## _wp07266

## Source details

**Canonical URL:** [_wp07266](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07266.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07266.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07266.pdf.json)

---

### I. Introduction — purpose and contribution
- Central questions:
  - Size and stability of China’s trade elasticities determine how changes in external conditions or the exchange rate affect China’s growth and current account.
  - Stability of trade elasticities is crucial for methodologies that compute underlying or equilibrium exchange rates (for example, the macroeconomic-balance approach).
- Motivation:
  - If trade elasticities are low, changes in external conditions or the exchange rate are unlikely to have large effects on China’s economy or current account.
  - If trade elasticities are unstable, projections of how the economy reacts to shocks are unreliable.
- Contribution:
  - Uses both aggregate and disaggregated data to analyze China’s trade elasticities and their stability.
  - Finds relatively stable aggregate import demand and price elasticities, but export elasticities have increased over time.
  - Attributes much of the increase in aggregate export elasticities to changes in the composition of China’s trade, notably increasing sophistication of exports and rising domestic content of processing trade.

### II. How China’s trade changed — structural drivers
- China’s position and trade dynamics:
  - After a decade of rapid trade growth, China is now the world’s third largest exporter, behind the United States and Germany.
  - China’s trade surplus has widened sharply as import growth has lagged export growth.
- Structural changes affecting elasticities:
  - Privatization of the export sector: in 1995 more than one third of total exports was produced by state-owned enterprises; by 2006 this share had fallen to about one fifth.
  - Trade liberalization: unweighted average tariffs declined from about 43 percent in 1992 to [text truncated in source]; import tariffs: simple average fell from 16.4 percent in 2000 to under 10 percent by 2005; import quotas and licensing abolished by 2005.
  - Composition shifts: increasing share of exports with relatively high elasticities; rising domestic content of processing trade as processing exports move away from pure assembly operations.
- Composition and sophistication:
  - Primary products share: declined from 15 percent in 1995 to 3 percent in 2006.
  - Machinery (including electronics) share: rose from 20 percent in 1995 to about 60 percent in 2006.
  - Sophistication index: trade-weighted PRODY shows increase in sophistication across major categories; overall index rose more sharply than individual categories.
- Processing trade and domestic content:
  - Processing trade ≈ more than fifty percent of total exports (customs classification).
  - Assembly exports share declined to less than 10 percent by 2006, about half its share in 1992.
  - Domestic content of nonassembly processing exports (excluding equipment imports) rose from about 20 percent to close to 35 percent during 1992−2005.
  - Net processing imports of basic materials to total processing exports: fell from 25 percent to 5 percent by 2005.
  - Ratio of machinery imports to total exports of processing trade: fell from 40 percent (early 1990s) to 7 percent by 2005.
  - Implication: rising domestic content increases sensitivity of processing exports to exchange rate changes.

### III. Empirical framework and data
- Model:
  - Reduced-form partial-equilibrium trade model with productivity term.
  - Export equation includes real foreign demand (D*), relative export price (Px), productivity (Prod).
  - Import equation includes domestic demand (D), relative import price (Pm).
- Data and measures:
  - Relative price for exports: CPI-based real effective exchange rate.
  - Relative price for imports: PPI-based trade-weighted ratios at SITC2 level.
  - Productivity proxy: ratio of per capita GDP in China to per capita GDP in the United States.
  - Sample: quarterly data 1995Q1−2006Q4 (data prior to 1994 excluded).
- Estimation:
  - Dynamic OLS (DOLS) with one lead and one lag, seasonal dummies; Monte Carlo properties favor DOLS for short samples.

### IV. Aggregate empirical results (key estimates)
- Aggregate estimated elasticities (full sample):
  - Export elasticity to foreign demand (D*): 3.8 (standard deviation 0.29) ***
  - Export elasticity to relative price (Px): −1.6 (standard deviation 0.27) ***
  - Import elasticity to domestic demand (D): 1.3 (standard deviation 0.04) ***
  - Import elasticity to relative price (Pm): 0.9 (standard deviation 0.20) ***
  - Productivity (Prod) coefficient in exports: 1.3 (standard deviation 0.09) ***
- Statistical fit:
  - Export equation: R-squared 0.99; Adj. R-squared 0.99.
  - Import equation: R-squared 0.98; Adj. R-squared 0.98.
- Robustness and comparisons:
  - Estimates are within ranges of other studies and satisfy Marshall-Lerner condition.
  - Panel/bilateral studies cited show export demand elasticities ~4.27, 4.33; price elasticities −1.32, −3.26.

### V. Time variation, structural breaks, and stability evidence
- Time variation:
  - Export elasticities increased over time in absolute value for both demand and price elasticities.
  - Import elasticities showed less variability; slight decline in import demand elasticity consistent with increased domestic sourcing.
- Sub-sample changes (selected numbers):
  - Export demand elasticity rose from 3.56 (1995−1999) to 4.26 (2000−2006) in reported specifications.
  - Export price elasticity increased from −1.31 (1995−1999) to −2.00 (2000−2006).
  - Import price elasticity examples: 1.03 (1995−1999) and 0.41 (2000−2006).
- Breakpoint/testing:
  - Chow tests indicate 1999Q4 is a breakpoint for both export and import equations (F-statistic and log likelihood tests significant).
  - Bai-Perron test confirms breakpoint in import equation at 1999Q4; fails to detect break in export equation.
  - CUSUM tests do not detect large abrupt one-time shifts, consistent with continuous policy changes.
- Cointegration and unit roots (Appendix I):
  - ADF tests: most variables nonstationary except relative import price index (P_m) where t-Statistic -3.20 Probability 0.03.
  - Trace tests: export equation indicates 2 cointegrating eqn(s) at the 0.05 level; import equation indicates 1 cointegrating eqn(s) at the 0.05 level.
  - Residuals ADF: export residuals t-Statistic -4.2 Probability 0.0; import residuals t-Statistic -2.7 Probability 0.1 (stationary).

### VI. Sectoral heterogeneity — sectoral elasticities and drivers
- Sectoral export demand elasticities (selected full-sample values):
  - Primary goods: 2.09 (standard error 0.16) ***
  - Manufactures (example): 2.94 (0.21) ***
  - Chemical: 4.14 (0.94) ***
  - Machinery: 7.03 (0.87) ***
  - Electric & electronics machinery: 9.71 (0.95) ***
  - Clothing: 2.47 (0.51) ***
  - Other manufactures: 2.90 (0.43) ***
- Sectoral export price elasticities (selected full-sample values):
  - Primary goods: −0.96 (0.23) ***
  - Manufactures (example): −1.28 (0.20) ***
  - Chemical: −2.06 (0.38) ***
  - Machinery: −1.64 (0.33) ***
  - Electric & electronics machinery: −2.17 (0.38) ***
  - Clothing: −0.61 (0.48)
  - Other manufactures: −0.97 (0.40) ***
- Import sector elasticities:
  - Import price elasticities (selected): chemicals 1.5; manufacturing by material 1.4 (sectoral tables provide more detail).
  - Import equations include sector-specific price indices and, for electric/electronics, export volume (Xelec).
- Weighted elasticities:
  - Rising share of electronics and electric machinery drove rising weighted export elasticities.
  - Weighted import elasticities remained broadly unchanged.
- Rolling-window sectoral estimates show substantial cross-sectoral heterogeneity in magnitudes and volatility of elasticities.

### VII. Sources of rising export elasticities — empirical tests
- Identified contributors:
  1. Changes in sectoral composition (shift toward high-elasticity sectors like electronics and machinery).
  2. Rising domestic content in processing trade.
  3. Increased product sophistication.
- Processing vs nonprocessing:
  - Processing export relative price elasticity increased over time; nonprocessing export elasticity remained broadly unchanged.
  - Regression interaction Px × DC (DC = domestic content of processing trade): coefficient −0.09 (standard deviation 0.04) ** (sample 2001/Q1–2006/Q4) — implies rising domestic content raises price elasticity of processing trade.
  - Tests for assembly share interaction (AS): D* × AS and Px × AS insignificant (sample 1995/Q1–2006/Q4).
- Product sophistication interactions (selected coefficients, Table 6):
  - D* coefficient (exports): 6.21 (standard deviation 1.54) ***
  - Px coefficient (exports): −2.01 (standard deviation 0.40) **
  - D* × Sx: 16.27 (standard deviation 7.01) ** — higher export sophistication increases demand elasticity.
  - Px × Sx: −15.71 (standard deviation 6.97) ** — higher export sophistication increases price sensitivity (more negative).
  - For imports: D × Sm: 6.58 (standard deviation 3.91) *; Pm × Sm not significant.

### VIII. Macroeconomic implications and numerical examples
- Increased external dependence:
  - A one percent decline in external demand could lead to about 4½ percent contraction in China’s export or about ¾ percent decline in GDP (taken at face value).
- Change in export price elasticity:
  - Shift from −1.3 during 1995−1999 to −2.0 during 2000−2006.
  - Example implication: for a 10 percent reduction in exports, the required real appreciation would be about one-third less than previous estimates.
- Interaction example:
  - A 10 percent real appreciation would, ceteris paribus, reduce exports by 2 percent; if external demand grew by 0.5 percent concurrently, that export reduction might not show in aggregate trade numbers.

### IX. Policy-relevant conclusions and cautions
- Main conclusions:
  - Trade elasticities for China during 1995−2006 changed significantly at both aggregate and sectoral levels.
  - Causes: trade composition shifts, increased domestic content of processing exports, and rising product sophistication.
- Policy implications:
  - Because export elasticities have risen considerably and trade structure is changing rapidly, policymakers should exercise caution when assessing how China’s economy and trade balance will respond to external demand shocks or exchange rate changes.
  - Methods that rely on stable trade elasticities (for example, macroeconomic-balance approaches to exchange rate misalignment) may yield unreliable results if elasticities continue to change.
- Structural drivers to monitor:
  - Ongoing privatization, trade liberalization, rising sophistication of exports, and increasing domestic content of processing trade should be monitored as they materially affect trade elasticities and transmission of external shocks.

*Source: _wp07266*

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

### _wp07266 - References..............................................................................................................

### I. Introduction
- Central questions:
  - Size and stability of China’s trade elasticities determine how changes in external conditions or the exchange rate affect China’s growth and current account.
  - Stability of trade elasticities is crucial for methodologies that compute underlying or equilibrium exchange rates (for example, the macroeconomic-balance approach).
- Motivation:
  - If trade elasticities are low, changes in external conditions or the exchange rate are unlikely to have large effects on China’s economy or current account.
  - If trade elasticities are unstable, projections of how the economy reacts to shocks are unreliable.
- Contribution:
  - Uses both aggregate and disaggregated data to analyze China’s trade elasticities and their stability.
  - Finds relatively stable aggregate import demand and price elasticities, but export elasticities have increased over time.
  - Attributes much of the increase in aggregate export elasticities to changes in the composition of China’s trade, notably increasing sophistication of exports and rising domestic content of processing trade.

### II. How China’s Trade Has Changed
- China’s position and trade dynamics:
  - After a decade of rapid trade growth, China is now the world’s third largest exporter, behind the United States and Germany.
  - China’s trade surplus has widened sharply as import growth has lagged export growth.
- Structural changes affecting trade elasticities:
  - Privatization of the export sector.
  - Trade liberalization.
  - Increasing share of exports with relatively high elasticities.
  - Rising domestic content of exports as processing exports move away from pure assembly operations.
- Ownership shifts (1995–2006):
  - In 1995 more than one third of total exports was produced by state-owned enterprises; by 2006 this share had fallen to about one fifth.
- Trade barrier reduction:
  - Unweighted average tariffs declined from about 43 percent in 1992 to [text truncated in source].

### III. Key Quantitative Findings on Elasticities
- Changes in export elasticities (Appendix II):
  - Elasticity with respect to external demand rose from 3.6 in 1995−1999 to around 4.3 by 2000−2006.
  - Price elasticity increased from −1.3 in 1995−1999 to −2.0 by 2000−2006.
- Interpretation and implications:
  - These are large shifts occurring over a short period.
  - The rising export demand and price elasticities imply that relatively moderate changes in external demand or in the exchange rate could have large effects on the economy and the trade balance.
  - Given rapid structural change, it is dangerous to expect these new elasticity estimates to remain stable over the medium term.
  - Trade composition itself is likely to be affected by changes in the exchange rate.

### IV. Figures and Tables Referenced (themes only)
- Figures illustrating:
  - Share of Exports in the World.
  - Trade developments (export growth, import growth, trade balance).
  - Trade by ownership (state-owned, foreign invested, domestic private).
  - Composition and sophistication of trade; evolution and domestic input of processing trade.
  - Regional trade balance and various elasticity estimation diagnostics (aggregate, recursive, CUSUM, sectoral, weighted).
- Tables covering:
  - Sectoral composition of processing exports.
  - Aggregate trade elasticity.
  - Chow test of breakpoint at 1999/Q4.
  - Impact of domestic content and assembly trade.
  - Trade sophistication indices and elasticities.
- Appendices include:
  - Unit root and cointegration tests.
  - Elasticities in sub-samples and at the sectoral level.
  - Sectoral export/import elasticity tables and rolling sample period figures.

### V. Policy-Relevant Conclusions
- Caution in policy assessment:
  - Because export elasticities have risen considerably and trade structure is changing rapidly, policymakers should exercise caution when assessing how China’s economy and trade balance will respond to external demand shocks or exchange rate changes.
- Model and measurement implications:
  - Methods that rely on stable trade elasticities (for example, macroeconomic-balance approaches to exchange rate misalignment) may yield unreliable results if elasticities continue to change.
- Structural drivers to monitor:
  - Ongoing privatization, trade liberalization, rising sophistication of exports, and increasing domestic content of processing trade should be monitored as they materially affect trade elasticities and transmission of external shocks.

*Source: _wp07266 - References..............................................................................................................*

### 16.4 percent in 2000—just prior to China’s

### _wp07266 - 16.4 percent in 2000—just prior to China’s

### Trade liberalization and tariffs
- Import tariffs: simple average fell from 16.4 percent in 2000 to under 10 percent by 2005.
- Import quotas and licensing: abolished by 2005.
- Effect: increased responsiveness of enterprises to market signals (reference to Cerra and Saxena (2003) evidence).

### Composition and sophistication of trade
- Exports by price-sensitivity category:
  - Primary products: declined from 15 percent in 1995 to 3 percent in 2006.
  - Machinery (including electronics): rose from 20 percent in 1995 to about 60 percent in 2006.
- Sophistication index:
  - Measured using trade-weighted PRODY based on Hausman, Huang, and Rodrik (2006).
  - Increase in sophistication occurred across all major categories and rose more sharply for the overall index than for individual categories.

### Processing trade — structural shift and domestic content
- Customs classification: processing trade ≈ more than fifty percent of total exports (tax purpose classification).
- Two processing subcategories:
  - (i) Processing based on contractual agreements with foreign suppliers/buyers (no import taxes or VAT).
  - (ii) Processing using imported inputs (pay taxes then claim rebates).
- Assembly exports:
  - Share declined to less than 10 percent by 2006, about half its share in 1992.
- Domestic content of nonassembly processing exports (excluding equipment imports):
  - Rose from about 20 percent to close to 35 percent during 1992−2005.
- Ratios and shares:
  - Net processing imports of basic materials to total processing exports: fell from a high of 25 percent to 5 percent by 2005.
  - Ratio of machinery imports to total exports of processing trade: fell from a peak of 40 percent in the early 1990s to 7 percent by 2005.
- Implication: rising domestic content increases sensitivity of processing exports to exchange rate changes.

### Sectoral composition of processing exports (Table 1)
- Shares of processing exports by sector:
  - Basic material: 1992 18.5; 1995 23.2; 2000 17.3; 2005 13.0
  - Textiles: 1992 38.5; 1995 29.0; 2000 20.1; 2005 9.1
  - Machinery: 1992 30.1; 1995 37.1; 2000 52.7; 2005 71.0
  - Others manufacturing: 1992 12.9; 1995 10.7; 2000 10.0; 2005 6.9

### Empirical framework and data
- Model: reduced-form partial-equilibrium trade model with productivity term added.
  - Export equation includes real foreign demand (D*), relative export price (Px), productivity (Prod).
  - Import equation includes domestic demand (D), relative import price (Pm).
- Key data and choices:
  - Relative price for exports: CPI-based real effective exchange rate.
  - Relative price for imports: PPI-based trade-weighted ratios at SITC2 level.
  - Productivity proxy: ratio of per capita GDP in China to per capita GDP in the United States.
  - Sample: quarterly data 1995Q1−2006Q4 (data prior to 1994 excluded).
- Estimation method: Dynamic OLS (DOLS) with one lead and one lag, seasonal dummies; Monte Carlo properties favor DOLS for short samples.

### Aggregate empirical results (Table 2 and interpretation)
- Aggregate estimated elasticities (full sample):
  - Export elasticity to foreign demand (D*): 3.8 (standard deviation 0.29) ***
  - Export elasticity to relative price (Px): −1.6 (standard deviation 0.27) ***
  - Import elasticity to domestic demand (D): 1.3 (standard deviation 0.04) ***
  - Import elasticity to relative price (Pm): 0.9 (standard deviation 0.20) ***
  - Productivity (Prod) coefficient in exports: 1.3 (standard deviation 0.09) ***
- Statistical fit:
  - Export equation: R-squared 0.99; Adj. R-squared 0.99.
  - Import equation: R-squared 0.98; Adj. R-squared 0.98.
- Comparison and robustness:
  - Estimates are within ranges of other studies and satisfy Marshall-Lerner condition.
  - Panel and bilateral studies cited show similar magnitudes (examples: export demand elasticities ~4.27, 4.33; price elasticities −1.32, −3.26).

### Time variation and structural breaks
- Rolling subsample and recursive estimates:
  - Export elasticities increased over time in absolute value for both demand and price elasticities.
  - Import elasticities showed less variability; slight decline in import demand elasticity consistent with increased domestic sourcing.
- Breakpoint tests:
  - Chow tests indicate 1999Q4 is a breakpoint for both export and import equations (F-statistic and log likelihood tests significant).
  - Bai-Perron test confirms breakpoint in import equation at 1999Q4; fails to detect break in export equation.
  - CUSUM tests do not detect breaks (no large abrupt one-time shifts), consistent with continuous policy changes.

### Sources of changes in elasticities — empirical evidence
- Three main contributors:
  1. Changes in sectoral composition of trade.
  2. Rising domestic content in processing trade.
  3. Increased product sophistication.
- Sectoral elasticities (Figure 13 highlights):
  - Export demand elasticities (selected):
    - Machinery: 7.0
    - Electric & electronics machinery: 9.7
    - Primary goods: 2.1
  - Export price elasticities (selected):
    - Electric & electronics machinery: −2.2
    - Manufacturing by material: −2.1
    - Machinery: −1.6
  - Import demand elasticities: small variations across sectors.
  - Import price elasticities (selected): chemicals 1.5; manufacturing by material 1.4.
- Weighted trade elasticities:
  - Rising share of electronics and electric machinery (largest increases in export shares) drove rising weighted export elasticities.
  - Weighted import elasticities remained broadly unchanged.

### Processing vs nonprocessing exports (tests and results)
- Processing export relative price elasticity increased over time; nonprocessing export elasticity remained broadly unchanged.
- Formal regression including interaction term Px × DC (DC = domestic content of processing trade):
  - Coefficient on Px × DC: −0.09 (standard deviation 0.04) ** (sample 2001/Q1–2006/Q4) — implies rising domestic content raises price elasticity of processing trade.
- Tests for assembly share interaction (AS):
  - Interaction terms D* × AS and Px × AS insignificant (sample 1995/Q1–2006/Q4), implying assembly share had no detectable effect on demand or price elasticities.

### Product sophistication and elasticity interaction (Table 6)
- Equations include interaction terms between demand/price and sophistication indices (Sx for exports, Sm for imports).
- Key coefficients and interpretations:
  - D* coefficient in export equation: 6.21 (standard deviation 1.54) ***
  - Px coefficient in export equation: −2.01 (standard deviation 0.40) **
  - D* × Sx (interaction): 16.27 (standard deviation 7.01) ** — indicates higher export sophistication increases demand elasticity.
  - Px × Sx (interaction): −15.71 (standard deviation 6.97) ** — indicates higher export sophistication increases price sensitivity (more negative).
  - For imports, D × Sm: 6.58 (standard deviation 3.91) * (affects demand elasticity); Pm × Sm not significant.
- Conclusion: as products become more sophisticated, export elasticities rise in absolute terms for both demand and price channels; import sophistication mainly affects demand elasticity.

### Macroeconomic implications and quantitative examples
- Increased demand elasticity implies greater dependence of the Chinese economy on external conditions:
  - A one percent decline in external demand could lead to about 4½ percent contraction in China’s export or about ¾ percent decline in GDP (taken at face value).
- Change in export price elasticity across periods:
  - Export price elasticity shifted from −1.3 during 1995−1999 to −2.0 during 2000−2006.
  - Example implication: for a 10 percent reduction in exports, the required real appreciation would be about one-third less than previous estimates.
- Interaction example:
  - A 10 percent real appreciation would, ceteris paribus, reduce exports by 2 percent; if external demand grew by 0.5 percent concurrently, that export reduction might not show in aggregate trade numbers.

### Conclusion and caution for policy/analysis
- Trade elasticities for China during 1995−2006 changed significantly at both aggregate and sectoral levels.
- Causes: trade composition shifts, increased domestic content of processing exports, and rising product sophistication.
- Policy/analysis note: extra caution needed when using trade elasticities to estimate responses to price and demand shocks because elasticities are time-varying and linked to structural changes in China’s production and trade organization. Quantitative assessments that ignore these structural shifts risk producing misleading results.

*Source: _wp07266 - 16.4 percent in 2000—just prior to China’s (PDF chapter/section).*

### APPENDIX I: UNIT ROOT AND CONINTEGRATION TESTS

### APPENDIX I: UNIT ROOT AND CONINTEGRATION TESTS

### Augmented Dickey–Fuller (ADF) tests — original variables
- Augmented Dikey−Fuller tests show that all the variables, with the exception of relative import price index, used in the aggregated export and import equations are nonstationary.
- ADF t-Statistic and Probability (original variables):
  - Aggregate exports X: t-Statistic 2.79  Probability 1.00
  - External demand D*: t-Statistic -1.13  Probability 0.70
  - Real exchange rate P_x: t-Statistic -1.41  Probability 0.57
  - Relative productivity Prod: t-Statistic 0.41  Probability 0.98
  - Export quality index Xqu: t-Statistic -0.52  Probability 0.88
  - Aggregate imports M: t-Statistic 0.06  Probability 0.96
  - Domestic demand D: t-Statistic 2.18  Probability 1.00
  - Relative import price P_m: t-Statistic -3.20  Probability 0.03
  - Import quality index Mqu: t-Statistic -1.16  Probability 0.68
- Table I.1. Augmented Dickey-Fuller Test Statistic of Original Variables

### Cointegration tests (Unrestricted Cointegration Rank Test — Trace)
- Cointegration tests show that variables used in both export and import equations are cointegrated.
- The variables for the export equation may have upto two cointegration vectors, while the import equation only has one.
- Unrestricted Cointegration Rank Test (Trace) results (as reported):
  - Export equation (Variables: X, D*, P_x, Prod)
    - Number of CE(s): None * ; Eigenvalue 0.6985 ; Trace statistic 5.3847 ; Critical value 7.860 ; Probability 0.00
    - At most 1 * ; Eigenvalue 0.3530 ; Trace statistic 0.5429 ; Critical value 29.800 ; Probability 0.04
    - At most 2 ; Eigenvalue 0.1810 ; Trace statistic 0.1315 ; Critical value 15.490 ; Probability 0.27
    - At most 3 ; Eigenvalue 0.020 ; Trace statistic 0.853 ; Critical value 3.840 ; Probability 0.36
    - Trace test indicates 2 cointegrating eqn(s) at the 0.05 level
  - Import equation (Variables: M, D, P_m, ...)
    - Number of CE(s): None * ; Eigenvalue 0.3231 ; Trace statistic 1.0129 ; Critical value 29.800 ; Probability 0.04
    - At most 1 ; Eigenvalue 0.1712 ; Trace statistic 2.561 ; Critical value 15.490 ; Probability 0.13
    - At most 2 ; Eigenvalue 0.073 ; Trace statistic 3.533 ; Critical value 3.840 ; Probability 0.06
    - At most 3 ; Eigenvalue 0.020 ; Trace statistic 0.853 ; Critical value 3.840 ; Probability 0.36
    - Trace test indicates 1 cointegrating eqn(s) at the 0.05 level
- 1/ * Denotes rejection of the hypothesis at the 0.05 level; and **MacKinnon-Haug-Michelis (1999) p-values.
- Table I.2. Unrestricted Cointegration Rank Test (Trace) 1/

### Stationarity of residuals (ADF tests of residuals)
- Augmented Dickey−Fuller tests show that residuals of the export and import equations are stationary, confirming that long−run relationships exist as hypothesized in the export and import equations.
- ADF t-Statistic and Probability (residuals):
  - Export equation residuals: t-Statistic -4.2  Probability 0.0
  - Import equation residuals: t-Statistic -2.7  Probability 0.1
- Table I.3. Augmented Dickey-Fuller Test Statistic of Residuals

*Source: APPENDIX I: UNIT ROOT AND CONINTEGRATION TESTS (content provided)*

---

### APPENDIX II: ELASTICITIES IN SUB-SAMPLES AND AT THE SECTORAL LEVEL

### Sub-sample elasticities — overview
- Elasticities for two sub-sample periods, based on the Chow breaking point test, are presented.
- Note that export elasticity estimated for the first period, 1995Q1−1999Q4 is not significant, possibly a result of major trade policy reform during that period.

### Sub-period elasticity estimates (selected reported coefficients)
- Dependent Variable: M (imports) and X (exports); sub-periods shown include 1995/Q1–1999/Q4, 2000/Q1–2006/Q4, 2001/Q1–2006/Q4.
- Selected coefficient excerpts (with significance markers and Newey-West HAC standard errors where provided):
  - For one specification:
    - D*: 3.56 0.89 *** ; 4.26 0.45 *** 
    - P_x: -1.31 0.93 ; -2.00 0.41 ***
    - P_m: 1.03 0.59 ** ; 0.41 0.18 **
    - Prod: 1.08 0.57 * ; 1.08 0.17 ***
    - C: -2.59 2.87 ; 0.00 1.10
    - ΔD* (-1): -10.44 4.00 *** ; -3.81 2.67
    - ΔP_m (-1): 0.65 1.26 ; -1.04 0.45 **
    - Dummy for Q1: -0.20 0.05 *** ; -0.15 0.03 ***
    - R-squared: 0.85 0.99 ; 0.80 0.97
    - Adj. R-squared: 0.80 0.99 ; 0.75 0.96
  - Notes:
    - 1/ * significant at 10 percent, **significant at 5 percent, and ***significant at 1 percent. Δ, first difference. (-1) first lag.
    - 2/ Based on Newey-West HAC standard errors.
    - Period label examples: 2000/Q1–2006/Q4 ; 1995/Q1–1999/Q4 ; 2001/Q1–2006/Q4

### Export elasticities by sector (full sample) — selected coefficients
- Table II.2. Export Elasticities by Sector Using Full Sample 1/ reports sectoral coefficient estimates and standard errors for sectors including Primary, Manufactures, Chemical, Machinery, Electronics, Clothing, Manufactures (Other).
- Selected coefficient excerpts (format: coefficient  standard error significance where given):
  - D* coefficients in different sector regressions: 2.09    0.16 *** ; 2.94    0.21 *** ; 4.14    0.94 *** ; 7.03    0.87 *** ; 9.71    0.95 *** ; 2.47    0.51 *** ; 2.90    0.43 ***
  - P_x coefficients: -0.96    0.23 *** ; -1.28    0.20 *** ; -2.06    0.38 *** ; -1.64    0.33 *** ; -2.17    0.38 *** ; -0.61    0.48 ; -0.97    0.40 ***
  - Prod coefficients: 0.73    0.28 *** ; 0.67    0.26 *** ; 0.26    0.07 *** ; 0.86    0.16 *** ; 1.08    0.17 ***
  - C (constant) examples: 3.81    0.75 *** ; 0.98    0.07 *** ; -0.51    3.15 ; -16.51    2.78 *** ; -25.34    2.97 ***
  - ΔD* (-1) examples: -7.41    3.14 ** ; -7.00    1.34 *** ; -7.38    2.61 *** ; -9.84    2.78 *** ; -11.91    3.76 *** ; -4.64    2.41 *
  - ΔP_x (1): -1.34    0.46 ***
  - ΔProd(-1): -1.07    0.54 *
  - Dummy for Q1 examples: -0.27    0.03 *** ; -0.12    0.03 *** ; -0.17    0.03 *** ; -0.13    0.03 *** ; -0.24    0.04 ; -0.20    0.02 ***
  - R-squared reported across sector regressions: 0.98, 0.99, 0.96, 0.98, 0.98, 0.89 (varies by sector)
- Notes:
  - 1/ * significant at 10 percent, **significant at 5 percent, and ***significant at 1 percent. Δ, first difference. (-1) first lag, (1) first lead.
  - 2/ Based on Newey-West HAC standard errors.
  - 3/ Per capita electronics output in China relative to that in Japan is used as Prod as discussed in section III. Full sample seasonality is imposed for exports in this sector by seasonally adjust the data using dummy variables to accommodate the issue of a small sample (especially in sub-sample regressions) and a large number of independent variables.
  - 1/ (footnote in Figure): Including transportation; excluding electric and electronics.

### Import elasticities by sector (full sample)
- Table II.3. Import Elasticities by Sector Using Full Sample 1/ reports sectoral import elasticity estimates.
- Selected coefficient excerpts and notes:
  - Dependent-variable-specific coefficients for D, P_m (sector-specific price indices), X_elec included for electric/electronics sector (Xelec 4/ 0.72 0.10 ******).
  - ΔP_m (-1) examples: -0.59    0.24 *** ; -1.64    0.39 *** ; -0.72    0.25 *** ; -1.44    0.44 *** ; -2.18    0.79 ***
  - Dummy quarter coefficients and R-squared values reported; R-squared examples: 0.93, 0.98, 0.93, 0.89, 0.99, 0.93, 0.97
  - 2/ Based on Newey-West HAC standard errors.
  - 3/ The price indices used are sector specific; they are import weighted ratio of China's sector PPI to U.S. import price indices at SITC2.
  - 4/ Export volume of electric machinery and electronics, included in this sector's import equation to reflect the large share of processing exports.

### Rolling (subsample) sectoral export elasticities — summary of figures
- Figure II.1. Sectoral Export Elasticities by Rolling Sample Periods presents rolling-period estimates with:
  - Full sample elasticity line
  - Rolling period elasticity line
  - ±2 std. E of period elasticity bands
- Rolling-period x-axis labels run from 95Q1-02Q4 through 99Q1-06Q4 (various rolling windows such as 95Q1-02Q4, 95Q2-03Q1, ... , 99Q1-06Q4).
- Sector-specific charts shown include:
  - Primary: Export Elasticity to External Demand; Export Elasticity to Real Exchange Rate
  - Chemical: Export Elasticity to External Demand; Export Elasticity to Real Exchange Rate
  - Manufacturing by Material: Export Elasticity to External Demand; Export Elasticity to Real Exchange Rate
  - Machinery (including transportation; excluding electric and electronics): Export Elasticity to External Demand; Export Elasticity to Real Exchange Rate
  - Electric Machinery and Electronics: Export Elasticity to External Demand; Export Elasticity to Real Exchange Rate
  - Clothing: Export Elasticity to External Demand; Export Elasticity to Real Exchange Rate
  - Miscellaneous Manufacturing: Export Elasticity to External Demand; Export Elasticity to Real Exchange Rate
- Visuals indicate substantial cross-sectoral heterogeneity in magnitudes and volatility of elasticities across rolling samples (figures provided in the source).

*Source: APPENDIX II: ELASTICITIES IN SUB-SAMPLES AND AT THE SECTORAL LEVEL (content provided)*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp07266.pdf_
