## _wp0426

## Source details

**Canonical URL:** [_wp0426](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0426.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0426.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0426.pdf.json)

---

### Introduction — research question and approach
- Research question: How much does a country’s long-term economic growth depend on economic conditions in the rest of the world?
- Scope and sample:
  - Panel data for the period 1960–99.
  - 101 industrial and developing economies.
  - Observations are five-year averages (initial GDP per capita takes first-year value of each five-year period).
- Trade weights:
  - Time series of export shares constructed from IMF Direction of Trade Statistics (IMF, 2002) for 1960–99; average share over a five-year period is used.
- Estimation focus: fixed-effects panel regressions to capture long-run impacts rather than short-run business-cycle effects.

### Empirical framework and key variables
- Dependent variable:
  - Real GDP per capita growth (1995 constant US$).
- Standard control variables:
  - ln(initial GDP per capita), population growth, investment/GDP, secondary school enrollment, inflation, trade/GDP.
- Variables of interest:
  - Trading partners’ real per capita GDP growth.
  - Ratio of domestic real per capita GDP to trading partners’ real per capita GDP (and alternatively trading partners’ real per capita GDP).
  - Interaction terms: trading partners’ growth × Trade/GDP.
  - World growth, non-trading-partner growth, and distance-weighted growth to capture regional/global trends.
- Methodology notes:
  - Fixed-effects estimated (Hausman test justifies fixed effects for most specifications).
  - Pooled-panel and cross-section estimates used for robustness.

### Key empirical findings — trading-partner growth effects
- Main quantitative result:
  - "A 1 percentage point increase in economic growth among a country’s trading partners, keeping all else equal, is correlated with an increase in domestic growth of as much as 0.8 percentage points."
- Selected trading-partner growth coefficients (Table references as reported):
  - 0.60 (t = 5.71)
  - 0.82 (t = 7.08)
  - 0.73 (t = 5.51)
  - 0.84 (t = 7.20)
  - 0.69 (t = 5.33)
  - 0.38 (t = 1.93) in one specification including interaction term
  - 0.43 (t = 2.71), 0.44 (t = 2.77), 0.38 (t = 2.30) when controlling for distance-weighted growth
  - 0.69 (t = 5.27) for 1980–99; 0.64 (t = 2.91) in a 1960–99 robustness column
- Interaction with openness:
  - Interaction term of trading partners’ growth × Trade/GDP reported as 0.01 (t-statistics ~2.3–2.5) in several specifications.
  - Interpretation given: "with every 10 percent increase in the trade share, the impact of a 1 percent increase in trading partners’ growth on home growth increases by 0.1 percentage points."

### Key empirical findings — trading-partner income (relative income) effects
- Ratio of domestic GDP per capita to trading partners’ GDP per capita:
  - Negative and statistically significant coefficients reported:
    - -1.45 (t = -3.87) in Table 4, regression (2)
    - -1.29 (t = -3.45) and -1.28 (t = -2.73) in other specifications
    - -1.24 (t = -3.12) and -1.26 (t = -3.18) when controlling for distance-weighted growth (Table 5)
  - Example interpretation: a rise in trading partner GDP that lowers the ratio of domestic to foreign GDP by 10 percentage points is correlated with an increase in domestic growth of 0.13 percentage points.
- Direct level of trading partners’ GDP per capita:
  - Reported as positive but not statistically significant in one specification: 0.68 (t-statistic 1.36) in Table 4, regression (1).

### Regional versus global effects
- Distance-weighted growth coefficients (Table 5):
  - 0.57 (t = 3.61), 0.55 (t = 3.49), 0.54 (t = 3.37).
- Once distance-weighted (regional) and trading-partner growth are included, world GDP per capita growth is not statistically significant (Table 5).
- Conclusion: trading-partner effects are not simply global or regional trends; trading partners matter beyond common global or regional shocks.

### Robustness and subsample findings
- Robustness:
  - Results robust to pooled-panel and cross-section estimations.
  - Robust to alternative fixed-period weights based on early 1990s.
- Period and openness subsamples:
  - Period 1980–99: larger estimated coefficients for trading partners’ growth and relative GDP per capita when interaction terms are included (Table 6: trading-partner growth 0.69).
  - Open vs. closed economies (Sachs-Warner openness definition):
    - Trading partners’ growth coefficient positive and significant for both groups: 0.72 for open economies and 0.80 for closed economies in reported 1960–99 split (Table 6).
    - Relative GDP per capita effect is significant only for open economies in some specifications.
- Sensitivity:
  - Excluding regions (e.g., East Asia) does not drive results (details available from authors).

### Quantitative highlights and model fit
- Correlation of trade weights over time:
  - Successive five-year periods: 0.93.
  - Successive ten-year periods: 0.88.
  - First half of 1960s vs. second half of 1990s: 0.70 overall; 0.88 for industrial, 0.66 for developing countries.
- Selected t-statistics and coefficients (reiterated):
  - Trading-partner growth: values reported above with corresponding t-statistics.
  - Relative income ratio: -1.45 (t = -3.87); -1.29 (t = -3.45); -1.28 (t = -2.73).
  - Distance-weighted growth: ~0.57, 0.55, 0.54 (t-statistics reported above).
  - Trade/GDP interaction with trading-partner growth: 0.01 (t-statistics ~2.3–2.5).
- Adjusted R-squared ranges in reported tables:
  - Table 3: 0.31 to 0.47.
  - Table 4: 0.46–0.47.
  - Table 5: 0.47–0.48.
  - Table 6: 0.50–0.63 (varies by subsample).

### Interpretation and mechanisms
- Two distinct influences from trading partners identified:
  - Growth effect: faster-growing trading partners raise demand for a country’s exports and facilitate spillovers.
  - Relative income (conditional convergence) effect: trading with relatively richer partners correlates with faster domestic growth because conditional convergence is stronger when partners are richer relative to the home country.
- Possible channels discussed:
  - Aggregate demand effects from trading partners’ expansion.
  - Technological and R&D spillovers, specialization in technologically advanced sectors, and higher import content of advanced inputs for sectors exporting to advanced partners.
  - Trade-induced spillovers appear larger for more open economies and in more recent decades; world trade share rose from 26 percent in the 1960s to 42 percent in the 1990s.

### Overall conclusions
- A country’s growth is positively associated with both:
  - The growth rate of its trading partners, and
  - The relative income of its trading partners (trading with relatively richer partners correlates with faster domestic growth).
- These associations hold after controlling for world growth, non-trading-partner growth, and distance-weighted regional growth, indicating effects beyond global or regional shocks.
- The effects appear stronger for more open economies and for more recent decades (1980–99), consistent with increased global integration.

*Source: _wp0426 - conclusions of the trade and growth literature, as well as with those of a few recent papers that (PDF; regressions and tables reported for 1960–99; data sources: IMF Direction of Trade Statistics; World Development Indicators).*

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

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

### Listed Tables
- 1.   Most Important Trading Partners and Number of Countries for Which Each of Them Is Among the 10 Most Important Trading Partners ...........................................................13
- 2.   Ranking Economies by Their Trading Partners’ per Capita GDP Growth........................14
- 3.   Growth and Trading Partners’ Growth: Fixed Effects Panel Regressions, 1960–99 ........15
- 4.   Growth and Trading Partners’ Income: Fixed Effects Panel Regressions, 1960–99.........16
- 5.   Growth and Trading Partner Growth: Fixed Effects Panel Regressions, Controlling for Regional Trends, 1960–99 ...........................................................................................17
- 6.   Growth and Trading Partners: Fixed Effects, Panel Regressions, Robustness Tests ........18

### Introduction — research question and approach
- Research question: How much does a country’s long-term economic growth depend on economic conditions in the rest of the world?
- Literature context:
  - General conclusion cited: trade openness has a positive impact on growth.
  - Common view: with growing economic integration, developments in a country are significantly influenced by developments abroad.
  - Noted gap: quantification of the relationship between foreign economic conditions and domestic economic growth has been relatively neglected.
- Approach: empirical analysis using panel data to capture the strength of the impact over the longer run rather than the short-run business-cycle horizon.

### Key empirical findings
- Sample and period:
  - Panel data for the period 1960–99.
  - 101 industrial and developing economies.
- Main quantitative result:
  - "A 1 percentage point increase in economic growth among a country’s trading partners, keeping all else equal, is correlated with an increase in domestic growth of as much as 0.8 percentage points."
- Interpretation:
  - A country’s economic growth is positively influenced by both:
    - the relative income level of its trading partners, and
    - the growth rate of its trading partners,
    after controlling for other growth determinants.
  - Implications:
    - Developing countries benefit from trading with industrial countries, which have relatively higher income levels.
    - Industrial countries benefit from trading with fast-growing developing countries.
- Consistency:
  - The positive sign of the relationship is consistent with the broader literature linking trade openness and growth.

_References section of IMF Working Paper (as presented in the source content)._

### conclusions of the trade and growth literature, as well as with those of a few recent papers that

### _wp0426 - conclusions of the trade and growth literature, as well as with those of a few recent papers that

### Research question and scope
- Focus: whether economic conditions in a country’s trading partners matter for its growth (as distinct from whether openness per se matters).
- Sample: 101 countries, period 1960–99; observations are five-year averages (initial GDP per capita takes first-year value of each five-year period).
- Trade weights: time series of export shares constructed from IMF Direction of Trade Statistics (IMF, 2002) for 1960–99; average share over a five-year period is used.

### Empirical framework and key variables
- Estimation: fixed-effects panel regressions (Hausman test justifies fixed effects for most specifications); pooled panel and cross-section estimates used for robustness.
- Baseline growth regression specification:
  - Dependent variable: real GDP per capita growth (1995 constant US$).
  - Standard control variables: ln(initial GDP per capita), population growth, investment/GDP, secondary school enrollment, inflation, trade/GDP.
  - Additional variables of interest: trading partners’ real per capita GDP growth; ratio of domestic real per capita GDP to trading partners’ real per capita GDP (and alternatively trading partners’ real per capita GDP); interaction terms with trade/GDP; world growth, non-trading-partner growth, and distance-weighted growth to capture regional/global trends.

### Main empirical findings
- Impact of trading partners’ growth on domestic growth:
  - Higher growth in trading partners is positively correlated with domestic growth. Estimates across regressions:
    - As large as 0.82 (Table 3, regression (2)) — a 1 percent higher trading-partner growth correlated with up to 0.8 percent higher domestic growth.
    - 0.60 (Table 3, regression (1)); 0.73, 0.84, 0.69 in other specifications (Table 3).
    - When controlling for distance-weighted growth, the coefficient falls to about 0.43–0.44 (Table 5, regressions (1) and (2)) and 0.38 in one specification (Table 5, regression (3)).
  - Interaction with openness: interaction term of trading partners’ growth × Trade/GDP is positive and small (0.01 in Table 3, regression (6); statistically significant in some specifications). Interpretation: the positive impact of trading partners’ growth on home growth increases with openness; specifically, “with every 10 percent increase in the trade share, the impact of a 1 percent increase in trading partners’ growth on home growth increases by 0.1 percentage points.”
- Impact of trading partners’ income level (relative incomes):
  - Ratio of domestic GDP per capita to trading partners’ GDP per capita has a negative and statistically significant coefficient (Table 4: -1.45 in regression (2); -1.29 and -1.28 in regressions (3) and (4)).
  - Quantified example: a rise in trading partner GDP that lowers the ratio of domestic to foreign GDP by 10 percentage points is correlated with an increase in domestic growth of 0.13 percentage points.
  - Direct level of trading partners’ GDP per capita (unscaled) is positive but not statistically significant in the reported specification (Table 4, regression (1): 0.68, t-statistic 1.36).
- Regional vs. global effects:
  - Distance-weighted growth (closer countries weighted more heavily) has a positive and statistically significant coefficient of about 0.57, 0.55, 0.54 across Table 5 regressions.
  - Once regional (distance-weighted) and trading-partner growth are included, world GDP per capita growth is not statistically significant (Table 5).
  - Conclusion: trading-partner effects are not simply global or regional trends; trading partners matter beyond common global or regional shocks.
- Robustness and subsample results:
  - Results robust to pooled-panel and cross-section estimations; robust to alternative fixed-period weights based on early 1990s.
  - Period 1980–99: larger estimated coefficients for trading partners’ growth and relative GDP per capita when interaction terms are included (Table 6, 1980–99: trading-partner growth 0.69).
  - Open vs. closed economies (Sachs-Warner openness definition): trading partners’ growth coefficient positive and significant for both groups — 0.72 for open economies and 0.80 for closed economies in the 1960–99 split reported (Table 6). Relative GDP per capita effect is significant only for open economies in some specifications.
  - Excluding regions (e.g., East Asia) does not drive results (details available from authors).

### Interpretation and mechanisms
- Two distinct influences from trading partners:
  - Growth effect: faster-growing trading partners raise demand for a country’s exports and facilitate spillovers (corresponds to the positive trading-partner growth coefficients).
  - Relative income (conditional convergence) effect: trading with relatively richer partners is positively correlated with domestic growth because conditional convergence is stronger the richer a country’s trading partners are relative to itself (negative coefficient on domestic-to-partner GDP per capita ratio implies countries with relatively poorer domestic incomes grow faster when trading partners are richer).
- Possible channels:
  - Aggregate demand effects from trading partners’ expansion.
  - Technological and R&D spillovers, specialization in technologically advanced sectors, and higher import content of advanced inputs for sectors exporting to advanced partners.
  - Trade-induced spillovers appear larger for more open economies and in more recent decades (consistent with increased global integration: world trade share rose from 26 percent in the 1960s to 42 percent in the 1990s).

### Quantitative highlights (selected coefficients and statistics)
- Correlation of trade weights over time:
  - Successive five-year periods: 0.93.
  - Successive ten-year periods: 0.88.
  - First half of 1960s vs. second half of 1990s: 0.70 overall; 0.88 for industrial, 0.66 for developing countries.
- Trading-partner growth coefficients (selected):
  - 0.60 (t = 5.71), 0.82 (t = 7.08), 0.73 (t = 5.51), 0.84 (t = 7.20), 0.69 (t = 5.33) — Table 3.
  - 0.38 (t = 1.93) in one specification including interaction term — Table 3.
  - 0.43 (t = 2.71), 0.44 (t = 2.77), 0.38 (t = 2.30) when controlling for distance-weighted growth — Table 5.
  - 0.69 (t = 5.27) for 1980–99; 0.64 (t = 2.91) in a 1960–99 robustness column — Table 6.
- Relative income ratio coefficient:
  - -1.45 (t = -3.87) in Table 4, regression (2); -1.29 (t = -3.45) and -1.28 (t = -2.73) in other specifications — Table 4.
  - -1.24 (t = -3.12) and -1.26 (t = -3.18) when controlling for distance-weighted growth — Table 5.
  - In Table 6, ratio estimates reported include -0.98 (t = -1.90) for 1980–99 and -4.15 (t = -3.95) in one 1960–99 column (sample/specified differences across columns).
- Distance-weighted growth coefficient: ~0.57 (t = 3.61), 0.55 (t = 3.49), 0.54 (t = 3.37) — Table 5.
- Trade/GDP interaction with trading-partner growth: 0.01 (t-statistics ~2.3–2.5) in several specifications — Tables 3 and 4.
- Adjusted R-squared values in reported tables:
  - Table 3: 0.31 to 0.47 depending on specification.
  - Table 4: 0.46–0.47.
  - Table 5: 0.47–0.48.
  - Table 6: 0.50–0.63 (varies by subsample).

### Overall conclusions (paper’s concluding points)
- A country’s growth is positively associated with both:
  - The growth rate of its trading partners, and
  - The relative income of its trading partners (trading with relatively richer partners correlates with faster domestic growth).
- These associations hold after controlling for world growth, non-trading-partner growth, and distance-weighted regional growth, indicating effects beyond global or regional shocks.
- The effects appear stronger for more open economies and for more recent decades (1980–99), consistent with increased global integration.

*Source: _wp0426 - conclusions of the trade and growth literature, as well as with those of a few recent papers that (PDF; regressions and tables reported for 1960–99; data sources: IMF Direction of Trade Statistics; World Development Indicators).*

### References

### _wp0426 - References

### References

- Agénor, Pierre-Richard, John McDermott, and Eswar Prasad, 1999, “Macroeconomic 
Fluctuations in Developing Countries: Some Stylized Facts,” IMF Working Paper, 
WP/99/35. 

- Ahmed, Shaghil, and Prakash Loungani, 1999, “Business Cycles in Emerging Market 
Economies,” Monetaria, Volume XXII, Number 4, October-December, Centro de 
Estudios Monetarios Latinoamericanos (CEMLA). 

- Arora, Vivek and Athanasios Vamvakidis, 2003, “The Impact of U.S. Economic Growth on 
the Rest of the World: How Much Does it Matter?” Journal of Economic Integration, 
forthcoming. 

- Baldwin, Robert E., 2003, “Openness and Growth: What’s the Empirical Relationship?,”  
 NBER Working Paper 9578. 

- Barro, Robert, and Xavier Sala-i-Martin, 1995, “Economic Growth,” New York: McGraw 
Hill. 

- Ben-David, Dan, 1993, “Equalizing Exchange: Trade Liberalization and Income 
Convergence,” The Quarterly Journal of Economics 108, 653-679. 

- Bhagwati, Jagdish N., and T. N. Srinivasan, 1985, “Trade Policy and Development,” in  
 Dependence and Interdependence, Edited by Gene Grossman, Essays in Development 
Economics, Series 2, Cambridge, Mass.: MIT Press. 

- ———, 2002, “Trade and Poverty in Poor Countries,” American Economic Review 92, 
180-83 

- Brunner, Allan D., 2003, “The Long-Run Effects of Trade on Income and Income Growth,”  
 IMF Working Paper, WP/03/37.  

- Caselli, Francesco, Gerardo Esquivel. and Fernando Lefort, 1996, “Reopening the 
Convergence Debate: A New Look at Cross-Country Growth Empirics,” Journal of  
 Economic Growth 1, 363-89. 

- Coe, David, and Elhanan Helpman, 1995, “International R&D spillovers,” European 
Economic Review 39, 859-887. 

- ———, and Alexander Hoffmaister, 1997, “North-South R&D Spillovers,” The Economic 
Journal 107, 134-149. 

- Dollar, David, 1992, “Outward-Oriented Developing Economies Really do Grow More 
Rapidly: Evidence from 95 LDCs, 1976-1985,” Economic Development and Cultural 
Change 40, 523-544. 

- Economist, 2002, “United We Fall,” September 26. 

- Edwards, Sebastian, 1998, “Openness, Productivity and Growth: What Do we Really  
 Know?” Economic Journal 108, 383-398.   

- Frankel, Jeffrey A., and David Romer, 1999, “Does Trade Cause Growth?” American  
 Economic Review, 89, 379-399. 

- Greenaway, David, Wyn Morgan, and Peter Wright, 1998, “Trade Reform, Adjustment and 
Growth: What does the Evidence Tell us?” Economic Journal, 108, 1547-1561.    

- Harrison, Ann, 1996, “Openness and Growth: A Time-series, Cross-Country Analysis for 
Developing Countries,” Journal of Development Economics, 48, 419–447. 

- Levine, Ross, and David Renelt, 1992, “A Sensitivity Analysis of Cross-Country Growth 
Regressions,” American Economic Review, September, 82, 942–63. 

- Navaretti, Giorgio Barba, and  David Tarr, “International Knowledge Flows and Economic 
Performance: A Review of the Evidence,” World Bank Economic Review 14, 1-15. 

- Rodriguez, Francisco, and Dani Rodrik, 1999, “Trade Policy and Economic Growth: A 
Skeptic’s Guide to the Cross-National Evidence,” NBER Working Paper 7081. 

- Sachs, Jeffrey D., and Andrew Warner, 1995, “Economic Reform and the Process of Global 
            Integration,”            Brooking            Papers of Economic Activity, 1-95. 

- Spilimbergo, Antonio, 2000, “Growth and Trade: The North Can Lose,” Journal of Economic 
Growth 5, 131-46. 

- Vamvakidis,  Athanasios,  1998,  “Regional  Integration  and  Economic  Growth,”  World  Bank  
Economic Review 12 (2), 251-270. 

- ———, 1999, “Regional Trade Agreements or Broad Liberalization: Which Path Leads to 
Faster Growth?,” IMF Staff Papers, No. 46 (March), p. 42-68. 

- ———, 2002, “How Robust is the Growth-Openness Connection? Historical Evidence,” 
Journal of Economic Growth, 7:1, 57-80. 

- Warner, Andrew, 2002, “Once More into the Breach: Economic Growth and Global 
Integration,” Harvard University, unpublished manuscript. 

- Williamson, Jeffrey and Michael Clemens, 2002, “Why did the Tariff-Growth Correlation 
Reverse After 1950?,” NBER Working Paper 9181.

*Source: _wp0426 - References*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0426.pdf_