## _wp1259

## Source details

**Canonical URL:** [_wp1259](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1259.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1259.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1259.pdf.json)

---

### I. Introduction — purpose and contributions
- Research question: whether increases in the “sophistication” of a country’s production and exports (goods and services) matter for economic development and growth.
- Four contributions:
  - Systematically documents changes in export sophistication over the past 20 years in low-income countries (LICs) and middle-income countries (MICs), including regional and natural-resource–rich differences.
  - Explicitly analyzes to what extent increasing sophistication of production and exports translates into overall economic growth, and through which mechanisms.
  - Examines what enables a country to increase production sophistication (institutional factors, structural reforms, policy measures).
  - Extends the analysis beyond goods to include a new measure of the sophistication of services (modern services vs. traditional services).
- Framing points:
  - Sophistication measure based on whether products exported by a country are typically exported by high-income or low-income economies.
  - Goods and services analyzed separately because services are “gradually becoming more productive, tradable, and unbundled” and can follow alternative development paths.

### II. Stylized facts — structure and dynamics of exports
- Aggregate and trade magnitudes:
  - Services account for 70 percent of total output.
  - In most developing countries, services account for 55 percent of total GDP growth (post–2000 contribution highlighted).
  - Recorded cross-border service exports in 2007 amounted to $3.3 trillion, or 20 percent of total world trade.
  - The share of services in trade rises to almost 50 percent if transactions are measured in value-added rather than in gross terms.
  - The sale of services by foreign affiliates of multinational firms for fifteen OECD countries is about $1.5 trillion in 2007.
- Evolution of export composition:
  - HICs and fast-growing economies moved from resource-based to manufactured exports.
  - LICs and SSA exports remain dominated by natural resources and primary products; LIC export baskets are relatively static and have moved into relatively few new products.
  - Modern services expanded post–late 1990s in many countries; LICs and SSA have largely failed to share in these trends.
- Selected region- and country-specific figures:
  - By the end of the sample period, manufactures accounted for almost 90 percent of China’s goods exports.
  - Resource-based products account for 90 percent of SSA’s exports of goods.
  - SSA’s manufactures: 7 percent of total goods exports in the early 1980s, expanding to 13 percent by 2009.
  - In other LICs, the share of manufactures in total exports of goods decreased over time to just over 20 percent.
  - India’s share of computer information services in the service export basket grew from 0 to 51 percent by 2009.
- RCA-based product-classification used: “Classical”, “Disappearing”, “Marginal”, “Emerging” (definitions preserved as in source).

### III. Export sophistication — measurement and patterns
- Measurement framework:
  - EXPY/PRODY framework (Hausmann, Hwang and Rodrik, 2007) applied to goods, manufactured goods, and services.
  - PRODY: productivity level associated with a product equals weighted average of per capita GDPs, weights from revealed comparative advantage/value-shares.
  - EXPY: country-level export sophistication equals weighted average of PRODYs for products in the country‟s export basket.
  - PRODYs held constant at their average during 2005–09 (static PRODYs); EXPYs constructed for each country-year.
- Broad patterns:
  - Export sophistication for goods, manufactures, and services has risen over time.
  - Levels of sophistication are generally low in LICs and especially in SSA.
  - Goods and manufacturing sophistication rose steadily in Asia, China, and India; SSA lags behind.
  - Service-export sophistication grew rapidly post–2000; India stands out as relatively sophisticated in services; China displays average sophistication in services.
- Correlates:
  - Positive partial correlations between sophistication and tertiary years of schooling, information flows, and external liberalization.
  - Countries exporting more high-technology goods and services tend to be more sophisticated exporters.
- PRODY by category (Appendix Table 2):
  - Primary Products: 10,425
  - Resource Based: 14,827
  - Low Tech: 13,635
  - Medium Tech: 19,070
  - High Tech: 21,814
- Service-category mean PRODYs and selected annual series (Appendix Table 3):
  - Transportation — Mean: 11,563; Standard Deviation: 792; 2004: 12,656
  - Travel — Mean: 8,608; Standard Deviation: 381; 2004: 9,086
  - Communication — Mean: 6,886; Standard Deviation: 611; 2004: 7,932
  - Construction — Mean: 9,405; Standard Deviation: 799; 2004: 9,494
  - Insurance — Mean: 13,055; Standard Deviation: 2,232; 2004: 14,894
  - Financial — Mean: 24,158; Standard Deviation: 1,053; 2000: 24,409
  - Computer & Information — Mean: 18,585; Standard Deviation: 874; 2000: 19,852
  - Royalties & License Fees — Mean: 14,480; Standard Deviation: 1,072
  - Other business services — Mean: 11,921; Standard Deviation: 1,024
  - Personal, cultural and recreational services — Mean: 14,089; Standard Deviation: 529
- Summary statistics (Appendix Table 4):
  - GOODS EXPY examples:
    - 1980: Mean 9,706; Standard Deviation 4,592
    - 2000: Mean 12,182; Standard Deviation 4,586
    - 2007: Mean 10,004; Standard Deviation 2,012
  - SERVICE EXPY examples:
    - 1990: Mean 10,768; Standard Deviation 4,783
    - 2000: Mean 9,921; Standard Deviation 1,822
    - 2009: Mean 12,714; Standard Deviation 4,848

### IV. Economic growth and export sophistication — empirical evidence
- Data and estimation:
  - Dynamic panel estimation using over 100 countries over 1990–2008.
  - Dependent variable: three-year average growth rate of output per capita.
  - Main regressors: sophistication measures for goods exports and for service exports (analyzed separately).
  - Controls include: initial income per capita, human capital (total years of schooling), external liberalization (composite KOF index with weights: hidden import barriers 22%; mean tariff rate 28%; taxes on international trade/current revenue 27%; capital account restrictions 23%), financial development (private sector credit/GDP), country and time fixed effects.
  - Export sophistication normalized by multiplying by the relevant (goods or services) export ratio.
- Main growth effects:
  - Initial export sophistication of both goods and services is associated with subsequent output growth, controlling for standard determinants.
  - A one standard deviation increase in sophistication:
    - goods → 0.6 percentage points increase in average annual growth rate.
    - services → 0.4 percentage points increase in average annual growth rate.
  - If developing countries raised sophistication to levels observed in advanced economies:
    - goods → per capita growth rate would increase by 1.1 percentage points.
    - services → per capita growth rate would increase by 0.5 percentage points.
  - Regression highlights (selected coefficients from Tables 2 and 3):
    - Table 2 (Goods EXPY): Log initial Goods EXPY coefficients range including 0.65*; 0.654*; 0.663*; 0.694*; 0.843*; 0.626*.
    - Table 3 (Service EXPY): Log initial Service EXPY coefficients include 0.721***; 0.46***; 0.447**; 0.455**; 0.322*; 0.537**.
    - Years of Schooling and External Liberalization frequently positive and significant across specifications; Financial Development coefficients not uniformly positive.
- Heterogeneity and interactions:
  - Hypothesis: growth impact of sophistication stronger with appropriate macro policy, liberalized markets, and good information flows.
  - Additional measures:
    - Overall Macroeconomic Policy proxied by index of real exchange rate overvaluation (asymmetric, censoring overvaluation below zero).
    - Information Flows composite KOF index (internet users per 1000 people 36%; televisions per 1000 people 37%; trade in newspapers/GDP 28%).
  - Findings:
    - Goods EXPY interactions (Table 4):
      - Overvaluation * Goods EXPY: -3.585*** (standard error (0.55)).
      - External Liberalization * Goods EXPY: 1.633*** (standard error (0.46)).
    - Service EXPY interactions (Table 5):
      - Information Flows * Service EXPY: 2.232*** (standard error (0.21)).
      - External Liberalization * Service EXPY: 0.994*** (standard error (0.42)).
  - Interpretation:
    - Sophistication of goods exports has greater growth impact when economy is liberalized and macroeconomic policy is appropriate; overvaluation reduces positive spillovers.
    - Sophistication of service exports has greater growth impact when economy is liberalized and when information flows are good.

### V. Determinants of export sophistication (goods vs. services)
- Methodology:
  - FMOLS for heterogeneous cointegrated panels when non-stationarity detected.
  - Determinants: human capital (total years; tertiary years for services), external liberalization, overall macroeconomic policy, information flows.
- Core findings:
  - Educated workforce, external liberalization, and good information flows are significantly associated with greater sophistication of exports of both goods and services.
  - Appropriate macroeconomic policy particularly important for sustaining sophisticated goods exports; real exchange rate overvaluation reduces goods-sophistication, undervaluation offers no offsetting benefits.
  - Highly skilled labor and good information flows especially critical for sophisticated service exports.
- Magnitudes and selected FMOLS results:
  - Table 6 (Sophistication of Goods Exports):
    - Log GDP per capita coefficients: 0.97***; 0.47***; 0.60***; 0.73***; 0.87*** (t-statistics in parentheses preserved in source).
    - Years of Schooling: 0.34***; 0.34***; 0.17***; 0.10***.
    - External Liberalization: 0.29; 0.03; 0.06**.
    - Overvaluation: -0.02; -0.09***.
    - Information Flows: 0.17**.
    - Observations: 684; Number of Countries: 38.
  - Table 7 (Sophistication of Service Exports):
    - Log GDP per capita: 1.52***; 0.86***; 0.65***; 0.76***; 0.49***.
    - Years of Tertiary Schooling: 0.40***; 0.40***; 0.32***; 0.34***.
    - External Liberalization: 0.02***; 0.02***; 0.02***.
    - Overvaluation: 0.01***; 0.00***.
    - Information Flows: 0.50***.
    - Observations: 684; Number of Countries: 38.
- Elasticities and counterfactual gap reductions (preferred specifications preserved exactly):
  - Goods sophistication:
    - A one standard deviation increase in human capital → 0.1 standard deviation increase in goods-export sophistication (preferred specification: Table 6, column 5).
    - A one standard deviation increase in external liberalization → 0.06 standard deviation increase in goods-export sophistication (preferred specification: Table 6, column 5).
    - If developing economies raised total years of schooling to advanced-economy levels → gap in goods-export sophistication would shrink by 15 percent.
    - If developing economies raised external liberalization to advanced-economy levels → gap in goods-export sophistication would shrink by 8 percent.
  - Services sophistication:
    - A one standard deviation increase in tertiary human capital → 0.34 standard deviation increase in services-export sophistication (preferred specification: Table 7, column 5).
    - A one standard deviation increase in information flows → 0.5 standard deviation increase in services-export sophistication (preferred specification: Table 7, column 5).
    - If developing economies raised tertiary schooling to advanced-economy levels → gap in services-export sophistication would shrink by 42 percent.
    - If developing economies raised information flows to advanced-economy levels → gap in services-export sophistication would shrink by 53 percent.

### VI. Conclusions — summary insights and policy implications
- Stylized facts reiterated:
  - Rich and fast-growing countries shifted from resource-based to manufactured exports; LICs and SSA remain relatively resource-dependent with static export baskets.
  - Services are an increasingly important component of output growth and export performance; many countries increased modern services, but LICs and SSA largely have not shared these gains.
  - Export sophistication of both manufactures and services has increased over time but less so in LICs and particularly in SSA.
- Policy-relevant conclusions:
  - Increasing sophistication of exports (goods and services) can be an important contributor to overall economic growth.
  - Sophisticated export sectors are more likely to generate broader economy-wide growth when:
    - the economy is liberalized;
    - the exchange rate is not over-valued (overvaluation reduces spillovers);
    - information flows are good (enhancing spillovers and diffusion).
  - Determinants of sophistication:
    - Educated workforce, external liberalization, and good information flows are critical for increasing export sophistication.
    - Appropriate macroeconomic policy is particularly important for sophisticated goods exports.
    - Highly skilled labor and good information flows are especially critical for sophisticated service exports.
- Research directions:
  - Explore alternative measures of sophistication (e.g., whether goods or services are typically consumed rather than produced in advanced economies) to test robustness.

*Source: _wp1259 - References (paper text provided).*

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

### _wp1259 - References .............................................................................................................

### I. Introduction — purpose and contributions
- Frames the research question: whether increases in the “sophistication” of a country’s production and exports (goods and services) matter for economic development and growth.
- Four stated contributions:
  - Systematically documents changes in export sophistication over the past 20 years in low-income countries (LICs) and middle-income countries (MICs), including regional and natural-resource–rich differences.
  - Explicitly analyzes to what extent increasing sophistication of production and exports translates into overall economic growth, and through which mechanisms.
  - Examines what enables a country to increase production sophistication (institutional factors, structural reforms, policy measures).
  - Extends the analysis beyond goods to include a new measure of the sophistication of services (modern services vs. traditional services).

### Key framing points (as stated)
- The paper adopts a measure of sophistication based on whether products exported by a country are typically exported by high-income or low-income economies.
- Goods and services are analyzed separately because services are “gradually becoming more productive, tradable, and unbundled” and can follow alternative development paths (e.g., China: manufacturing-led; India: service-export–led).

### II. Stylized facts — structure and dynamics of exports
- Services and manufacturing in the global economy:
  - Services account for 70 percent of total output.
  - In most developing countries, services account for 55 percent of total GDP growth (post–2000 contribution highlighted).
- Cross-border and related service-trade magnitudes:
  - Recorded cross-border service exports in 2007 amounted to $3.3 trillion, or 20 percent of total world trade.
  - The share of services in trade rises to almost 50 percent if transactions are measured in value-added rather than in gross terms.
  - The sale of services by foreign affiliates of multinational firms for fifteen OECD countries is about $1.5 trillion in 2007.

- Evolution of export composition (goods and services):
  - HICs and fast-growing economies have moved from resource-based to manufactured exports.
  - LICs and SSA exports remain dominated by natural resources and primary products; LIC export baskets are relatively static and have moved into relatively few new products.
  - Modern services expanded in many countries post–late 1990s, but LICs and SSA have largely failed to share in these trends.

- Selected region- and country-specific figures:
  - By the end of the sample period, manufactures accounted for almost 90 percent of China’s goods exports.
  - Resource-based products account for 90 percent of SSA’s exports of goods.
  - SSA’s manufactures: 7 percent of total goods exports in the early 1980s, expanding to 13 percent by 2009.
  - In other LICs, the share of manufactures in total exports of goods decreased over time to just over 20 percent.
  - India’s share of computer information services in the service export basket grew from 0 to 51 percent by 2009.

- Classification of export-product dynamics (RCA-based):
  - Products are classified as “Classical”, “Marginal”, “Disappearing”, and “Emerging” based on Revealed Comparative Advantage (RCA) in 1990–94 and 2005–09:
    - Classical: RCA >1 in both 1990–94 and 2005–09.
    - Disappearing: RCA >1 in 1990–94 and <1 in 2005–09.
    - Marginal: RCA <1 in both periods.
    - Emerging: RCA <1 in 1990–94 and >1 in 2005–09.

### III. Export sophistication — measurement and patterns
- A continuous measure of sophistication is developed for goods, manufactures, and services (building on Hausmann, Hwang and Rodrik, 2007 for goods).
- Intuition: the measure captures whether a country’s export basket consists primarily of products typically exported by high-income economies (more sophisticated) or low-income economies (less sophisticated).
- Broad patterns:
  - Export sophistication for goods, manufactures, and services has risen over time.
  - Levels of sophistication are generally low in LICs and especially in SSA.
  - Distinct trends: goods and manufacturing sophistication rose steadily in Asia, China, and India; SSA lags behind.
  - Service-export sophistication grew rapidly post–2000; India stands out as relatively sophisticated in services, China displays average sophistication in services.
- Correlates:
  - Positive partial correlations between sophistication and (tertiary) years of schooling, information flows, and external liberalization.
  - Countries exporting more high-technology goods and services tend to be more sophisticated exporters.

### IV. Economic growth and export sophistication — empirical evidence
- Data and estimation:
  - Dynamic panel estimation using over 100 countries over 1990–2008.
  - Dependent variable: three-year average growth rate of output per capita (to abstract from business-cycle noise).
  - Main regressors: sophistication measures for overall goods exports and for service exports (analyzed separately due to collinearity).
  - Controls include: initial income per capita, human capital (total years of schooling), external liberalization (composite KOF index with weights: hidden import barriers 22%; mean tariff rate 28%; taxes on international trade/current revenue 27%; capital account restrictions 23%), financial development (private sector credit/GDP), country and time fixed effects.
  - Export sophistication normalized by multiplying by the relevant (goods or services) export ratio.

- Main growth effects (statistical and economic significance):
  - Initial export sophistication of both goods and services is associated with subsequent output growth, controlling for standard determinants.
  - A one standard deviation increase in sophistication:
    - goods → 0.6 percentage points increase in average annual growth rate.
    - services → 0.4 percentage points increase in average annual growth rate.
  - If developing countries raised sophistication to levels observed in advanced economies:
    - goods → per capita growth rate would increase by 1.1 percentage points.
    - services → per capita growth rate would increase by 0.5 percentage points.
  - These magnitudes are based on the preferred specifications in Table 2, column 4, and Table 3, column 4, and remain stable over time and larger in the developing-country sub-sample.

- Heterogeneity and enabling conditions (interaction tests):
  - Hypothesis: the growth impact of sophistication is stronger when macroeconomic policy is appropriate, markets are liberalized, and information flows are good (facilitating reallocation and spillovers).
  - Constructed additional measures:
    - Overall Macroeconomic Policy: proxied by an index of real exchange rate overvaluation (asymmetric specification, censoring overvaluation below zero).
    - Information Flows: composite KOF index (internet users per 1000 people 36%; televisions per 1000 people 37%; trade in newspapers/GDP 28%).
  - Findings:
    - Sophistication of goods exports has a greater impact on growth when the economy is liberalized and macroeconomic policy is appropriate (Table 4).
    - Overvaluation reduces positive growth spillovers from a sophisticated export sector; undervaluation yields no corresponding benefits.
    - Sophistication of service exports has a greater impact when the economy is liberalized and when information flows are good (Table 5).

### V. Determinants of export sophistication (goods vs. services)
- Methodology:
  - Estimation using FMOLS for heterogeneous cointegrated panels when some non-stationarity was detected.
  - Determinants considered: human capital (total years of schooling; tertiary years when focusing on services), external liberalization, overall macroeconomic policy, and information flows.

- Core findings:
  - An educated workforce, external liberalization, and good information flows are significantly associated with greater sophistication of exports of both goods and services across multiple specifications (Tables 6 and 7).
  - Appropriate macroeconomic policy is particularly important for sustaining sophisticated goods exports; real exchange rate overvaluation reduces goods-sophistication, undervaluation offers no offsetting benefits.
  - Highly skilled labor and good information flows are especially critical for sophisticated service exports.

- Magnitudes (selected elasticities/effects preserved exactly as reported):
  - Goods sophistication:
    - A one standard deviation increase in human capital or in external liberalization is associated with a, respectively, 0.1 or 0.06 standard deviation increase in the sophistication of goods exports (preferred specifications: Table 6, column 5).
    - If developing economies raised total years of schooling or external liberalization to levels observed in advanced economies, the gap in goods-export sophistication would shrink by, respectively, 15 percent and 8 percent.
  - Services sophistication:
    - A one standard deviation increase in tertiary human capital or in information flows is associated with, respectively, 0.34 or 0.5 standard deviation increases in the sophistication of services exports (preferred specifications: Table 7, column 5).
    - If developing economies raised tertiary schooling or information flows to levels observed in advanced economies, the gap in services-export sophistication would shrink by, respectively, 42 percent and 53 percent.

### VI. Conclusions — summary insights and policy implications
- Stylized facts reiterated:
  - Rich and fast-growing countries have moved from resource-based to manufactured exports; LICs and SSA remain relatively resource-dependent with static export baskets.
  - Services are an increasingly important component of output growth and export performance; many countries have increased the relative importance of modern services, but LICs and SSA have largely not shared these gains.
  - Export sophistication of both manufactures and services has increased over time but less so in LICs and particularly in SSA.

- Policy-relevant conclusions:
  - Increasing sophistication of exports (goods and services) can be an important contributor to overall economic growth.
  - Sophisticated export sectors are more likely to generate broader economy-wide growth when:
    - the economy is liberalized;
    - the exchange rate is not over-valued (overvaluation reduces spillovers);
    - information flows are good (enhancing spillovers and diffusion).
  - Determinants of sophistication:
    - Educated workforce, external liberalization, and good information flows are critical for increasing export sophistication.
    - Appropriate macroeconomic policy is particularly important for sophisticated goods exports.
    - Highly skilled labor and good information flows are especially critical for sophisticated service exports.

- Research directions:
  - The paper suggests exploring alternative measures of sophistication (e.g., whether goods or services are typically consumed rather than produced in advanced economies) to test robustness.

*Source: _wp1259 - References (paper text provided).*

### REFERENCES

### _wp1259 - REFERENCES

### References (selected)
- Barro, Robert, and Jong-Wha Lee, 2010, “A New Data Set of Educational Attainment in the World, 1950–2010,” NBER Working Paper No 15902.
- Dreher, Axel, 2006, “Does Globalization Affect Growth? Evidence from a New Index of Globalization,” Applied Economics 38 (10): 1091–1110.
- Escaith, Hubert, 2008, “Measuring Trade in Value Added in the New Industrial Economy: Statistical Implications,” MPRA Paper 14454, University Library of Munich, Germany.
- Francois, Joseph, and Bernard Hoekman, 2010, “Services Trade and Policy,” Journal of Economic Literature 48(3): 642–92, September.
- Ghani, Ejaz, and Homi Kharas, 2010, “The Service Revolution in South Asia: An Overview,” in: Ghani, Ejaz, ed., The Service Revolution in South Asia (Oxford University Press).
- Gwartney, James, Joshua Hall, and Robert Lawson, 2010, Economic Freedom of the World: 2010 Annual Report (Vancouver, BC: The Fraser Institute).
- Hausmann, Ricardo, Jason Hwang, and Dani Rodrik, 2007, “What You Export Matters,” Journal of Economic Growth 12(1): 1–25.
- Johnson, Simon, Jonathan Ostry, and Arvind Subramanian, 2010, “Prospects for Sustained Growth in Africa: Benchmarking the Constraints,” IMF Staff Papers 57: 119–171.
- Lall, Sanjaya, 2000, “The Technological Structure and Performance of Developing Country Manufactured Exports, 1985–98,” Oxford Development Studies 28(3).
- Lall, Sanjaya, John Weiss, and Jinkang Zhang, 2005, “The „Sophistication‟ of Exports: A New Measure of Product Characteristics,” Queen Elizabeth House Working Paper No 123, Oxford University.
- Mishra, Saurabh, Susanna Lundstrom, and Rahul Anand, 2011, “Service Export Sophistication and Economic Growth,” World Bank Working Paper No 5606.
- Triplett, Jack, and Barry Bosworth, 2004, Productivity in the U.S. Services Sector: New Sources of Economic Growth (Washington, D.C: Brookings Institution).
- World Bank, 2010, World Development Indicators (Washington, D.C.: World Bank).
- World Trade Organization, 2009, International Trade Statistics 2009 (Geneva: World Trade Organization).

### A. Data Description
- Trade data source: IMF Balance of Payments statistics; sample covers approximately 100 countries during 1990–2007.
- GDP per capita data source: World Bank Indicators database.
- External Liberalization and Hidden Import Barriers: Source Gwartney et al. (2010). Index based on Global Competitiveness Report survey question: “In your country, tariff and non-tariff barriers significantly reduce the ability of imported goods to compete in the domestic market.”
- Taxes on International Trade (percent of current revenue): Source World Bank (2010). Taxes include import duties, export duties, profits of export or import monopolies, exchange profits, and exchange taxes. Current revenue definition provided in full in source.

### Appendix Table 1 — Breakdown of Export Services (categories and explanations)
- Transportation: covers all transportation (sea, air, land, internal waterway, space, pipeline, etc) services performed by residents of one economy for those of another; includes carriage of passengers, goods (freight), rentals of carriers with crew, and related supporting services.
- Travel: covers goods and services acquired during visits (less than a year); consumer moves to provider location; international carriage of traveler covered under transportation; proxy for international tourism, business traveling and international student (even if staying longer than one year).
- Communication: covers (i) telecommunication and (ii) postal and courier between residents and nonresidents international transactions.
- Construction: construction services performed by employees outside enterprise country; includes goods employees bring abroad to perform task; local goods expenditures recorded under Other business services.
- Insurance: insurance provided by a resident to a nonresident and vice versa (often freight insurances and other direct services).
- Financial: financial intermediary and auxiliary services (except insurance enterprises and pension funds) between residents and nonresidents; examples include fees for letters of credit, lines of credit, financial leasing, foreign exchange transactions, transactions in securities, asset management.
- Computer Information: computer data and new-related service transactions between residents and nonresidents (databases, data processing, hardware consultancy, software implementation, maintenance and repair of computers, news agency services, etc).
- Royalties & license fees: payments between residents and nonresidents for use of intangible and nonfinancial assets or property rights (patents, copy rights, franchising, manuscripts, films, etc).
- Other business: includes (i) Merchanting, (ii) Operational leasing without operators (resident-nonresident leasing and charter without crew), and (iii) Miscellaneous services including legal, accounting, management consulting, public relation services; advertising and market research; research and development; architectural, engineering and other technical services; agricultural, mining and on-site processing services; and other services between residents and nonresidents.
- Personal & cultural: divided into (i) audiovisual (services and fees for motion pictures—including to actors and producers, radio and television programs and musical recordings) and (ii) other (museum, library, sporting, correspondence courses, etc).

### B. Constructing the Measure of Export Sophistication (EXPY / PRODY)
- Framework: EXPY constructed using Hausmann, Hwang and Rodrik (2007) framework to capture productivity level associated with a country‟s exports; computed for three categories: goods, manufactured goods, and services.
- Ranking: Each export category is ranked according to income levels of exporting countries; products exported by rich countries rank higher than those exported by poor countries (controlling for overall economic size).
- PRODY definition: productivity level associated with product k in category p equals the weighted average of per capita GDPs, where weights represent the revealed comparative advantage of each country in that product. The weight numerator is the value-share of the product in the country‟s category p export basket; the denominator aggregates value-shares across all countries exporting that product in that category.
- EXPY definition: country-level export sophistication is the weighted average of PRODYs for products in that country‟s export basket, with weights equal to product share in the country‟s exports.
- Construction specifics:
  - EXPYs constructed for each country-year with available data.
  - PRODYs held constant at the average value during 2005–09 (static PRODYs).
  - Interpretation: increases in EXPY reflect shifts from low PRODY to high PRODY products (higher share of high PRODY goods, manufactures, and services).
- Empirical patterns noted:
  - Higher-value-added goods and services have higher recorded PRODY (Appendix Table 2).
  - Modern services (insurance, financial, computer & information, royalties & license fees, other business services, personal, cultural & recreational services) generally have higher PRODY and have been growing more rapidly than traditional services (transportation, travel, communication, construction) (Appendix Table 3).
  - Both mean and standard deviation of goods and services EXPY have increased over the years, indicating increasing cross-country diversity in sophistication and higher potential EXPY (Appendix Table 4).
  - Countries with more sophisticated exports (goods or services) generally enjoy higher GDP per capita (Appendix Figure 1).

### C. Outlier Regressions (qualitative finding)
- Regressions control for: log income per capita; log income per capita squared; size proxied by log population; plus regional dummies.
- Notable outliers:
  - China and India are both positive, and typically statistically significant, outliers.
  - India is a large positive, significant outlier in the sophistication of services.
  - Sub-Saharan Africa (SSA) sophistication was roughly as expected, given its stage of development and size.

### Key regression results — Panel Regressions (selected coefficients and statistics)
- Table 2. Dependent Variable: Growth in Income Per Capita (3-Year Average). Independent Variable: Sophistication of Goods Exports
  - Log initial GDP per capita: -3.969***; -7.866***; -8.211***; -7.216***; -7.974***; -5.448*** (standard errors in parentheses: (0.92) (1.13) (1.15) (1.32) (1.58) (1.24))
  - Log initial Goods EXPY: 0.65*; 0.654*; 0.663*; 0.694*; 0.843*; 0.626* (standard errors: (0.36) (0.37) (0.38) (0.38) (0.44) (0.39))
  - Years of Schooling: 3.798***; 2.611***; 2.533***; 3.835***; 2.799*** (standard errors: (0.78) (0.87) (0.88) (1.08) (1.01))
  - External Liberalization: 1.147***; 1.181***; 2.071***; 0.933** (standard errors: (0.30) (0.30) (0.39) (0.37))
  - Financial Development: -0.508; -0.894**; -0.32 (standard errors: (0.36) (0.40) (0.42))
  - Observations: 630; 527; 486; 483; 398; 308
  - R-squared: 0.039; 0.114; 0.150; 0.155; 0.249; 0.151
  - Number of countries: 153; 123; 111; 111; 110; 78

- Table 3. Dependent Variable: Growth in Income Per Capita (3-Year Average). Independent Variable: Sophistication of Service Exports
  - Log initial GDP per capita: -5.63***; -7.664***; -7.752***; -7.497***; -7.249***; -5.759*** (standard errors: (0.98) (1.16) (1.18) (1.34) (1.61) (1.25))
  - Log initial Service EXPY: 0.721***; 0.46***; 0.447**; 0.455**; 0.322*; 0.537** (standard errors: (0.14) (0.17) (0.18) (0.18) (0.19) (0.22))
  - Years of Schooling: 2.904***; 1.889*; 1.885*; 3.075**; 2.345** (standard errors: (0.87) (0.96) (0.97) (1.19) (1.13))
  - External Liberalization: 1.014**; 1.034***; 2.045***; 0.766** (standard errors: (0.30) (0.30) (0.40) (0.38))
  - Financial Development: -0.116; -0.371; 0.182 (standard errors: (0.34) (0.38) (0.39))
  - Observations: 664; 553; 507; 504; 412; 329
  - R-squared: 0.069; 0.099; 0.127; 0.128; 0.197; 0.132
  - Number of Countries: 152; 123; 111; 111; 110; 80

### Interaction and robustness checks (selected table highlights)
- Table 4. Goods EXPY interactions
  - Log initial GDP per capita: -3.969***; -4.747***; -5.469***; -4.977***; -6.015*** (standard errors shown)
  - Log initial Goods EXPY: 1.656*; 1.857*; 4.247***; 1.824; 0.728
  - Overvaluation * Goods EXPY: -3.585*** (standard error (0.55))
  - External Liberalization * Goods EXPY: 1.633*** (standard error (0.46))
  - Observations: 630; 559; 559; 530; 530
  - R-squared: 0.039; 0.042; 0.130; 0.060; 0.082
  - Number of Countries: 153; 137; 137; 123; 123

- Table 5. Service EXPY interactions
  - Log initial GDP per capita: -5.63***; -7.363***; -9.465***; -6.996*** 
  - Log initial Service EXPY: 0.721***; 0.353**; 0.026; 0.524*** 
  - Information Flows: 1.418***; 0.02 (standard errors (0.31) (0.38))
  - Information Flows * Service EXPY: 2.232*** (standard error (0.21))
  - External Liberalization * Service EXPY: 0.994*** (standard error (0.42))
  - Observations: 664; 664; 664; 553
  - R-squared: 0.069; 0.106; 0.164; 0.088
  - Number of Countries: 152; 152; 152; 122

### Cointegrating panel regressions — institutional / structural / policy factors
- Table 6. Dependent Variable: Sophistication of Goods Exports (FMOLS co-integrated heterogeneous panels)
  - Log GDP per capita: 0.97***; 0.47***; 0.60***; 0.73***; 0.87*** (t-statistics in parentheses: (32.64) (17.47) (14.45) (15.36) (9.13))
  - Years of Schooling: 0.34***; 0.34***; 0.17***; 0.10*** (t-statistics: (5.23) (5.83) (4.82) (5.64))
  - External Liberalization: 0.29; 0.03; 0.06** (t-statistics: (1.54) (0.96) (2.02))
  - Overvaluation: -0.02; -0.09*** (t-statistics: (1.31) (4.02))
  - Information Flows: 0.17** (t-statistic (2.02))
  - Observations: 684
  - Number of Countries: 38

- Table 7. Dependent Variable: Sophistication of Service Exports (FMOLS co-integrated heterogeneous panels)
  - Log GDP per capita: 1.52***; 0.86***; 0.65***; 0.76***; 0.49*** (t-statistics: (72.40) (29.13) (22.41) (20.96) (14.46))
  - Years of Tertiary Schooling: 0.40***; 0.40***; 0.32***; 0.34*** (t-statistics: (14.33) (15.15) (17.11) (19.68))
  - External Liberalization: 0.02***; 0.02***; 0.02*** (t-statistics: (4.62) (4.45) (4.93))
  - Overvaluation: 0.01***; 0.00*** (t-statistics: (3.10) (2.61))
  - Information Flows: 0.50*** (t-statistic (8.04))
  - Observations: 684
  - Number of Countries: 38

*Source: _wp1259 - REFERENCES (Appendix A–C, Tables 1–7).*

### Appendix Table 2. PRODY for Various Categories of Goods and Services

### _wp1259 - Appendix Table 2. PRODY for Various Categories of Goods and Services

### PRODY for Goods and Services (Appendix Table 2)
- GOODS average PRODY by category:
  - Primary Products: 10,425
  - Resource Based: 14,827
  - Low Tech: 13,635
  - Medium Tech: 19,070
  - High Tech: 21,814
- SERVICES average PRODY by category:
  - Transportation: 11,563
  - Travel: 9,405
  - Communication: 14,480
  - Construction: 18,585
  - Insurance: 24,158
  - Financial: (value not separately listed in Appendix Table 2; see Appendix Table 3 for service-category detail)
  - Royalty and License Fees: (see Appendix Table 3)
  - Computer and Information: (see Appendix Table 3)
  - Financial (repeated in Appendix Table 3): (see Appendix Table 3)

### Sophistication of Service Exports by Category (Appendix Table 3)
- Annual values by service category (selected years and summary statistics):
  - Transportation:
    - 2000: 10,516
    - 2001: 11,213
    - 2002: 11,559
    - 2003: 11,870
    - 2004: 12,656
    - Mean: 11,563
    - Standard Deviation: 792
  - Travel:
    - 2000: 8,082
    - 2001: 8,408
    - 2002: 8,696
    - 2003: 8,769
    - 2004: 9,086
    - Mean: 8,608
    - Standard Deviation: 381
  - Communication:
    - 2000: 6,553
    - 2001: 6,552
    - 2002: 6,467
    - 2003: 6,926
    - 2004: 7,932
    - Mean: 6,886
    - Standard Deviation: 611
  - Construction:
    - 2000: 10,422
    - 2001: 9,695
    - 2002: 9,185
    - 2003: 8,232
    - 2004: 9,494
    - Mean: 9,405
    - Standard Deviation: 799
  - Insurance:
    - 2000: 10,856
    - 2001: 10,496
    - 2002: 13,842
    - 2003: 15,187
    - 2004: 14,894
    - Mean: 13,055
    - Standard Deviation: 2,232
  - Financial:
    - 2000: 24,409
    - 2001: 25,160
    - 2002: 24,862
    - 2003: 22,486
    - 2004: 23,871
    - Mean: 24,158
    - Standard Deviation: 1,053
  - Computer & Information:
    - 2000: 19,852
    - 2001: 18,058
    - 2002: 17,906
    - 2003: 17,961
    - 2004: 19,150
    - Mean: 18,585
    - Standard Deviation: 874
  - Royalties & License Fees:
    - 2000: 14,973
    - 2001: 15,318
    - 2002: 14,828
    - 2003: 14,670
    - 2004: 12,611
    - Mean: 14,480
    - Standard Deviation: 1,072
  - Other business services:
    - 2000: 10,763
    - 2001: 11,132
    - 2002: 11,812
    - 2003: 12,725
    - 2004: 13,175
    - Mean: 11,921
    - Standard Deviation: 1,024
  - Personal, cultural and recreational services:
    - 2000: 14,282
    - 2001: 14,613
    - 2002: 13,442
    - 2003: 13,615
    - 2004: 14,491
    - Mean: 14,089
    - Standard Deviation: 529
- Source note: "Source: Authors‟ calculations."

### Summary Statistics: Goods and Services EXPY (Appendix Table 4)
- GOODS EXPY (selected years):
  - 1980: Mean 9,706; Standard Deviation 4,592
  - 1990: Mean 9,235; Standard Deviation 1,799
  - 1995: Mean 9,492; Standard Deviation 1,903
  - 2000: Mean 12,182; Standard Deviation 4,586
  - 2005: Mean 9,865; Standard Deviation 2,000
  - 2007: Mean 10,004; Standard Deviation 2,012
- SERVICE EXPY (selected years):
  - 1990: Mean 10,768; Standard Deviation 4,783
  - 2000: Mean 9,921; Standard Deviation 1,822
  - 2007: Mean 12,509; Standard Deviation 4,645
  - 2009: Mean 12,714; Standard Deviation 4,848
- Source note: "Source: Authors‟ calculations."

### Export Sophistication: Outlier Regressions (Appendix Table 5)
- Regression structure: Columns correspond to (1)-(9) across three panels: Goods Export Sophistication (1991, 2000, 2006), Manufacturing Export Sophistication (1991, 2000, 2006), Service Export Sophistication (1991, 2000, 2006).
- Selected coefficient estimates (with standard errors in parentheses):
  - log GDP per capita:
    - Goods 1991: 0.592 (0.671)
    - Goods 2000: 0.424 (0.454)
    - Goods 2006: 0.705** (0.336)
    - Manufacturing 1991: -0.730 (0.518)
    - Manufacturing 2000: -0.286 (0.316)
    - Manufacturing 2006: -0.193 (0.325)
    - Services 1991: 0.257 (0.247)
    - Services 2000: -0.182 (0.244)
    - Services 2006: -0.168 (0.235)
  - (log GDP per capita)^2:
    - Goods 1991: -0.0148 (0.0381)
    - Goods 2000: -0.00917 (0.0255)
    - Goods 2006: -0.0232 (0.0184)
    - Manufacturing 1991: 0.0527* (0.0289)
    - Manufacturing 2000: 0.0282 (0.0177)
    - Manufacturing 2006: 0.0224 (0.0179)
    - Services 1991: -0.0109 (0.0141)
    - Services 2000: 0.0155 (0.0139)
    - Services 2006: 0.0159 (0.0132)
  - Geographical Size:
    - Goods 1991: 0.0678*** (0.0135)
    - Goods 2000: 0.0408*** (0.0126)
    - Goods 2006: 0.0521*** (0.00913)
    - Manufacturing 1991: 0.00493 (0.0129)
    - Manufacturing 2000: 0.0145 (0.0119)
    - Manufacturing 2006: 0.00450 (0.0111)
    - Services 1991: 0.0179** (0.00724)
    - Services 2000: 0.000860 (0.00995)
    - Services 2006: 0.00146 (0.00806)
  - India dummy:
    - Goods 1991: 0.236** (0.0931)
    - Goods 2000: 0.108 (0.0655)
    - Goods 2006: 0.154** (0.0619)
    - Manufacturing 1991: 0.0479 (0.113)
    - Manufacturing 2000: 0.0597 (0.0752)
    - Manufacturing 2006: 0.186** (0.0759)
    - Services 1991: 0.0911** (0.0390)
    - Services 2000: 0.400*** (0.0671)
    - Services 2006: 0.457*** (0.0516)
  - China dummy:
    - Goods 1991: -0.00385 (0.160)
    - Goods 2000: -0.0320 (0.156)
    - Goods 2006: -0.157 (0.121)
    - Manufacturing 1991: 0.167 (0.171)
    - Manufacturing 2000: 0.105 (0.152)
    - Manufacturing 2006: 0.252* (0.146)
    - Services 1991: -0.0445 (0.0813)
    - Services 2000: 0.0645 (0.129)
    - Services 2006: 0.152 (0.103)
  - SSA dummy:
    - Goods 1991: -0.0618 (0.0958)
    - Goods 2000: -0.116* (0.0676)
    - Goods 2006: -0.00775 (0.0715)
    - Manufacturing 1991: 0.0108 (0.116)
    - Manufacturing 2000: -0.0352 (0.0787)
    - Manufacturing 2006: -0.00707 (0.0821)
    - Services 1991: -0.0660 (0.0413)
    - Services 2000: 0.00821 (0.0605)
    - Services 2006: 0.0461 (0.0501)
  - Constant terms:
    - Goods 1991: 4.646 (2.958)
    - Goods 2000: 6.031*** (2.061)
    - Goods 2006: 4.507*** (1.532)
    - Manufacturing 1991: 11.58*** (2.343)
    - Manufacturing 2000: 9.540*** (1.433)
    - Manufacturing 2006: 9.271*** (1.484)
    - Services 1991: 7.308*** (1.091)
    - Services 2000: 9.508*** (1.097)
    - Services 2006: 9.553*** (1.067)
- Observations and R-squared:
  - Goods regressions:
    - Observations: 1991: 103; 2000: 125; 2006: 122
    - R-squared: 1991: 0.710; 2000: 0.698; 2006: 0.752
  - Manufacturing regressions:
    - Observations: 1991: 103; 2000: 125; 2006: 122
    - R-squared: 1991: 0.363; 2000: 0.552; 2006: 0.576
  - Services regressions:
    - Observations: 1991: 114; 2000: 124; 2006: 118
    - R-squared: 1991: 0.372; 2000: 0.311; 2006: 0.509
- Notes: "These regressions indicate where different regions and countries lie in terms of export sophistication, compared to the global norm, after controlling for the stage of development, size, and other independent variables. *, **, and *** denote significance at the, respectively, 10 percent, 5 percent, and 1 percent level. Robust standard errors are in parentheses."

### Figures and Trends (Selected captions and data highlights)
- Figure 1: Share of Manufacturing and Services in GDP versus Stage of Development, 2008 (axes labeled with 2008 Manufacturing value added (Percent of GDP) and 2008 Services value added (Percent of GDP); 2008 Ln(Per Capita GDP); regional markers: SSA, East Asia, LA-Carib, BRIC).
- Figure 2: Panel A and Panel B present Growth Accounting by Sectors, 1995–09 and 2000–08; source: World Development Indicators, 2011. Series include Service, Industry, Agriculture across regions: Sub Saharan Africa, Upper Middle Income, Developing, Low Income, India, China, USA.
- Figure 3: Export Share in GDP (1990–2006 series) with regional series SSA, HICs, MICs, Other, LICs, Asia (excl. India and China), India, China.
- Figure 4: Share of Goods and Service Exports in Total Exports, 1990 and 2009 (IMF Balance of Payments, 2011) showing Services vs Goods across regions including SSA, HICs, MICs, Other, LICs, Asia (excl. India and China), India, China.
- Figures 5–9: Various breakdowns of manufacturing and resource export composition over periods 80-85 and 07-09; categories include Manufacturing, Primary and Resource Based; High-Tech, Medium-Tech, Low-Tech; and classifications Emerging, Marginal, Disappearing, Classical.
- Figures 10–12: Growth and shares of Modern and Traditional Service Exports (2000–09, 1990 & 2009, 2000–07) with series for World and selected countries/regions (Sub-Saharan Africa, Brazil, China, India, Ireland, United States).
- Figure 13: Service Exports / Service Value Added (percent) time series 1986–2008 for SSA, Other LICs, MICs, China, India.
- Figure 14: Export Sophistication over Time, for Goods, Manufactures, and Services, 1990-2009; vertical scales: 8.6 to 10 (Goods), 8.6 to 9.8 (Manufacturing), 8.6 to 9.8 (Services); country/regional series: SSA, HIC, MIC, LIC, Asia, India, China, Brazil.
- Figures 15–20: Cross-sectional and map visualizations of sophistication of manufactured and service exports versus GDP per capita, world maps of sophistication for goods, manufacturing, services, and relationships with high-technology exports. Notes indicate bubble sizes denote trade ratios and colors represent regions.

*Source: Authors‟ calculations.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2012/_wp1259.pdf_