## _030514

## Source details

**Canonical URL:** [_030514](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_030514.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_030514.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_030514.pdf.json)

---

### EXECUTIVE SUMMARY — overview and main messages
- Diversification and structural transformation play important roles in influencing the macroeconomic performance of low-income countries (LICs).
- Early-stage increases in income per capita are typically accompanied by changes in production and export structures: diversification into new products and trading partners and increases in the quality of existing products.
- Analysis uses a newly-constructed cross-country dataset (1962–2010 time coverage for many series), complemented by country case studies.
- Main empirical conclusions:
  - Export and domestic production diversification are conducive to faster economic growth in LICs.
  - Increased diversification is associated with lower output volatility and greater macroeconomic stability.
  - There is both a growth payoff and a stability payoff to diversification.
  - LICs have historically concentrated on a narrow set of primary products and a small number of export markets, but patterns have been changing since the mid–1990s with marked cross-country and regional variation.
  - LICs as a group have diversified away from agriculture into manufactured goods since the mid–1990s, though substantial scope remains to upgrade export quality, including in agriculture.

### Conceptual framework and measurement
- Export diversification decomposed into:
  - Extensive export diversification: increases in the number of products or trading partners.
  - Intensive export diversification: distribution of export volumes across active products/partners; measured with the Theil index (lower values indicate higher diversification); Theil index is decomposable into extensive and intensive margins.
- Output diversification measured by an output diversification index across seven broad sectors (value-added data).
- Quality upgrading measured via unit values corrected for confounders and expressed relative to the world quality frontier (the 90th percentile in each product-year).
- Structural transformation: reallocation of resources across sectors/products; sectoral productivity = real value added per worker; aggregate labor productivity growth decomposed into sectoral productivity growth and sectoral shifts.

### Key empirical findings and quantitative magnitudes
- Diversification effects on growth:
  - A one standard deviation increase in LICs’ export product diversification raises their growth rate by about 0.8 percentage points.
  - A one standard deviation increase in output diversification in LICs raises their average annual growth rate by about 1.4 percentage points.
- Quality upgrading and growth:
  - Average annual growth was more than 1 percentage point higher in LICs with fast quality convergence compared to LICs with slow quality convergence (post-1995).
  - In Africa only, the growth dividend from quality upgrading exceeds ½ percentage points per annum in the post-1995 period.
  - Small states with fast quality convergence experienced additional annual per capita growth of 2 percentage points relative to slow converging peers.
- Diversification and volatility:
  - Export and output diversifications are associated with lower output volatility; effect mainly through a more evenly balanced product mix (intensive margin).
  - Identified 61 export product diversification spurts in the post–1962 period involving 51 developing countries.
  - Diversification spurts last 13 years on average, rising to 20 years in East Asia and Pacific.
  - In LICs, output volatility diminishes by 0.5 percentage points following diversification spurts (effect in the rest of the sample is about half that in LICs).

### Sectoral patterns, quality ladders, and domestic reallocation
- Sectoral shifts accompanying diversification:
  - East and South Asia: agricultural export share fell from 40 percent in 1965-1970 to 15 percent in 2006-2010; manufactured export share rose from 17 percent to 66 percent over the same periods.
  - SSA: sizable shift away from agricultural exports but only a small rise in manufacturing export share.
- Quality ladder observations:
  - Manufactures typically have the highest potential for quality upgrading; natural resources lower.
  - East Asia and Pacific saw steady increases in export quality since the mid-1980s; SSA experienced a steady decrease in overall export quality driven by agriculture.
  - Quality convergence is most rapid at early development stages and largely completes by upper middle-income status (GDP per capita around $25,000-$30,000 marks reconcentration in advanced economies).
- Domestic sectoral reallocation:
  - Over 1990-2010, economies with greater export diversification and quality upgrading experienced larger reductions in agricultural shares and more productivity-enhancing sectoral reallocation.
  - In recent decades, the decline in agriculture’s share in LIC output was largely absorbed by non-tradable activities (construction, wholesale trade, transportation) rather than manufacturing in many cases.

### Drivers and policy-relevant associations
- Cross-country associations (interpret as correlations, not strict causal claims):
  - Human capital and institutional quality positively associated with export diversification and quality upgrading.
  - Financial depth (private credit to GDP) associated with higher export diversification for developing countries.
  - Globalization (KOF index) and rising South-South trade associated with higher export diversification.
  - Proximity matters: landlocked countries are less diversified.
  - Reforms correlated with diversification: trade liberalization and agricultural reforms (agricultural reforms have large impacts at early development stages).
  - Some evidence real exchange rate (RER) undervaluation correlates with higher diversification, but persistent undervaluation has macroeconomic costs.
- Sectoral productivity drivers (LIC-focused 1995–2010):
  - Manufacturing employment shares associated with liberalized electricity markets, easier credit, and reduced labor/business regulation.
  - Education linked to higher value added shares in manufacturing and services.
  - Reduced average import tariffs associated with labor moving out of agriculture.
  - Agricultural productivity gains associated with removing tariff barriers and reducing interest rate controls.
  - Manufacturing productivity linked to domestic finance reforms, capital account liberalization (FDI), road infrastructure, and education.
  - Services productivity benefits from liberalization of network industries, e.g., telecommunications.
- Policy lessons from case studies:
  - Macroeconomic stabilization, removal of barriers to entry, investment in infrastructure and people, and adaptive reform waves aligned to evolving comparative advantage are common features of successful diversification episodes.
  - Industry-targeted support has mixed outcomes: can boost exports but may increase concentration and risks of policy capture; measures facilitating agglomeration (industrial zones, EPZs) have worked by promoting scale and linkages.

### Practical policy recommendations
- Make diversification and structural transformation central to LIC development strategies to boost growth and reduce volatility.
- Maintain a role for agriculture in development policy:
  - Agriculture remains the main employer for many poor households; productivity and quality improvements in agriculture are essential for inclusive growth.
  - Agriculture offers scope for quality upgrading and higher value-added products.
- Policy instruments with demonstrated associations to diversification and upgrading:
  - Improve infrastructure and trade networks.
  - Invest in human capital.
  - Promote financial deepening.
  - Reduce barriers to entry for new products.
  - Improve governance.
  - Implement agricultural reforms.
- Policy sequencing and country specificity:
  - Early-stage priorities: domestic price liberalization, agricultural reforms, trade liberalization to exploit low-cost labor.
  - Later-stage priorities: skills development and financial deepening to support upgrading.
  - Policy choices should be flexible and tailored given heterogeneous country circumstances (natural-resource rich, small states, fragile states).

### Diversification toolkit and data resources
- Toolkit capabilities:
  - Access to highly disaggregated product-level export diversification and product quality data.
  - Computation of Theil indices (overall, extensive/between, intensive/within) and normalized quality indices relative to the 90th percentile world frontier.
  - Quality estimates for 851 SITC 4-digit-plus product categories covering more than 20 million product-exporter-importer-year observations.
  - Export diversification indices and quality measures aggregated to higher SITC levels and to sectoral groupings; sample coverage generally 1962–2010 for trade data.
- Methodology highlights:
  - Theil index definitions follow Cadot et al. (2011); intensive and extensive margins explicitly computed.
  - Quality estimated with a unit-value-based method (modified Hallak, 2006; baseline per Henn et al., 2013); estimates normalized to the 90th percentile in each product-year and aggregated using current trade values as weights.
  - Quality values typically range between 0 and 1.2 after normalization.
- Practical uses:
  - Identify country-specific bottlenecks, export product space, remaining potential for quality upgrading, and decomposition of productivity growth into reallocation and within-sector effects to guide policy.

### Heterogeneity, country examples, and small states
- Country case study highlights (Rwanda, Tanzania, Bangladesh, Angola, Vietnam, Malaysia):
  - Diverse experiences: some LICs (e.g., Vietnam, Rwanda, Tanzania) achieved notable diversification and quality upgrading; others experienced re-concentration or limited experimentation (e.g., Bangladesh, Malaysia’s later re-concentration).
  - Experimentation dynamics differ: differences in the number of new product varieties, survival rates, and growth of surviving varieties explain varied diversification outcomes.
- Small states (population below 1.5 million):
  - Small size constrains diversification; many small states face narrow export bases and vulnerability to shocks.
  - Small states have upgraded manufacturing quality considerably since the early 2000s and approach the world frontier in manufactures for many cases (16 of 25 small states in sample above 0.8 on the world possibility frontier in manufacturing).
  - Larger remaining potential for quality upgrading in agricultural and commodity sectors for small states.
  - Services data limitations (notably tourism) mean full diversification assessments for small states are incomplete.

*Source: IMF staff analysis in “Diversification and Transformation in LICs” (chapter excerpts and boxes as provided).*

### EXECUTIVE SUMMARY

### EXECUTIVE SUMMARY

### Overview
- Diversification and structural transformation play important roles in influencing the macroeconomic performance of low-income countries (LICs).
- Increases in income per capita at early stages of development are typically accompanied by a transformation in a country’s production and export structure, including diversification into new products and trading partners and increases in the quality of existing products.
- The paper uses a newly-constructed cross-country dataset, complemented by country case studies, to examine patterns of diversification and transformation in LICs since the mid-1960s.

### Key empirical findings
- Diversification in exports and in domestic production has been conducive to faster economic growth in LICs.
- Increased diversification is associated with lower output volatility and greater macroeconomic stability.
- There is both a growth payoff and a stability payoff to diversification.
- Most LICs have historically been heavily dependent on a narrow range of traditional primary products and on a small number of export markets for the bulk of their export earnings and sources of growth.
- These patterns have been changing over the past two decades, with significant variation in the extent of diversification both across LICs and within regions.
- LICs as a whole have diversified away from agriculture and into manufactured goods since the mid–1990s, but there is still ample scope to upgrade the quality of LICs’ exports including in agriculture.
- LICs typically produce at relatively low quality levels, implying considerable potential for quality upgrading; opportunities are strongest in manufacturing but also exist in agriculture.

### Analytical contributions and tools
- The paper extends existing analysis by considering diversification in external trade and the broader domestic economy, and by examining “quality upgrading” (production of higher-quality varieties of existing products).
- A new diversification toolkit developed by Fund staff provides easy access to highly disaggregated, product-level data on export diversification and product quality, enabling country authorities and mission teams to conduct more detailed, country-specific analysis.
- The data used for analysis will be provided in an accompanying toolkit.

### Drivers of diversification (cross-country evidence)
- General policy and reform measures associated with promoting diversification and structural transformation in LICs include:
  - improving infrastructure and trade networks;
  - investing in human capital;
  - encouraging financial deepening;
  - reducing barriers to entry for new products;
  - improvements in governance;
  - agricultural reforms.
- There is no one-size-fits-all recipe; country case studies document diverse experiences.

### Policy conclusions and recommendations
- Development strategies in LICs should place further emphasis on diversification, given its critical role in increasing economic growth and reducing output volatility.
- Development policies in LICs should include rather than abandon agriculture. Agriculture is the least productive sector in LICs and offers fewer opportunities to upgrade product quality than the manufacturing sector, but there is substantial scope to improve quality and introduce high value-added products.
- Authorities in LICs can help promote diversification by:
  - improving infrastructure and trade networks;
  - reducing barriers to entry for new products;
  - improving governance;
  - promoting financial deepening;
  - increasing human capital;
  - implementing agricultural reforms.
- The diversification toolkit will help identify country-specific bottlenecks and construct appropriate policy responses.

### Limitations and data issues
- Data limitations constrain analysis:
  - Analysis of export diversification and export quality is based on merchandise trade data alone, thereby missing the increasingly important role of service sector exports.
  - The quality of data on sectoral output and employment levels for LICs is uneven and typically available only at a relatively high level of aggregation, limiting the granularity of structural transformation analysis.
- These data deficiencies should not affect the robustness of the main conclusions but suggest more detailed analysis requires significant data improvements.

### Objectives and structure of the paper
- Main objectives:
  - Assess the pace and breadth of economic diversification and structural transformation in LICs—encompassing increases in sectoral productivity, export and output diversification, and quality upgrading—using a newly-constructed cross-country dataset complemented by country case studies.
  - Augment understanding of the relationships among diversification, growth, and volatility, attempting to establish more robust relationships.
  - Identify the policy drivers of diversification.
  - Inform the policy debate on how diversification and structural transformation can help LICs enhance macroeconomic stability and achieve continued and robust growth over the medium term.
- Organization highlights:
  - Cross-country patterns of diversification by income and geographic groups, covering external trade and output diversification, and patterns of quality upgrading.
  - Links between diversification and volatility and growth.
  - Analysis of main drivers of diversification, sectoral productivity, and quality upgrading.
  - Policy conclusions on how diversification can help LICs enhance macroeconomic stability and promote growth.

*Prepared by SPR under the overall guidance of Seán Nolan and Catherine Pattillo; paper led by Chris Papageorgiou (SPR) with Sarwat Jahan, Giang Ho, Ke Wang, Lisa Kolovich (SPR), Camelia Minoiu (RES), Alun Thomas (AFR), Alex Pitt (APD) and Nikola Spatafora (WB). Substantial contributions by Christian Henn (WTO), Nick Gigineishvili (AFR), Samuele Rosa (OBP), Thelma Choi, Nombulelo Duma and Jules Tapsoba (APD). March 5, 2014.*

### 11.      Export diversification can occur across either products or trading partners. Product

### 11.      Export diversification can occur across either products or trading partners. Product

### Conceptual framework and measurement
- Export diversification occurs across products or trading partners and can be decomposed into:
  - Extensive export diversification: increase in the number of export products or trading partners.
  - Intensive export diversification: considers the shares of export volumes across active products or trading partners; a country is less diversified when export revenues are driven by only a few sectors or trading partners.
- The Theil index is used to measure the extent of diversification across a country’s exports; lower values indicate higher diversification. The Theil index can be decomposed into extensive and intensive margins.
- Output diversification index reflects diversification in domestic production across seven broad sectors (value-added data).
- Quality upgrading is measured using unit values corrected for confounding factors; quality estimates are expressed relative to the world quality frontier (the 90th percentile of quality in each product-year combination).
- Structural transformation is defined as the reallocation of resources across different sectors and products over time; sectoral productivity is real value added per worker; growth in aggregate labor productivity can be decomposed into sectoral productivity growth and sectoral shifts.

### Stylized empirical findings on product diversification
- Higher per capita income is broadly associated with greater export product diversification up to advanced-economy status (with GDP per capita of $25,000-$30,000).
- The relationship between diversification and income is evident in country-year observations and in cross-sectional averages (1962-2010); the dataset’s extended time dimension is critical to confirming the relationship.
- There is substantial heterogeneity in diversification after controlling for income per capita:
  - Many small states (population less than 1.5 million) are relatively undiversified due to narrow resource bases, small domestic markets, and economies of scale in many sectors, particularly manufacturing.
  - Commodity exporters are relatively undiversified; higher incomes are not typically associated with greater diversification for commodity exporters.
- Historical patterns:
  - For over three decades (1965-1995), few LICs diversified their exports even when controlling for size and commodity-exporting status.
  - Since the mid–1990s, LICs have experienced some success in diversifying exports across products.
  - South Asian countries have made significant progress with export diversification, gaining strong momentum after 1995.
  - Experiences in SSA have been mixed: some countries (e.g., Tanzania, Uganda, and Kenya) have experienced significant export diversification in the recent period, whereas many others have not.

### Sectoral shifts accompanying diversification
- Changes in export product diversification are broadly paralleled by shifts between agricultural and manufacturing exports:
  - East and South Asia: share of agricultural exports fell from 40 percent in 1965-1970 to 15 percent in 2006-2010.
  - East and South Asia: mean share of manufactured exports rose from 17 percent in 1965-1970 to 66 percent in 2006-2010 (almost a quadrupling).
  - SSA: sizable shift away from agricultural exports but only a small rise in the share of manufacturing exports.

### Geographic diversification
- All regions, including SSA, have made significant progress in diversifying exports across partners.
- The extensive margin shows a significant increase in exports to new partners, linked to globalization and shifts in LICs’ trade away from the European Union (EU) and towards Asia (China in particular) and SSA.
- Time coverage and datasets: product-level diversification indices use an updated UN–NBER dataset harmonizing COMTRADE bilateral trade flow data at the 4-digit SITC (Rev. 1) level, extending back to 1962 and covering most LICs.

### Quality upgrading and quality ladders
- Quality upgrading—producing higher quality varieties of existing products—can increase export revenue potential by using more physical- and human-capital intensive production techniques.
- Potential for quality upgrading (length of a product’s quality ladder) varies by product:
  - Natural resources tend to have lower potential for quality upgrading than agricultural products or manufactures.
  - Manufactures typically have the highest quality upgrading potential.
- For early-stage developing countries, diversifying into products with longer quality ladders may be necessary before large gains from quality improvement are reaped; alternatively, quality upgrading within existing products may be more feasible for small LICs.
- Regional and temporal patterns in quality:
  - Since the mid-1980s, East Asia and Pacific has seen a steady increase in overall export quality levels.
  - SSA experienced a steady decrease in overall export quality, driven by a significant fall in quality upgrading in agriculture despite limited quality upgrading in manufacturing.
  - Other regions remained relatively stagnant in overall quality.
- Quality convergence:
  - Quality upgrading is particularly rapid during early stages of development, with quality convergence to the world frontier largely completed as a country reaches upper middle-income status.
  - There is wide variation in quality upgrading experiences across countries and a strong association between quality upgrading and GDP per capita growth.
  - Quality upgrading opportunities exist not only in manufacturing but also in agriculture, typically associated with a shift toward higher value products and increased productivity.
- Countries’ positions on sectoral quality ladders vary by sector; LICs generally occupy low rungs on these ladders, implying considerable potential for quality upgrading. Asian LICs have initiated diversification into low-end manufactures (apparel/misc. manufactured articles), with some emergence in East Africa.

### Data and toolkit notes
- Toolkit components:
  - Export diversification indices across products and partners (extensive and intensive margins), main source UN–NBER harmonized COMTRADE data, methodology combines importer- and exporter-reported data to maximize comprehensiveness, covering 1962–2010.
  - Quality upgrading database covering 178 countries and 851 products over 1962–2010 (based on more than 20 million product-exporter-importer-year observations); quality measures correct unit values for confounding factors and normalize to the 90th percentile world frontier in each product-year combination.
- Sample details:
  - Cross-section averages often reported for the period 1962-2010.
  - Commodity exporters sample includes 44 commodity producers (classification: either fuel or primary commodities are the main source of export earnings).
  - Small states defined as population less than 1.5 million.

*Source: IMF staff analysis in “Diversification and Transformation in LICs” (chapter excerpts and boxes as provided).*

### 21.      Measures of diversification should go beyond trade, to capture domestic sector

### 21.      Measures of diversification should go beyond trade, to capture domestic sector diversification and the underlying dynamic process of structural transformation

### Overview and motivation
- Diversification analysis should include sectoral value-added shares and sectoral allocation of labor in addition to export measures.
- Historical growth takeoffs in LICs have been accompanied by productivity surges in broad sectors—agriculture, industry, and services.
- As economies diversified production, export diversification (type and quality of export products) also increased (Papageorgiou and Spatafora, 2012).
- The paper examines sectoral diversification and structural transformation using sectoral value-added and labor allocation data (see Box 1 for data sources).

### Output and sectoral diversification: patterns and trends
- Output has become more diversified in all regions over the past 40 years (Figure 8).
- Both SSA and South Asia experienced output diversification; however, there has been a significant widening of the gap between these two regions since the early 1980s.
- Over the past decade, the share of agriculture in output in LICs has declined significantly.
- Analysis of seven key sectors shows the gap left by agriculture was largely filled by non-tradable activities: construction, wholesale trade, and transportation, rather than by manufacturing (Figure 9).
- Significant cross-country variation exists in:
  - the magnitude of resources shifting out of agriculture, and
  - the identity of sectors expanding in agriculture’s place.
- Geography matters: resource-rich, coastal non-resource-rich, and land-locked non-resource-rich countries in SSA may need different growth models (Ndulu et al., 2007).

### Export product diversification, quality upgrading, and sectoral reallocation
- Export product diversification and quality upgrading are associated with:
  - larger reductions in agricultural shares over 1990-2010,
  - faster movement out of agriculture for economies with faster quality upgrading,
  - more productivity-enhancing sectoral reallocation (workers moving from lower-productivity to higher-productivity activities and sectors).
- Over 1990-2010, LICs with a more diversified export base experienced larger reductions in agricultural shares.
- Sectoral reallocation tended to be more productivity-enhancing in economies with greater export diversification and quality upgrading (Figures 10 top, middle, bottom panels).
- The link between export diversification/quality convergence and domestic sectoral shifts suggests significant scope for positive spillovers even when not all products are tradable.

### Case studies: varied country experiences and mechanisms
- Six country case studies (Rwanda, Tanzania, Bangladesh, Angola, Vietnam, Malaysia) illustrate heterogeneous diversification experiences across LICs and MICs in SSA and East Asia.
- Case study selection spans:
  - LICs with per capita income below $1,000 (Rwanda, Tanzania, Bangladesh),
  - a resource-dominated MIC facing development challenges (Angola),
  - countries at or beyond MIC thresholds with successful diversification experiences (Vietnam, Malaysia).
- Key cross-case findings:
  - Export Product Diversification:
    - Angola: steady increase in export concentration driven by oil.
    - Bangladesh: initial diversification with garments replacing agricultural exports, later re-concentration.
    - Tanzania: major changes since the 1990s; traditional agricultural cash crops declined.
    - Rwanda: diversified exports starting from a low base.
    - Malaysia: historically most diversified, later some re-concentration as electronic products dominated in the 1990s and primary commodities regained prominence.
    - In all countries, agricultural share in total exports declined; manufacturing shares increased in all but Malaysia.
  - Output Diversification:
    - Increased in five of six sample countries; exception Vietnam where overall diversification declined in the 2000s despite rapid structural change (big rise in manufacturing share and sharp decline in agriculture).
    - Countries with the fastest GDP per capita growth also experienced the fastest growth in manufacturing output.
  - Quality Upgrading:
    - Malaysia: highest average quality level.
    - Vietnam and Rwanda: most improvement in export quality since the 1990s.
    - Tanzania: average quality relative to frontier decreased despite increased export product diversification (Figure 12).

### Export experimentation and extensive margin dynamics
- Export product diversification depends on:
  - frequency of new product introductions (extensive margin),
  - survival rates of new varieties,
  - growth prospects of surviving varieties.
- Over 1990–2010, countries showed significant differences in:
  - number of new product varieties introduced annually,
  - survival rates of new varieties,
  - growth rates of surviving varieties.
- Country-level observations:
  - Malaysia: lowest growth in product varieties over time; greatest share of exports from incumbent varieties; low experimentation possibly linked to re-concentration.
  - Vietnam: significant new entry, high survival probability for new varieties.
  - Tanzania: less growth in number of new varieties, but surviving varieties grew more—achieved reduction in incumbent share.
  - Rwanda: performed well in the early 2000s on new entry.
  - Bangladesh: less experimentation and less growth in surviving varieties, contributing to unusually high concentration (Figure 13).
- Note: A variety is defined as a specific product exported to a specific country (Asmundson forthcoming). Angola omitted from experimentation analysis due to insufficient data.

### Diversification, growth, and stability: empirical associations
- The paper establishes a robust relationship between growth and diversification in LICs using recently developed econometric methods.
- Quality upgrading and sectoral reallocation of resources are positively linked to growth.
- Export diversification (across products and trading partners) is associated with lower output volatility.
- After a long period of poor performance, many LICs have grown at very high rates over the past 15 years; some of the world’s fastest growing countries are LICs, and SSA is the world’s second fastest-growing region, behind only developing Asia.
- Contributors to recent LIC growth include favorable commodity prices in the 2000s, improved macroeconomic stability, better policy making, healthier institutions, and structural reforms.
- Heterogeneity:
  - Not all fragile countries experienced the post–1995 growth spurt.
  - Many LICs diversified rapidly; progress was more halting in others, especially fragile states.
- Export product diversification patterns are on average positively associated with subsequent growth, though the relationship is heterogeneous (Figure 14).
- Diversification spurts (sharp increases in diversification) are associated with subsequent sharp increases in growth; growth accelerations are associated with subsequent increases in export product diversification (Figure 15).
- Results are more pronounced in non-fragile LICs.
- To address causality concerns, the empirical approach leverages a method designed to allow for a large set of potential growth determinants when causality is in question (Eicher et al., 2009) (See Supplement Note 1).

*Source: _030514 - 21.      Measures of diversification should go beyond trade, to capture domestic sector*

### 33.      The results suggest that diversification in both exports and output is among the key

### _030514 - 33.      The results suggest that diversification in both exports and output is among the key

### Diversification and growth — core findings
- Diversification in both exports and output is identified as a key factor driving growth, but only in LICs.
- Posterior inclusion probabilities indicate that export diversification is among the most effective determinants of growth in LICs.
- Results are similar when diversification is measured at the extensive or intensive margin.
- Economic significance:
  - A one standard deviation increase in LICs’ export product diversification raises their growth rate by about 0.8 percentage points.
  - A one standard deviation increase in output diversification in LICs raises their average annual growth rate by about 1.4 percentage points.
- These findings are robust to endogeneity and model uncertainty using an IVBMA estimator.

### Quality upgrading and growth
- Quality upgrading is associated with higher growth; the correlation is strongest for manufacturing quality upgrading.
- Agriculture also offers ample quality upgrading opportunities; upgrading often involves rebalancing toward higher value products and typically leads to productivity increases that free up resources.
- Sectoral patterns:
  - Manufacturing's share of GDP tends to expand as quality in the sector increases.
- Empirical magnitudes:
  - Average annual growth was more than 1 percentage point higher in LICs with fast quality convergence compared to LICs with slow quality convergence (post-1995).
  - When focusing on Africa only, the growth dividend from quality upgrading is still in excess of ½ percentage points per annum in the post-1995 period.
  - Small states with fast quality convergence experienced additional annual per capita growth of 2 percentage points relative to slow converging peers.

### Sectoral reallocation, sectoral productivity, and heterogeneity
- Growth in economy-wide labor productivity decomposes into within-sector productivity growth and sectoral shifts.
- Regional decomposition highlights heterogeneity:
  - Agriculture and manufacturing were primary sources of productivity growth in Asia.
  - Services were relatively more dominant in Europe and Central Asia (ECA), Middle East and North Africa (MENA), and Sub-Saharan Africa (SSA).
  - Mining played an important role in Latin America and the Caribbean (LAC).
- Labor reallocation:
  - Labor reallocation contributed largest productivity gains in Asian countries.
  - Labor reallocation contributed negatively to productivity growth in LAC and particularly SSA.
- Agricultural productivity gaps:
  - Agricultural productivity (average real value added per worker in agriculture as a percentage of that in industry and services) averaged about 30 percent in 2005 for LICs, with large dispersion across countries.

### Diversification and volatility
- Both export and output diversifications are associated with lower volatility of output in developing countries.
- The effect on output volatility is primarily driven by a more evenly balanced product mix of the current export basket (intensive margin).
- The diversification–volatility relationship holds even for more trade-open countries (exports and imports to GDP).
- Related effects:
  - Diversification from agriculture into manufacturing and services is associated with lower volatility of inflation because large and volatile shocks to food supply play a relatively smaller role.
  - Diversification across partners also helps reduce volatility of GDP per capita.
- Large diversification spurts:
  - Staff analysis identifies a total of 61 export product diversification spurts in the post–1962 period involving 51 developing countries.
  - Diversification spurts last 13 years on average, rising to 20 years in the East Asia and Pacific region.
  - Export product diversification spurts are associated with reduction in the volatility of GDP per capita.
    - In LICs, output volatility diminishes by 0.5 percentage points in the wake of diversification spurts.
    - In the rest of the sample the effect is about half of that in LICs.

### Drivers of diversification, quality upgrading, and sectoral productivity (associations)
- Cross-country regression analysis indicates common economic fundamentals and policy variables associated with diversification dimensions; relationships should be interpreted as associations rather than causal effects.
- Key associated drivers:
  - Human capital and institutional quality are important for both export product diversification and quality upgrading.
  - Agricultural reforms and financial liberalization have contributed to export product diversification, quality upgrading, and sectoral productivity.
- Case study evidence:
  - No universal diversification trajectory exists.
  - Successful diversification is linked to removal of barriers to entrepreneurship and investment in infrastructure and people.

*Source: IMF staff calculations and analysis from the cited chapter.*

### 42.      Regression analysis reveals that export product diversification is associated with a

### _030514 - 42.      Regression analysis reveals that export product diversification is associated with a

### Regression findings on export product diversification
- Empirical specification accounts for a hump-shaped pattern of export product diversification along the development path and includes GDP per capita and its quadratic term to capture the turning point (TP).
- Countries below a certain level of income per capita (US$ 25,000 in 2005 purchasing power parity dollars) are in the diversification stage while countries above it reconcentrate their export structure.
- Associations uncovered (using a specification that accounted for the turning point):
  - Higher levels of education and institutional quality, including better protection of property rights, are associated with higher export product diversification; the institutional quality effect is economically significant only for developing countries (those below the TP).
  - Deeper financial systems (measured by the private credit to GDP ratio) are associated with higher export diversification in developing countries (those below the TP).
  - Globalization (measured by the overall KOF index) is associated with higher export diversification for all countries.
  - Increasing trends in South-South trade and higher export shares to all developing economies are correlated with increased export diversification.
  - Proximity to markets is important for diversification; landlocked countries are found to be less diversified.
  - Several reforms correlate with diversification, notably trade liberalization and agricultural reforms; agricultural reforms have the highest impact on countries at early stages of development.
  - Some statistical evidence indicates real exchange rate (RER) undervaluation (measured relative to a PPP-based measure of the equilibrium RER that allows for Balassa-Samuelson effects) is associated with higher export diversification, but persistent use of a depreciated exchange rate can lead to consistently higher inflation or sustained reserve accumulation—neither supportive of growth.

### Drivers of sectoral shifts and sectoral productivity
- Sectoral reallocation responds to the infrastructure and regulatory environment affecting labor mobility (analysis of LICs over 1995-2010).
- Policies associated with higher manufacturing employment shares:
  - Liberalizing the electricity market.
  - Easing credit.
  - Reducing labor market and business regulations.
- Improving education is linked to increased value added shares for both manufacturing and service sectors.
- Reducing average import tariff rates is associated with labor moving out of agriculture in the analyzed LIC group.
- Reforms associated with sectoral productivity gains (sector-specific):
  - Agriculture: Removing tariff barriers and reducing interest rate controls boost agricultural productivity in LICs (channels: better market and technology access, cheaper imported inputs, greater competition).
  - Manufacturing: Domestic financial sector reforms, capital account liberalization (FDI), improvement in road infrastructure, and education matter for manufacturing productivity.
  - Services: Liberalization of network industries, specifically telecommunication, generates productivity gains via greater competition, innovation, more FDI, and better access at lower prices.
- Data note: Employment data in LICs are of low quality and uneven coverage; analysis combines household surveys, GGDC database, WDI, ILO, and supplements employment measures with value added from UN National Accounts.

### Drivers of quality upgrading
- Human capital and institutional quality are positively associated with the growth rate of product quality.
  - Regression estimates indicate movement up the value chain is associated with an increasingly skilled workforce.
  - Institutional quality encourages investments needed for process and product upgrading.
- Liberal trade regimes, agricultural policy, and domestic financial markets support faster quality upgrading; these relationships are statistically and economically significant.
- LICs’ potential for quality upgrading is not apparently constrained by destination-market demand:
  - The average quality level imported in LIC destination markets is much higher than the average quality of LIC exports, implying demand in existing markets is not the binding constraint.
  - Poorer exporters show greater remaining upgrading potential (larger gap between quality demanded and quality exported).
  - Fragile LICs and LICs in Africa show larger gaps than non-fragile peers and higher-income LICs; small states show more limited upgrading potential in existing destination markets.
  - Reorientation toward higher-income markets is not pressing at this stage to harness quality upgrading potential.

### Country case studies — patterns and reform lessons
- Common factors associated with successful diversification (no causation claimed):
  - Macroeconomic stabilization coincided with successful diversification in cases like Vietnam, Rwanda, and Tanzania.
  - Reduction of barriers to entry boosts diversification by lowering costs and encouraging entrepreneurship (e.g., reversal of collectivization in Vietnam; large divestment of state enterprises in Rwanda; dismantling of state distribution in Tanzania; removal of bureaucratic red tape in Bangladesh).
  - Investment in infrastructure lowers business costs and is an important state role (producer and/or regulator of infrastructure).
  - Investment in people is crucial for moving up the quality ladder (examples: Vietnam increased average years of education by about 50 percent in two decades; Rwanda expanded education through ninth grade for all students).
- Effective reform waves adapt to evolving comparative advantage:
  - Early-stage priorities: domestic price liberalization, agricultural reforms, and trade liberalization to exploit low-cost labor.
  - Later-stage priorities: development of domestic skills and deepening the financial system to match changing comparative advantage.
  - Examples: Vietnam’s transformation from low-end agricultural exporter to middle-range manufacturing exporter in less than two decades; Rwanda’s agricultural reforms boosting productivity and reducing poverty.
- Industry-targeted measures have mixed outcomes:
  - State support to specific industries (Malaysia, Vietnam, Bangladesh) sometimes increased exports but also led to greater export concentration.
  - SOEs in Vietnam largely failed to lead strategic industries.
  - Targeted subsidies can be risky where governance is weak and policy capture is possible.
  - Industrial policies facilitating agglomeration (industrial zones, export processing zones) have been effective in encouraging entrepreneurial activity via economies of scale, skills and knowledge spillovers, and backward/forward linkages.

### Policy views and recommendations
- Structural transformation and diversification should be central to LIC development strategy because they enhance growth and reduce volatility.
  - Sustainable development involves reallocating resources towards more productive sectors, diversifying into new products, and upgrading the quality of existing products.
  - Diversification and quality upgrading are often complementary; entry into products with long quality ladders offers particular scope for quality improvement.
  - Significant heterogeneity exists across regions and countries; natural resource rich countries, small states, and fragile states warrant special consideration.
- Agriculture should not be neglected:
  - A large share of LIC populations remain employed in agriculture; moving the poorest into growth gains requires agricultural development.
  - There is substantial scope for productivity and quality gains within agriculture (high value-added products, market-oriented production, quality upgrading).
- Role of government policies depends on presence and nature of market failures:
  - Common market failures in LICs: infrastructure provision, human capital accumulation, trade networks, functioning factor markets, and creation/management of ideas.
  - Horizontal policies with evidence of success: improving education, infrastructure, institutional and regulatory environment; enhancing access to new export markets; structural reforms such as trade liberalization, reduction of agricultural subsidies and price controls, and financial deepening through banking-sector reforms.
  - Industry-focused measures remain an open question; broad policies aimed at resolving specific market failures (e.g., facilitating agglomeration, providing information for nascent export-oriented industries) can promote entry into new activities.
  - Policy approach should be flexible and country-specific given heterogeneity in experiences and constraints.

*Source: IMF staff analysis in the chapter “Diversification and Transformation in LICs.”*

### 53.      To enable more detailed, country-specific analysis, Fund staff has developed a new

### _030514 - 53.      To enable more detailed, country-specific analysis, Fund staff has developed a new

### Diversification toolkit — purpose and capabilities
- Provides easy access to highly disaggregated, product-level data on export diversification and product quality.
- Enables country authorities, mission teams, and researchers to assess the potential for further transformation and quality upgrading.
- Produces quality estimates and diversification indices that support country-specific analysis and cross-country comparisons.

### Measurement of diversification and quality (methodology)
- Theil indices (overall, intensive/within, extensive/between) follow definitions and methods in Cadot et al. (2011).
  - Products are classified by dummies as “Traditional,” “New,” or “Non-traded” (classification rules specified in text).
  - Extensive Theil index formula: T_B = ∑_k (N_k/N) (μ_k/μ) ln(μ_k/μ).
  - Intensive Theil index formula: T_W = ∑_k (N_k/N) (μ_k/μ) {(1/N_k) ∑_{i∈I_k} (x_i/μ_k) ln(x_i/μ_k)}.
- Quality upgrading is estimated using a unit-value based method (modified Hallak, 2006; baseline methodology per Henn et al., (2013)):
  - Trade price (unit value) model: ln p_mxt = ζ0 + ζ1 ln θ_mxt + ζ2 ln y_xt + ζ3 ln Dist_mx + ξ_mxt (equation (1) in source).
  - A quality-augmented gravity equation is specified separately for each product (equation (2)).
  - Substitution and rearrangement yield an estimable form (equation (3)); estimation is carried out separately for each product.
  - Estimates obtained by two stage least squares for each product.
- Exact aggregation and normalization procedures:
  - Estimation yields quality estimates for more than 20 million product-exporter-importer-year combinations.
  - All quality estimates are first normalized by their 90th percentile in the relevant product-year combination.
  - Resulting quality values typically range between 0 and 1.2.
  - Quality estimates aggregated to SITC 4-, 3-, 2-, and 1-digit levels and to country-level totals using current trade values as weights; normalization to the 90th percentile is repeated at each aggregation step.
  - Aggregations also produced based on the BEC classification and on 3 broad sectors (agriculture, non-agricultural commodities, and manufactures).
  - Unit values are also normalized with the 90th percentile set equal to unity.

### Empirical scope and data coverage
- Quality estimated for 851 SITC 4-digit-plus product categories (estimation done separately for each of the 851 categories).
- Quality estimates cover more than 20 million product-exporter-importer-year combinations.
- Sectoral productivity and sectoral shifts:
  - Aggregate labor productivity growth decomposed into sectoral reallocation effect and within-sector productivity growth (notation and decomposition described in source).
  - Data sources and periods:
    - Value added shares in 7 sectors: 1970–2010 (UN (2013)).
    - Labor employment shares in 7 sectors: 1969–2009 (ILO (2011), 1-digit classification).

### Findings on commodity exporters (Annex Box 1)
- Commodity exporters defined as countries with more than 50 percent of total exports consisting of fuels and primary commodities.
- Key observations:
  - Commodity exporters are undiversified in exports, especially fuel-exporting countries (export diversification indices confirm this).
  - Output diversification (measured by output or employment) is a more nuanced picture: differences in output diversification between commodity and non-commodity exporters are significantly less pronounced than differences in export diversification.
  - Within commodity exporters, greater income is associated with greater output diversification.
  - Commodity exporters’ position on the sectoral quality ladders is relatively low across almost all sectors, particularly in manufactures.
  - Commodity exporters have not seen the same reduction in agricultural employment shares as non-commodity exporters, which may reflect fewer alternative productive opportunities.

### Practical implications for country analysis and policy
- The toolkit facilitates detailed country-specific diagnostics on:
  - Export product space and diversification dynamics (new vs. traditional products).
  - Quality upgrading across products and sectors using normalized quality indices.
  - Decomposition of productivity growth into reallocation and within-sector components to inform structural policies.
- Policy-relevant metrics produced by the toolkit support assessment of:
  - The potential for quality upgrading and product transformation.
  - The degree and nature of export concentration and its implications for volatility and inclusive growth (e.g., commodity dependence and employment outcomes).

*Source: _030514 - 53. To enable more detailed, country-specific analysis, Fund staff has developed a new (IMF PDF content provided).*

### Annex Box 2. Small States

### Annex Box 2. Small States

### Overview
- Definition: small states = population below 1.5 million.
- Small size constrains capacity to diversify; narrow and undiversified production and export bases increase vulnerability to external shocks.
- July 2012 IMF survey of mission chiefs in small states reports that diversification is particularly difficult to achieve in these economies.

### Quality upgrading patterns and development
- Quality rises with income, particularly for income levels below $10,000 (Henn et al., 2013).
- Small states have:
  - Upgraded their quality considerably in the manufacturing sector since the early 2000s, from already high levels.
  - Approached the world frontier quality in manufactures.
- In agriculture, export quality in small states remains low, although some quality upgrading has taken place since the mid-1990s.

### Sectoral positions and diversification potential
- Small states’ position on sectoral quality ladders is relatively high in:
  - manufactured goods,
  - chemicals,
  - machinery and transport equipment.
- Implication: in these high-quality sectors, horizontal diversification into other subsectors may need to precede further quality upgrading.
- Current larger potential for quality upgrading for small states as a group:
  - agricultural sectors,
  - commodity sectors.

### Empirical coverage and data caveats
- Data availability: a majority of the small states for which data were available for analysis (16 of 25) were above 0.8 on the world possibility frontier in the manufacturing sector.
- Methodological caveat: the data and methodology do not allow for a breakdown of commodities by position within a value-chain; countries can be recorded as exporting high-value final products while undertaking only the final assembly or finishing of these goods.
- Services data limitation: comprehensive data on services is very limited and often unreliable; tourism—an important sector in many small states—could not be included in the analytical work in this box. As data quality improves, it is necessary to assess overall diversification in small states by including tourism. 

*Source: Comtrade, IMF staff estimates.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/np/pp/eng/2014/_030514.pdf_