network analysis is used to capture direct and indirect links in GVCs—both the number of linkages and the volume of trade in value added embodied in those links—to estimate short-term output impacts of trade shocks.
- Scenario considered: The United States raises the tariff (τ) on imports of cars and car parts to 25 percent.
- Loss accounting:
  - A tariff of τ percent on exports of country c, sector s (exp_cs) implies a loss of α × τ × exp_cs in exports, where α is based on import elasticities.
  - The part of this loss borne by the exporting country equals (DVA_exp_cs) × α × τ × exp_cs, where DVA_exp_cs is the share of domestic value added in the exports of cars and car parts.
  - The remainder, (1 − DVA_exp_cs) × α × τ × exp_cs, accrues to the set of foreign country-sectors that comprise the foreign-value-added component of exp_cs.
- Import-elasticity calibration:
  - Calibrated α: a 10 percent tariff on imports implies a decline in demand of 8 percent in nominal terms (calibration: 8 percent).
  - The 8 percent calibration is positioned as around the midpoint of the literature range of 0 to 15 percent.
  - The analysis notes that losses scale linearly with the choice of elasticity and that country- and product-specific elasticities could differ.

### Channels and amplification under GVC linkages
- Direct domestic impact: lower demand for the directly targeted sector in the tariffed destination and on domestic suppliers through domestic supply chains.
- Indirect foreign impact: lower demand for foreign suppliers that provide inputs to the targeted sector, including their upstream suppliers.
- Nonlinear amplification: In multi-country value chains where value-added travels between countries and back, nonlinear cascading effects can generate larger reductions in demand and output than a simple gross-exports view would suggest. The current exercise abstracts from these nonlinearities.

### Empirical case: US car and car-parts tariffs (findings)
- Aggregate export context:
  - Total exports of cars and car parts to the United States in 2017: Total: 228 billion USD (top 10 exporters account for 99 percent of exports).
  - Major exporters to the United States include Canada, Mexico, Japan, and Germany.
- Domestic value-added shares in car exports:
  - Domestic value-added share across exporters generally about 40–70 percent, with large cross-country variation.
  - Slovak Republic and Hungary have particularly low domestic value-added shares, at 13 percent.
  - Czech Republic has about 50 percent value-added content in its car exports to the United States.
  - About a third of the value of German car exports can be attributed to other countries in its supply chain.
- Geographic contributors to Germany’s foreign value added in auto exports include: other older EU members (EU-15), China, the United States, and the newer EU members (NMS-11).
- Impacts in the illustrative 25 percent tariff scenario:
  - Gross exports of cars and car parts from the European Union to the United States: roughly 0.3 percent of EU GDP.
  - Mapping export losses through supply chains yields an estimated impact on the European Union of 0.1 percent of GDP.
  - Estimated impact on Germany (most affected country): about 0.15 percent of GDP; about half of this impact would be related to supply chains.
  - Distributional findings:
    - The Czech Republic does not appear among top exporters by gross exports to the United States but appears as the 4th country potentially most affected when supply chains are fully considered because of Czech value added embodied in other countries’ exports.
    - The Slovak Republic exports $1,600 million in cars and car parts to the United States (gross), but its domestic value-added share in those exports is only 13 percent.
    - The Czech Republic exports $156 million in cars and car parts to the United States (gross), but its domestic value-added share is about 50 percent; after accounting for indirect value added in other countries’ exports, the Czech Republic’s total loss under the 25 percent tariff scenario is higher in nominal terms than the Slovak Republic’s losses (even though losses as a share of GDP are similar).
- Caveats noted:
  - Network analysis here represents a short-term estimation and abstracts from potential policy responses to tariffs, confidence effects, nonlinearities, and other spillovers that may emerge at larger disruption thresholds.
  - The exercise abstracts from trade diversion that may occur if exporting countries redirect exports to alternative destinations.
  - The elasticity calibration is a simplification; more disaggregated elasticities could change loss magnitudes.

### Key takeaways for analysis and policy
- Using value-added export measures (GVC-based analysis) rather than conventional gross-export measures changes the measured distribution of tariff exposure and losses across countries and sectors.
- GVC-based analysis reveals that impacts of tariffs can be spread across more countries (and in different rankings) than gross-export data would indicate.
- Short-term estimated impacts of a 25 percent US tariff on cars and car parts:
  - EU aggregate impact: 0.1 percent of GDP.
  - Germany impact: 0.15 percent of GDP (about half related to supply chains).
- Policy implication: Granular, GVC-aware analysis is necessary to capture the full distributional effects of trade policy shocks and to identify countries and sectors that are indirectly exposed through supply chains.

*Source: Box 2. Europe’s Global Value Chain Participation, Trade Tensions, Global Value Chains, and Spillovers (figures and text as provided in the source content).*

### 1. Slovak Republic

### 1. Slovak Republic

### Short-term impact of tariff shocks in a supply chain network
- Scenario note: assumes a 25 percent tariff imposed on US imports of cars and car parts.
- Key conceptual point: when input trade is important, final-expenditure switching (consumers substituting away from higher-priced foreign final goods) can be counterbalanced by intermediate-expenditure switching (reduced demand for upstream suppliers). Which channel dominates is ultimately empirical.
- Implication for the Slovak Republic: participation in integrated European supply chains can reduce direct competitive exposure to partner price shocks measured in gross trade terms.

### Methodology and measures of openness
- Normalization approach:
  - Gross exports are divided by gross turnover (GO).
  - Value-added exports are divided by total value added (GDP).
- Rationale: dividing gross exports by GDP mixes a turnover measure (numerator) with a value-added measure (denominator); consistent normalization (turnover-to-turnover or value-added-to-value-added) yields a more accurate picture of exposure to external demand shocks.
- The chapter uses a partial-equilibrium demand system and introduces a value-added real effective exchange rate (VA-REER) that relies on the GDP deflator rather than the consumer price index.

### Differences between VA-REER and conventional GO-REER
- Three main dimensions of difference:
  1. Partner weights reflect trade in value added (double-export weights capture head-to-head competition scaled by export share).
     - Countries in highly integrated supply chains often have much lower VA-REER weights than GO-REER weights; in some cases weights can be negative (partners act as complements rather than substitutes).
     - Examples of countries with negative VA-REER partner weights: Luxembourg, Malta, and Cyprus (when they provide intermediate services to large exporters).
  2. Effective degree of substitutability varies across countries under the VA-REER framework, unlike the fixed and equal elasticity of substitution (EoS) assumed in conventional REER indices.
  3. Degree of openness differs when measured on a value-added basis versus gross trade basis.

### Empirical patterns and implications for small, open European economies
- Large-but-remote economies (United States and China) tend to become more important trading partners in value-added terms than in conventional terms.
- Concentration of exposure:
  - Conventional weights of the largest trading partner were 16 percent larger than value-added weights in 2000.
  - Conventional weights of the largest trading partner were 32 percent larger than value-added weights in 2014.
  - Interpretation: countries are less vulnerable to partner price shocks under a value-added framework than conventional gross-flow indices suggest.
- Germany as a trading partner:
  - Germany’s importance is smaller in VA-REER measures than in conventional GO-REER measures for many central and eastern European economies (including the Slovak Republic, Poland, and Hungary).
  - Germany’s weight in VA-REERs is about 10 percent lower than in gross trade terms among its European trading partners.
- Outside Europe, value-added REER weights are almost 50 percent larger on average than conventional weights for trading hubs such as the United States, China, and Japan, reflecting relatively more final-goods trade with higher value-added content between these hubs and European countries.

### Analytical and policy takeaways
- Measuring competitiveness and exposure using value-added concepts (VA-REER, value-added exports normalized by total value added) provides a different and often less acute view of vulnerability to partner price shocks for countries integrated in supply chains.
- For countries like the Slovak Republic participating in regional supply chains, policy analysis should:
  - Use value-added measures to assess exposure and spillovers.
  - Recognize that intermediate-input linkages can produce countervailing effects to final-demand switches and may even make some partners complements rather than substitutes.
  - Account for heterogeneous elasticities of substitution across countries when modeling responses to relative price changes.

*Source: IMF staff chapter excerpt (ttgvcsieea - 1. Slovak Republic).*

### 2. Average REER Weights for Hub Countries1. Germany’s Weight in Partners’ REER

### 2. Average REER Weights for Hub Countries — Germany’s Weight in Partners’ REER

### Determinants of the effective elasticity of substitution (EoS)
- The three main determinants of the size of the effective EoS:
  - The relative size of the EoS for final and intermediate goods (final demand typically more elastic than demand for intermediate goods).
  - The relative importance of final and intermediate trade in total trade flows.
  - The size of market shares in value-added terms.
- Assumptions used in the subsequent analysis:
  - Leontief production so that elasticities of substitution in production are zero.
  - Nested constant elasticity of substitution (CES) demand with a final demand EoS of 3 and Leontief production elasticities—between differentiated inputs and between total inputs and domestic value added—of zero.
- Empirical range:
  - In the sample of 32 European countries for this chapter, the effective EoS varies between 0.3 and 1.2 in the latest year for which input-output data are available.
- Qualitative determinants and examples:
  - Effective EoS is lower for countries that trade relatively more in intermediate goods (closer supply-chain relationships imply less short-run substitutability).
  - Specialized economies such as Luxembourg and Malta tend to have lower effective EoS because financial services represent business inputs with limited short-run substitutability for larger partners.
  - Large and more distant economies such as China, the United States, Japan, and the United Kingdom typically display larger EoS because they have larger market shares in value-added terms and can substitute more easily to third countries when relative prices change.

### Trade openness and value-added vs gross measures
- Key comparisons:
  - Countries are more open in value-added terms than in gross trade flow terms.
  - In 2014, the share of domestic value added used abroad was about 70 percent larger on average than the analogous measure based on gross output.
  - Despite lower EoS, higher shares of intermediate trade resulted in demand for value added that was about 50 percent more responsive to relative price changes than demand for gross output.
- Interpretation:
  - Value-added measures show more openness because firms trade more in intermediate consumption within countries than between countries, reducing gross turnover shares relative to value-added exports.

### Tariff shock simulation: 5 percent US tariff (with global retaliation)
- Model setup and limitations:
  - Analyze a 5 percent tariff imposed by the United States on its imports across the board, with retaliation by all countries using the same tariff.
  - Assumes full pass-through of a country’s own tariff to its own prices.
  - Simplifying assumption that prices and demand for countries’ exports are not affected by tariffs imposed on other countries (therefore model omits trade diversion).
  - The model considers only the demand side of trade—holding the supply side fixed—so impacts are short-run partial equilibrium effects.
  - The calculations use the Matlab code from Bems and Johnson (2017) and the 2016 vintage of the World Input-Output Database (WIOD) to conduct analysis for 43 countries, from 2000 to 2014.
- Aggregate impacts (Europe):
  - The impact of a 5 percent tariff on all US imports would be an average decrease in the demand for European value added of about 0.2 percent.
  - The impact in terms of demand for gross output would be 0.1 percent.
  - The decline in the demand for Europe’s output is 50 percent higher when the 5 percent tariff shock is measured in value-added terms than when measured in gross terms.
- Heterogeneity across country groups and selected countries:
  - The difference between the estimated impact in value-added terms and in gross terms is three times larger for emerging Europe (compared to advanced Europe).
  - For countries around the German value chain:
    - The decline in demand after the US tariff is twice as large in value-added terms as in gross terms for Germany.
    - The difference is three times larger for Poland.
    - The difference is four times larger for the Czech Republic, Slovenia, and the Slovak Republic.
  - Countries that provide large exports of financial services (for example Luxembourg) are among the least affected; Luxembourg’s short-run effect is even estimated to be positive as more domestic activity in the United States leads to more demand for financial services exports.
- Broad conclusion:
  - European countries are 50 percent more exposed to the United States in value-added terms than in gross trade terms; almost all European countries, with the exception of Luxembourg, Ireland, and Malta, would suffer a larger decline in demand in value-added terms than in gross terms following a US tariff increase.

### Value-Added Real Effective Exchange Rates (VA-REER) vs Conventional REER (GO-REER)
- Comovement and gaps:
  - VA-REER and conventional REER indices display high comovement over time, but cumulative gaps emerge for several countries.
  - VA-REER indices show more real appreciation than conventional indices (GO-REER) for most countries.
  - The gaps peaked in 2008, reaching almost 10 percentage points on average, before declining.
- Cross-country statistics and examples (2014 vs 2000):
  - On average, in 2014, European countries remained 6 percent less competitive in value-added terms relative to their competitiveness in 2000.
  - Substantial cross-country variation:
    - Romania and Russia are about 30 percent less competitive in value-added terms relative to the conventional measure.
    - Ireland is 13 percent more competitive according to the VA-REER.
  - Emerging economies were substantially less competitive in value-added terms than in gross trade flow terms compared to advanced economies.
  - The standard deviation of the VA-REER index was 45 percent larger in 2014 than the GO-REER.
- Implication:
  - Measuring competitiveness and REER in value-added terms yields different and often more pronounced competitive imbalances than conventional gross-output-based measures.

### Growth spillovers in the presence of global value chains
- Objective and approach:
  - Study how growth shocks in major hub economies (United States, China, Germany) are transmitted to European trading partners when accounting for global value chain exposures.
  - Use a macroeconometric framework to estimate dynamic spillovers of growth shocks originating in these hub countries to European economies.
  - The growth shocks considered are general (can embody fiscal, monetary, and financial drivers) and are broader than pure tariff shocks.
- Empirical relationship:
  - Simple scatterplots suggest growth correlations are positively associated with the degree to which European economies are exposed to hub economies in value-added terms.
  - The greater the value-added exports of an economy to a trade hub (United States, China, Germany), the greater the growth correlation between that economy and the respective hub country.
- Context and literature:
  - The chapter complements earlier calibrated, static analyses by providing econometrically estimated dynamic spillovers.
  - The role of global value chains is highlighted as a channel amplifying or shaping growth spillovers and business cycle synchronization.

*Source: IMF staff calculations and cited analyses in the chapter "Price Shocks, Demand for Value Added, and Global Value Chains."*

### 1. United States

### 1. United States

### Methodology: identifying spillovers through value-added trade linkages
- Two-step estimation:
  - Step 1: Estimate country-specific growth shocks as residual growth after purging cross-country commonalities using principal components analysis (PCA) on GDP (quarter-over-quarter) growth for a panel of 68 economies during 1995:Q1–2018:Q3.
    - The three principal components, on average, explain about 28 percent of the variance in GDP growth.
    - The first principal component explains about 21 percent of the variance.
  - Step 2: Trace the effects of these orthogonalized country-specific growth shocks on other economies using a local projections model (Jordà 2005; implementation following Iacoviello and Navarro 2018).
    - Model allows interactions between hub-country shocks and the share of bilateral value-added exports (value-added exports as percent of GDP) to capture how spillovers depend on trade exposure.
    - Europe is analyzed as the set of spillover destinations; results are reported for a “median” European economy (median value-added export exposure) and a “high-exposure” economy (90th percentile of exposure).
- Robustness and alternatives:
  - A variant replaces exogenous shocks with actual GDP growth in the hub countries (one-step specification); this confounds common shocks and spillovers and yields larger estimated spillovers.
  - PCA robustness: a version with only the first principal component yields very similar residual growth shocks.

### Key empirical findings on spillovers (exact estimates preserved)
- Sample and shock size:
  - Analysis period: 1995:Q1–2018:Q3.
  - Spillover experiments consider a 1 percent country-specific (negative) growth shock in the hub countries.
- United States:
  - For the median European economy, a 1 percent negative US growth shock → peak cumulative growth impact over a two-year horizon of about –0.6 percentage point.
  - For a high-exposure European economy (90th percentile), the corresponding impact is about –1.4 percentage points.
  - For a high-exposure economy in advanced Europe, the peak cumulative impact due to a 1 percent negative US growth shock is about –1.6 percentage points.
- China:
  - Spillovers from China are sizable but smaller than those from the United States (numerical peaks not restated beyond relative comparison in the source).
- Germany:
  - Growth spillovers from Germany are much smaller than those from the United States and China.
  - For the median European economy, a 1 percent negative German growth shock → effects are not statistically significant in the baseline (orthogonalized shocks).
  - For a high-exposure European economy, the peak cumulative impact is around –0.1 percentage point (baseline).
  - Using actual German growth (without purging common shocks), the peak cumulative impact on the median European economy from a 1 percent negative German growth is about –0.7 percentage point.
  - The size of spillover estimates from a Germany shock on advanced and emerging Europe is comparable.
- Advanced vs emerging Europe:
  - Spillovers are somewhat smaller for emerging Europe than for advanced Europe, particularly for US-originated shocks.
  - For a high-exposure economy, the peak cumulative impact due to a 1 percent negative US growth shock is about –1.6 percentage points in advanced Europe and about half that for a high-exposure emerging market European economy.
- Other quantitative statements on trade and value chains:
  - Europe: about 70 percent of total European exports are linked to forward and backward supply chains.
  - Accounting exercise example: estimates of the potential impact of 25 percent tariffs on US imports of cars and car parts from the European Union suggest that short-term losses could average about 0.1 percent of GDP; half of this impact would be direct and the other half through supply chains.
  - After a 5 percent US tariff on all its imports (with retaliation from the rest), the average reduction in the demand for Europe’s value added is 0.2 percent.
  - Germany’s importance for its European export partners is about 10 percent lower in value-added terms than in gross trade terms in the VA-REER; the United States and China have their weight almost 50 percent larger in value-added terms compared with gross trade.

### Interpretation, caveats, and channels
- Controlling for common shocks is essential to identify orthogonal country-specific shocks; correlations between estimated US, Chinese, and German-specific growth shocks are close to zero under the PCA approach.
- Local projections provide a dynamic profile and allow for nonlinearities with respect to value-added trade exposure.
- Caveats:
  - Extrapolating results to ongoing trade wars is not straightforward: tariff increases in one hub country could affect many economies simultaneously, and working with global (rather than country-specific) shocks could overstate spillovers.
  - Spillovers can operate through channels beyond trade (financial, commodity prices, confidence); these channels are not modeled explicitly in the baseline framework.
  - The analysis estimates spillovers to destination countries from a 1 percent country-specific growth shock; mapping tariff shocks in earlier chapters to these growth shocks is not straightforward.
  - Feedback effects from the rest of the world could scale up spillovers; reduced-form estimates reflect such feedbacks in addition to trade-channel effects.

### Implications for Europe and policy-relevant conclusions
- Europe is highly integrated in global value chains; supply-chain linkages within and outside Europe mean shocks to major economies can transmit widely across sectors and countries.
- Using value-added exports (rather than gross exports) materially changes the distribution and magnitude of estimated impacts from trade shocks: negative impacts spread across more countries and sectors under value-added measures.
- Relative hub importance shifts under value-added accounting: Germany’s relative importance declines while the United States’ and China’s importance rises.
- Policy relevance:
  - Findings can help quantify more precisely the impact of growing trade tensions on specific countries and sectors.
  - Understanding propagation through global value chains can assist policymakers in calibrating countercyclical measures and targeting social policies to citizens most likely to be affected.

*Italic: Source — ttgvcsieea - 1. United States (PDF chapter/section; canonical URL provided in content unit).*

### Conclusions

### Conclusions

### Annex 1 — Network analysis of tariff shock through global value chains
- Method:
  - Uses network of linkages between country-sector pairs (within and across countries) derived from global input-output tables provided by EORA and the methodology in Aslam, Novta, and Rodrigues-Bastos (2017).
  - Main variables: domestic value added (DVA) and foreign value added in exports of each country sector (aggregate country-level or bilateral country-sector level).
- Shock calculation steps for a tariff τ on country c, sector s exports (exp_cs):
  - Loss in exports assumed: α × τ × exp_cs, where α is based on import elasticities.
  - Loss to country c’s domestic value added: (DVA___exp_cs) × α × τ × exp_cs, where DVA___exp_cs is the share of domestic value added in exp_cs.
  - Remaining loss to foreign value-added component: (1 − DVA__exp_cs) × α × τ × exp_cs, traced back to origin countries using bilateral country-sector linkages.
  - Losses to country c can be decomposed into losses directly incurred by sector s and losses to other sectors within c that feed into s; these plus losses through participation in foreign supply chains are summed as supply‑chain‑related losses.
- Coverage:
  - Exercise repeated over all exporting countries captures direct and indirect losses across countries and sectors.

*Source: ttgvcsieea - Conclusions (PDF chapter).* 

### Annex 2 — Global value chains, demand modeling, and impact of tariffs
- Modeling demand for trade:
  - Aggregate good Q_i is specified as:
    - Q_i = ( (ω_i^v)^(1/(1−γ)) V_i^((γ−1)/γ) + (ω_i^x)^(1/(1−γ)) X_i^((γ−1)/γ) )^(γ/(γ−1)),
      with V_i denoting domestic value added and X_i a CES composite intermediate good (EoS ρ) aggregating varieties X_ji.
  - Final goods demand F_i follows Armington CES aggregation:
    - F_i = ( ∑_j (ω_ji^f)^(1/(1−σ)) F_ji^((σ−1)/σ) )^(σ/(σ−1)).
  - Market clearing: Q_i = Σ_j (F_ij + X_ij).
  - Framework has two types of goods and three elasticities: γ, ρ, and σ. Empirical support suggests σ (final demand elasticity) is higher than γ and ρ; results rely qualitatively on final demand being more elastic and can reverse if elasticities differ.
- Impact of tariffs:
  - Tariffs applied to countries’ gross exports; assume full pass-through of the tariff to buyer price (example: a 5 percent tariff on gross exports of country i to country j leads to a relative price change between 100 and 105).
  - Value-added price change characterization (from Bems and Johnson):
    - p^_VA = [1 − Ω] [diag(s_i^VA)]^−1 p^_GO.
    - Interpretation: value-added price changes are a weighted average of gross output price changes; weights are inverse of total cost shares.
  - Demand effect sequencing:
    - Relative value‑added (gross output) price changes affect demand for value added (gross output) via a single country price per country assumption (which precludes analysis of trade diversion).
    - Example for US value-added demand (holding overall demand F^_i^w constant):
      - V^_US = − Σ_j T_US,j p^_j^VA + F^_US^w,
        where T_US,j ≡ σ T^σ_US,j + ρ T^ρ_US,j + γ T^γ_US,j are value-added weights (linear combination of σ, ρ, γ).
  - Retaliation:
    - Modeled in parallel for partner countries; a 5 percent retaliatory tariff decreases each country’s demand for US exports and increases domestic and other partners’ demand mirroring equation (3).
    - Quantitative note: given the size and relative openness of the US economy, the impact of retaliation is on average about 10 times smaller than the impact of the US tariff.
  - Sensitivity:
    - Assuming a different pass‑through rescales effects but does not change qualitative results since demand is linear in prices.
    - Relaxing unique-price-per-country is beyond scope.

*Source: ttgvcsieea - Conclusions (PDF chapter).* 

### Annex 3 — Two-step method to estimate growth spillovers from country-specific shocks
- Step 1 — Principal components analysis:
  - Purges effects of common shocks to obtain country‑specific growth shocks.
  - Estimate three principal components using GDP (quarter-over-quarter) growth for a balanced panel of 68 economies during 1995:Q1–2018:Q3 (covers advanced Europe, emerging Europe, and other major advanced and emerging market and developing economies).
  - Country-specific shocks = residual growth after removing common components.
- Step 2 — Local projections model to trace spillovers:
  - Model specification for horizon h = 0,1,...,H:
    - y_{i,t+h} = u_{i,h} + β_h U_t + θ_h (U_t × GVC_{i,t−1}) + ψ_h(L) y_{i,t−1} + X_{i,t} Γ_h + ε_{i,t+h}.
    - y_{i,t+h} denotes (cumulative) GDP growth for country i between t and t+h.
    - U_t denotes the US-specific growth shock when tracing US-origin spillovers.
    - GVC_i,t−1 denotes value‑added exports as percent of GDP (measure of trade exposure).
    - Controls include own lagged growth, oil prices, global uncertainty (measured by stock market volatility index), and country fixed effects; US model includes a global financial crisis dummy.
  - Interaction term U_t × GVC_{i,t−1} allows nonlinear spillovers: marginal impact of shock depends on level of value‑added exports.
  - Normalization of exposure (following Iacoviello and Navarro (2018)):
    - GVC_{i,t} = (gvc_{i,t} − gvc_{50}) / (gvc_{90} − gvc_{50}), where gvc_{50} and gvc_{90} are sample median and 90th percentile.
    - Marginal impact for a median economy = β_h; for a high‑exposure economy (90th percentile) = β_h + θ_h.
  - Inference:
    - Confidence bands use Driscoll and Kraay (1998) standard errors allowing arbitrary correlation across countries and time; report two standard deviation confidence bands.
  - Robustness:
    - Results robust to including value‑added exports as a separate regressor.
  - Applications:
    - Same method used to estimate spillovers from China and Germany by changing shock definition and value‑added exports measure.

*Source: ttgvcsieea - Conclusions (PDF chapter).* 

### Key statistics from Annex Table 3.1 — Value‑Added Exports (Percent of GDP)
- Median: United States 2.50; China 0.7; Germany 2.3
- High-exposure (90th percentile): United States 4.1; China 1.4; Germany 5.4
- Select country examples (United States | China | Germany):
  - Austria: 3.3 | 1.4 | 5.4
  - Belgium: 4.4 | 1.6 | 5.0
  - Czech Republic: 4.0 | 1.3 | 11.7
  - Ireland: 5.2 | 0.7 | 2.3
  - Netherlands: 3.7 | 1.2 | 2.6
  - Switzerland: 5.7 | 2.3 | 5.3
  - Israel: 6.4 | 0.7 | 2.4
  - Moldova: 0.0 | 0.0 | 0.0
- Sources for table: EORA database; and IMF staff calculations.

*Source: ttgvcsieea - Conclusions (PDF chapter).*

---


_Source: https://www.imf.org/-/media/files/publications/dp/2019/english/ttgvcsieea.pdf_