## wp1886

## Source details

**Canonical URL:** [wp1886](https://www.imf.org/-/media/files/publications/wp/2018/wp1886.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2018/wp1886.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2018/wp1886.pdf.json)

---

### Executive summary and research focus
- Introduction of new products to the export and production basket typically embodies productivity improvement (“dynamic effect”), leading to higher long run growth.
- Export diversification is pursued as a deliberate growth strategy, particularly important in countries with limited indigenous sources of productivity growth.
- Small states face inherent constraints to diversification: scarce resources, inadequate economic infrastructure, acute skills shortage, and high transportation costs due to geographical isolation from main trading partners.
- Diversification requires channeling resources away from sectors with comparative advantage, so assessment of costs and benefits is warranted.
- Research questions:
  - Do more diversified small states tend to have higher GDP per capita growth and lower volatility?
  - Is export concentration less beneficial for growth in small states and does it matter whether countries are commodity intensive or tourism intensive?
  - Are activist development policies to promote diversification justifiable in small states?
- Sample: 33 small states (population ≤ 1.5 million; excluding advanced market economies and high-income oil exporters).

### Sample definition and paper structure
- Small states defined as sovereign countries with a population of 1.5 million people or fewer.
- Paper structure:
  - Section II: theoretical and empirical literature overview.
  - Section III: stylized facts comparing small states to larger peers.
  - Section IV: empirical analysis summary.
  - Section V: conclusions and policy implications.

### Theoretical and empirical background
- Theoretical rationale:
  - Structural models: diversification from primary exports into manufactured exports supports sustainable growth (Chenery, 1979; Syrquin, 1989).
  - Commodity export dependence creates export earnings instability; diversification can stabilize export earnings (Ghosh and Ostry, 1994; Bleaney and Greenaway, 2001; McMillan, Rodrik and Verduzco-Gallo, 2014).
  - Hausmann et al (2003): growth driven by diversification of investments into new activities and entrepreneurial cost-discovery; governments can promote entrepreneurship and incentives for new activities.
- Empirical findings (selected):
  - Love (1986); Al-Marhubi (2000); Agosin (2007); Lederman and Maloney (2007): evidence of export diversification contributing to higher growth.
  - Feenstra and Kee (2004): a 10 percent boost in export diversification in all industries would result in a 1.3 percentage point increase in a country’s productivity growth (sample of 34 countries, 1984-1997).
  - Michaely (1977): positive link between diversification and growth only among more-developed countries; not found for least-developed economies.
  - Haddad et al. (2012): trade openness lowers output volatility in sufficiently diversified economies, while it increases volatility in those with more concentrated export baskets.
- Contribution: fills gaps on diversification and output volatility for small states.

### Stylized facts — patterns in small states
- Income status:
  - All but one (Comoros) small states in the study have reached at least middle-income status.
  - Roughly half are categorized as upper-middle or high-income economies.
- Modes of diversification:
  - In small states, high production costs and lack of economies of scale make extensive-margin diversification (introduction of many new products) difficult.
  - Diversification in small states mostly takes place via the intensive margin: rebalancing mix of existing export products or trading partners.
- Trends and reversals:
  - General trend toward diversification between 1995 and 2005 across most small states.
  - Diversification reversed during and after the 2007-09 global financial crisis.
  - Drivers of reversal: countercyclical fiscal responses focused on leading sectors, retreat of FDI, terms-of-trade movements, loss of preferential trade agreements.
- Group-specific patterns:
  - Tourism-dependent economies: track growth in Advanced Economies, highly synchronized with global business cycle, most impacted by global recessions.
  - Primary commodity exporters: most exposed to international commodity price fluctuations and experienced the most drastic change toward specialization since the mid-90s.
  - Diversified economies: greatest resilience — least volatility and higher average growth rates among groups.
- Service sector: plays a more important role in diversifying exports in small states than in larger peers.
- Economic Complexity Index (ECI): generally lower in small states compared to Emerging Markets (EM) average and higher than Low-Income Countries (LIC) average; country examples: Mauritius (improvements), Jamaica (relatively static), Trinidad and Tobago (declining).

### Measurement of diversification and classification
- Diversification metric: Herfindahl index (HFI) = ∑ s_i^2; i = 1,2,...,n. Index varies between 0 and 1; larger values indicate more concentrated export structure; lower values indicate more diversified export structure.
- Indices constructed:
  - HFI1: goods only (SITC1), 1990-2014.
  - HFI4: goods only (SITC4), 1962-2014.
  - HFI1 with travel: goods + travel, 1990-2014.
  - HFIm: export markets (country level), 1988-2014.
  - HFI1 with service: goods + service (BPM6 1 digit), 1990-2014.
- Classification: 33 small states ranked by dominant industry (>50% export share) for periods 1990-2004 and 2005-15; eight groupings defined, with focus in the paper on:
  - traditionally tourism;
  - traditionally primary;
  - traditionally diversified;
  - transformed from diversified to tourism.
- “Diversified” defined as no dominant industry exists.

### Empirical framework (regression design)
- Sample and aggregation:
  - Regression data covers 1990-2014.
  - Averages taken for every 3 years to estimate impacts on long-term growth.
- Growth estimation (Solow specification):
  - ∆y_{i,t} = α + β Diversification_{i,t−1} + φ X_{i,t} + γ_t + ε_i + ε_{i,t}
  - ∆y_{i,t}: real GDP growth rate.
  - Diversification measures (lagged): export goods diversification (SITC1, SITC4), goods and services, export markets, real sector diversification.
  - Controls (X): population growth, shares of gross capital formation, trade openness, FDI flows. Education omitted due to data gaps; natural disaster dummy tested but not retained.
- Volatility estimation:
  - VOL_{i,t} = σ VOL_{i,t−1} + ρ Diversification_{i,t−1} + δ X_{i,t} + γ_t + ε_i + ε_{i,t}
  - VOL_{i,t}: standard deviation of real GDP growth over each non-overlapping 3-year period.
  - Controls: trade openness, inflation volatility, terms of trade volatility, exchange rate volatility.

### Main regression findings — Growth (selected coefficients and statistics)
- General result: Diversification (lower HFI) associated with higher growth and lower volatility in small states.
- Key control coefficients (examples across specifications):
  - Gross capital formation: 0.121*** (0.0402); 0.129*** (0.0410); 0.127*** (0.0452); 0.126*** (0.0457); 0.151*** (0.0442); 0.134*** (0.0494); 0.116*** (0.0429)
  - Trade openness: 4.000*** (1.400); 2.343 (2.161); 3.010 (2.304); 2.996 (2.302); 2.027 (2.225); 3.119 (2.827); 3.317 (2.124)
  - FDI flows: -0.305 (0.309); -0.585* (0.342); -0.496 (0.374); -0.492 (0.374); -0.524 (0.354); -0.516 (0.518); -0.618* (0.349)
- Diversification coefficients (selected):
  - HFI, 1 digit export goods: -1.596* (0.853) — negative and significant.
  - HFI, 4 digit export goods: -1.059** (0.495) — negative and significant.
  - HFI, 1 digit export goods and travel: -0.285 (1.113) — negative but insignificant.
  - HFI, 1 digit export goods and services: -0.230 (1.148) — negative but insignificant.
  - HFI, export markets: 0.120 (0.741) — insignificant.
  - HFI, real sector: -1.198 (2.865) — negative but insignificant.
- Model statistics and sample sizes:
  - Observations by specification: 170; 154; 147; 147; 148; 125; 154
  - Number of ifs code: 22 across specifications
  - R-squared: 0.283; 0.302; 0.278; 0.278; 0.322; 0.292; 0.283
  - Significance notation: *** p<0.01, ** p<0.05, * p<0.1
- Interpretation: Export diversification across goods at 1-digit and 4-digit levels helps economic growth; other diversification measures are negative but often statistically insignificant.

### Main regression findings — Volatility (selected coefficients and statistics)
- Key control coefficients (examples):
  - Lagged volatility: -0.206 (0.138); -0.180 (0.124); -0.155 (0.128); -0.171 (0.129); -0.0990 (0.135); -0.327 (0.202); -0.163 (0.144)
  - Trade openness: -4.213** (1.859); -1.837 (1.820); -3.052* (1.746); -3.135* (1.758); -3.036* (1.775); -5.186** (2.202); -3.948** (1.876)
  - Inflation volatility: -0.173* (0.0877); -0.135* (0.0794); -0.160 (0.119); -0.172 (0.120); -0.199 (0.119); -0.214* (0.118); -0.163* (0.0883)
  - Terms of trade volatility: coefficients reported as (1.36e-10), (1.22e-10), (1.27e-10), (1.28e-10), (1.26e-10), (1.45e-10), (1.37e-10)
  - Exchange rate volatility: 7.608 (6.563); 5.851 (5.900); 1.671 (6.662); 2.775 (6.593); 6.591 (6.180); 12.069 (11.54); 9.519 (6.822)
- Diversification coefficients (selected):
  - HFI, 1 digit export goods: 2.046*** (0.634) — positive and significant.
  - HFI, 1 digit export goods and travel: 1.680* (0.852) — positive and significant.
  - HFI, 1 digit export goods and services: 1.368* (0.759) — positive and significant.
  - HFI, 4 digit export goods: 0.595 (0.370) — positive but insignificant.
  - HFI, export markets: 1.211* (0.616) — positive and significant.
  - HFI, real sector: -2.786 (2.733) — negative but insignificant.
- Model statistics and sample sizes:
  - Observations by specification: 65; 65; 61; 61; 65; 61; 65
  - Number of ifs code: 16 across specifications
  - R-squared: 0.410; 0.540; 0.534; 0.526; 0.535; 0.527; 0.426
- Interpretation: Export goods and services diversification reduce growth volatility; export market diversification also reduces volatility by mitigating partner-specific shocks.

### Synthesis of empirical findings
- Cross-country evidence for small states (1990-2015):
  - More diversified small states experienced the lowest output volatility and a higher average growth rate than most other small states.
  - Export diversification has a more significant impact on reducing output volatility than on improving long run growth rates in small states.
- Caveats:
  - Diversification successes concentrated in a small number of small states.
  - Diversification trend in the late 1990s and early 2000s reversed during and after the 2007-09 global financial crisis.
  - Fundamental determinants (policy and institutional factors) mediate the relationship between diversification and growth.

### Box 3 — Diversification in Mauritius (case study)
- Structural transformation:
  - 1970s: mainly agriculture (e.g. sugar) → 1990s: manufacturing (e.g. garment and jewelry) → since early 2000s: broad range of service exports (tourism, finance, information technology, business process outsourcing).
- Macroeconomic outcomes:
  - Mauritius averted recession during the 2007-09 global financial crisis and maintained relatively stable output growth.
  - Growth volatility in 2006-15 more than halved compared to the previous decade.
  - The average growth rate achieved in the past decade is "around 4 percent".
- Competitiveness and business environment:
  - Ranks 49/190 in 2017 World Bank Doing Business Index (highest among all small states).
  - Ranks 45/138 in World Economic Forum Global Competitiveness Index (highest among all states).
  - Competitive labor costs, partly attributed to lower minimum wage requirements, supported manufacturing expansion.

### Policy implications and recommendations
- Diversification should be part of a cohesive development strategy emphasizing fundamental determinants:
  - Maintain macroeconomic policy stability.
  - Foster a supportive business environment: quality infrastructure, human capital, and essential business services.
  - Attract investment to create new comparative advantages and foster development of new sectors.
- Prefer cost-effective policies given limited fiscal space in most small states:
  - Emphasize policy stability and business environment improvements over activist development policies and compensative tax incentives.
- Risk-pooling and economic integration:
  - Strengthen risk-pooling arrangements (e.g., economic integration as a substitute for diversification) given prevalence of country-specific shocks (e.g., natural disasters) among small states.
- Trade openness considerations:
  - Trade openness can lower output volatility in sufficiently diversified economies but can increase volatility in concentrated exporters; policy should account for a country’s diversification status when liberalizing trade.

*Source: Excerpt from IMF working paper (wp1886) content provided*

### introduction of new products to the export and production basket, which usually embodies

### introduction of new products to the export and production basket, which usually embodies

### Executive summary and research focus
- Introduction of new products to the export and production basket typically embodies productivity improvement (“dynamic effect”), leading to higher long run growth.
- Export diversification is pursued as a deliberate growth strategy, particularly important in countries with limited indigenous sources of productivity growth.
- Small states face inherent constraints to diversification: scarce resources, inadequate economic infrastructure, acute skills shortage, and high transportation costs due to geographical isolation from main trading partners.
- Diversification requires channeling resources away from sectors with comparative advantage, so a clear assessment of costs and benefits is warranted.
- This paper considers concepts of economic diversification with respect to exports (including tourism) for a sample of 33 small states and empirically investigates the relationship between diversification and economic growth and volatility.
- Research questions include:
  - Do more diversified small states tend to have higher GDP per capita growth and lower volatility?
  - Is export concentration less beneficial for growth in small states and does it matter whether countries are commodity intensive or tourism intensive?
  - Are activist development policies to promote diversification justifiable in small states?

### Sample definition
- Small states are sovereign countries with a population of 1.5 million people or fewer.
- The sample is narrowed to exclude those defined as advanced market economies or high-income oil exporting countries.
- Sample size: 33 small states (list provided in Annex I of source).

### Paper structure
- Section II: theoretical and empirical literature overview.
- Section III: stylized facts comparing small states to larger peers.
- Section IV: empirical analysis summary.
- Section V: conclusions and policy implications.

### *Source: IMF working paper content provided*

---

### II. LITERATURE REVIEW — theoretical and empirical background
- Theoretical rationale:
  - Structural models: diversification from primary exports into manufactured exports supports sustainable growth (Chenery, H. 1979; Syrquin, 1989).
  - Commodity export dependence creates export earnings instability due to volatile market prices; diversification can stabilize export earnings (Ghosh and Ostry, 1994; Bleaney and Greenaway, 2001; McMillan, Rodrik and Verduzco-Gallo, 2014).
  - Hausmann et al (2003): growth driven by diversification of investments into new activities and entrepreneurial cost-discovery; governments can promote entrepreneurship and create incentives for investment in new activities.
- Empirical findings supporting diversification-led growth:
  - Love (1986): evidence of export diversification contributing to higher per capita income growth.
  - Al-Marhubi (2000): export diversification promotes robust economic growth across specifications.
  - Agosin (2007): stronger diversification effect when exports grow faster.
  - Lerderman and Maloney (2007): dynamic panel supports diversification-led growth.
  - Feestra and Lee (2004): a 10 percent boost in export diversification in all industries would result in a 1.3 percentage point increase in a country’s productivity growth (sample of 34 countries, 1984-1997).
  - Gutierrez de Pineres and Ferrantino (2000): positive interplay between export diversification and growth for Chile, Colombia, El Salvador, Paraguay, and Uruguay.
- Evidence of non-universality or conditional effects:
  - Michaely (1977): positive link between diversification and growth only among more-developed countries; not found for least-developed economies.
  - Di Salvo (2015): explored non-linearity between export diversification and economic growth using 1995-2010 dynamic panel.
- Small states and volatility:
  - Export diversification relevant for countries vulnerable to terms-of-trade (ToT) shocks.
  - Limited empirical work on small states: Sannassee, Seetanah and Lamport (2014) found positive relationship in Mauritius.
  - Trade openness interacts with diversification: Mona Haddad et al. (2012) indicate trade openness lowers output volatility in sufficiently diversified economies, while it increases volatility in those with more concentrated export baskets.
- Paper contribution: fills gaps on diversification and output volatility for small states.

---

### III. STYLIZED FACTS — patterns in small states
- Income status and development:
  - All but one (Comoros) small states in the study have reached at least middle-income status after notable economic development.
  - Roughly half are categorized as upper-middle or high-income economies.
  - Contributing factors: infrastructure investment, integration to international markets, institutional upgrades for doing business.
- Modes of diversification:
  - Diversification can occur across products, sectors, or trading partners, and via introduction of new products or expansion/upgrading of existing products.
  - In small states, high production costs and lack of economies of scale make extensive export diversification (introduction of new products) difficult.
  - Export diversification in small states mostly happens via the intensive margin: a more evenly balanced mix of existing export products or trading partners.
- Observed trends and reversals:
  - General trend toward diversification between 1995 and 2005 across most small states (including tourism-dependent economies).
  - Diversification reversed during and after the 2007-09 global financial crisis.
  - Possible drivers of reversal: countercyclical fiscal responses focused on leading sectors, retreat of foreign direct investment, terms-of-trade movements, and loss of preferential trade agreements (e.g., sugar and bananas for the Caribbean).
  - Several tourism-dependent Caribbean economies diversified their tourism source markets since the early 2000s.
  - Economies transformed from diversified to tourism are presently as concentrated as traditionally tourism-dependent economies despite being among the most diversified a decade earlier.
- Group-specific growth and volatility patterns:
  - Tourism-dependent economies: track growth in Advanced Economies, highly synchronized with global business cycle, most impacted by global recessions.
  - Primary commodity exporters: most exposed to international commodity price fluctuations.
  - Diversified economies: greatest resilience — least volatility and higher average growth rates among groups.
  - The traditionally tourism countries and those transformed from diversified to tourism would have had lower volatility and higher average growth rates if they had not specialized.
- Additional observations:
  - Primary-based economies experienced the most drastic change toward specialization since mid-90s alongside the commodity price boom.
  - Traditionally primary commodity exporters had the highest average growth rate mostly attributed to the commodity super cycle between 2000 and 2014; they also demonstrate the highest output volatility.

---

### Measurement of diversification and classification
- Diversification metric:
  - Herfindahl index (HFI) calculated for country-level export flows in products at SITC1 and SITC4 digit level from WITS and travel and service exports from IMF BOP database.
  - HFI = ∑ s_i^2; i = 1,2,...,n.
  - Index varies between 0 and 1; larger values indicate more concentrated export structure; lower values indicate more diversified export structure.
- Indices constructed:
  - HFI1: goods only (at SITC1 digit level), 1990-2014.
  - HFI4: goods only (at SITC4 digit level), 1962-2014.
  - HFI1 with travel: goods + travel, 1990-2014.
  - HFIm: export markets (at country level), 1988-2014.
  - HFI1 with service: goods + service (at BPM6 classification 1 digit level), 1990-2014.
- Classification approach:
  - For the study, the 33 small states are ranked by their dominant industry (export share larger than 50 percent of total) for two periods: 1990-2004 and 2005-15.
  - Countries categorized into eight groupings:
    1. traditionally tourism dependent economies;
    2. traditionally primary commodity exporters;
    3. traditionally manufacturing economies;
    4. traditionally diversified economies;
    5. traditionally other services dominant economies;
    6. transformed from diversified to tourism;
    7. transformed from diversified to manufacturing;
    8. transformed from primary to diversified economies (between the two periods).
  - For this paper, stylized facts focus on groups: “traditionally tourism”, “traditionally primary”, “traditionally diversified”, and “diversified to tourism” (these constitute more than three quarters of all small states).
- Note on “diversified” definition:
  - For the purposes of this study, a country is defined as “diversified” if no dominant industry exists.

---

### Empirical and indicator-based observations
- Export basket concentration:
  - Most small states are heavily dependent on a narrow range of traditional products and markets for bulk of export earnings.
  - Export basket of small states is generally more concentrated compared with other developing and low-income countries and has become increasingly concentrated over the past decade.
  - Compared to others, the service sector plays a more important role in diversifying exports in small states.
- Economic Complexity Index (ECI) patterns:
  - ECI measures knowledge translated into products; computes average ubiquity of products a country exports and average diversity of countries that make those products.
  - In general, ECI is lower in small states compared to Emerging Markets (EM) average and higher than Low-Income Countries (LIC) average.
  - Country examples over past twenty years in the sample:
    - Mauritius: some improvements in economic complexity.
    - Jamaica: relatively static complexity levels.
    - Trinidad and Tobago: declining complexity levels.
  - Implication: small states may need extra efforts to diversify export products and upgrade export baskets toward more complex products.

---

### Key implications and policy-relevant findings (from source text)
- Export diversification can promote higher and more stable per capita income growth, but effects are heterogeneous and may be non-linear.
- Diversification is particularly valuable for small states vulnerable to ToT shocks and export earnings instability.
- Intensive-margin diversification (rebalancing mix of existing products or partners) is more feasible in small states than extensive-margin diversification (introducing many new products) due to high production costs and lack of economies of scale.
- Structural transformations and diversification are sensitive to external shocks and policy responses; the 2007-09 global financial crisis coincided with a reversal of diversification trends in many small states.
- Policy levers highlighted by the literature:
  - Governments can support entrepreneurship and create incentives to invest in new activities (Hausmann et al. perspective).
  - Trade openness interacts with diversification: in sufficiently diversified economies trade openness can lower output volatility; in concentrated exporters trade openness can increase volatility.

*Source: Excerpt from IMF working paper (wp1886) content provided*

### Box 3. Diversification in Mauritius

### Box 3. Diversification in Mauritius

### Mauritius: case study and outcomes
- Economy transformed under a trade-led development: mainly agriculture (e.g. sugar) in 1970s → manufacturing (e.g. garment and jewelry) in 1990s → broad range of service exports (e.g. tourism, finance, information technology, and business process outsourcing) since the early 2000s.
- Political economy and macro outcomes:
  - Mauritius averted recession during the 2007-09 global financial crisis and maintained relatively stable output growth.
  - Growth volatility in 2006-15 more than halved compared to the previous decade (when the economy was less diversified).
  - The average growth rate achieved in the past decade is "around 4 percent".
- Competitiveness and business environment:
  - Ranks 49/190 in 2017 World Bank Doing Business Index (the highest rank among all small states).
  - Ranks 45/138 in World Economic Forum Global Competitiveness Index (also the highest among all states).
  - Competitive labor costs, partly attributed to lower minimum wage requirements, supported manufacturing expansion.

### Empirical analysis framework
- Sample and aggregation:
  - Regression data covers 1990-2014.
  - Averages taken for every 3 years to estimate impacts on long-term growth.
- Growth estimation (Solow Growth Model specification):
  - Growth equation: ∆y_{i,t} = α + β Diversification_{i,t−1} + φ X_{i,t} + γ_t + ε_i + ε_{i,t}
  - ∆y_{i,t} is the real GDP growth rate of country i at time t.
  - Diversification measures (lagged): export goods diversification at different disaggregated levels, export of goods and services diversification, export markets diversification, and real sector diversification.
  - Controls (X_i): population growth, shares of gross capital formation, trade openness, and FDI flows.
  - Education measures not included due to poor data availability for small states.
  - Natural disaster dummy tested but did not improve regression results (time period controlled by year dummies).
- Volatility estimation:
  - Volatility equation: VOL_{i,t} = σ VOL_{i,t−1} + ρ Diversification_{i,t−1} + δ X_{i,t} + γ_t + ε_i + ε_{i,t}
  - VOL_{i,t} is volatility of real GDP growth (standard deviation over each non-overlapping 3-year period).
  - Controls: trade openness, inflation volatility, terms of trade volatility, and exchange rate volatility.
  - Averages of diversification measures and controls taken for each 3-year period.

### Main regression findings (growth)
- General result: Diversification (lower HFI) associated with higher growth and lower volatility in small states.
- Growth regressions (Fixed Effects, 3-year average, 1990-2014) — selected coefficients and exact statistics:
  - Gross capital formation: 0.121*** (0.0402); 0.129*** (0.0410); 0.127*** (0.0452); 0.126*** (0.0457); 0.151*** (0.0442); 0.134*** (0.0494); 0.116*** (0.0429)
  - Trade openness: 4.000*** (1.400); 2.343 (2.161); 3.010 (2.304); 2.996 (2.302); 2.027 (2.225); 3.119 (2.827); 3.317 (2.124)
  - FDI flows: -0.305 (0.309); -0.585* (0.342); -0.496 (0.374); -0.492 (0.374); -0.524 (0.354); -0.516 (0.518); -0.618* (0.349)
  - HFI, 1 digit export goods: -1.596* (0.853) — negative and significant, implying lower HFI (greater diversification) associated with higher growth.
  - HFI, 4 digit export goods: -1.059** (0.495) — negative and significant.
  - HFI, 1 digit export goods and travel: -0.285 (1.113) — negative but insignificant.
  - HFI, 1 digit export goods and services: -0.230 (1.148) — negative but insignificant.
  - HFI, export markets: 0.120 (0.741) — insignificant.
  - HFI, real sector: -1.198 (2.865) — negative but insignificant.
  - Constant terms reported (examples): -22.44*** (8.082); -14.37 (12.63); -18.57 (14.11); -18.28 (13.99); -14.17 (13.17); -17.10 (16.88); -21.03 (13.37)
  - Observations by specification: 170; 154; 147; 147; 148; 125; 154
  - Number of ifs code: 22 across specifications
  - R-squared: 0.283; 0.302; 0.278; 0.278; 0.322; 0.292; 0.283
  - Significance notation: *** p<0.01, ** p<0.05, * p<0.1
- Interpretation: Export diversification across goods at 1-digit and 4-digit levels helps economic growth; other diversification measures negative but often insignificant (data limitations and small states characteristics noted).

### Main regression findings (volatility)
- Volatility regressions (Fixed Effects, 3-year average, 1990-2014) — selected coefficients and exact statistics:
  - Lagged volatility: -0.206 (0.138); -0.180 (0.124); -0.155 (0.128); -0.171 (0.129); -0.0990 (0.135); -0.327 (0.202); -0.163 (0.144)
  - Trade openness: -4.213** (1.859); -1.837 (1.820); -3.052* (1.746); -3.135* (1.758); -3.036* (1.775); -5.186** (2.202); -3.948** (1.876)
  - Inflation volatility: -0.173* (0.0877); -0.135* (0.0794); -0.160 (0.119); -0.172 (0.120); -0.199 (0.119); -0.214* (0.118); -0.163* (0.0883)
  - Terms of trade volatility: very small coefficients reported as (1.36e-10), (1.22e-10), (1.27e-10), (1.28e-10), (1.26e-10), (1.45e-10), (1.37e-10)
  - Exchange rate volatility: 7.608 (6.563); 5.851 (5.900); 1.671 (6.662); 2.775 (6.593); 6.591 (6.180); 12.069 (11.54); 9.519 (6.822)
  - HFI, 1 digit export goods: 2.046*** (0.634) — positive and significant, implying higher HFI (less diversification) associated with higher volatility; equivalently lower HFI (greater diversification) decreases volatility.
  - HFI, 1 digit export goods and travel: 1.680* (0.852) — positive and significant.
  - HFI, 1 digit export goods and services: 1.368* (0.759) — positive and significant.
  - HFI, 4 digit export goods: 0.595 (0.370) — positive but insignificant.
  - HFI, export markets: 1.211* (0.616) — positive and significant, suggesting export markets diversification is a channel to maintain macro stability by mitigating partner-specific shocks.
  - HFI, real sector: -2.786 (2.733) — negative but insignificant.
  - Constant terms reported (examples): 21.85** (8.504); 13.09 (8.081); 18.22** (7.870); 18.88** (7.921); 17.52** (7.989); 27.63** (9.939); 16.09 (10.20)
  - Observations by specification: 65; 65; 61; 61; 65; 61; 65
  - Number of ifs code: 16; 16; 16; 16; 16; 16; 16
  - R-squared: 0.410; 0.540; 0.534; 0.526; 0.535; 0.527; 0.426
- Interpretation: Export goods and services diversification matter for mitigating growth volatility; export market diversification also plays a significant role.

### Synthesis of findings
- Cross-country evidence for small states (1990-2015):
  - More diversified small states experienced the lowest output volatility and a higher average growth rate than most other small states.
  - Export diversification has a more significant impact on reducing output volatility than on improving the long run growth rate in small states.
- Observed caveats:
  - Diversification successes are limited to a small number of small states.
  - A general diversifying trend in the late 1990s and early 2000s was short-lived and reversed during and after the 2007-09 global financial crisis when diversification efforts were deprioritized.
  - Fundamental determinants (policy and institutional factors) mediate the relationship between diversification and growth.

### Policy implications and recommendations
- Consider diversification within a cohesive development strategy emphasizing fundamental determinants:
  - Maintain macroeconomic policy stability.
  - Foster a supportive business environment: quality of infrastructure, human capital, and essential business services.
  - Attract investment to create new comparative advantages and foster development of new sectors.
- Prefer cost-effective policies given limited fiscal space in most small states:
  - Emphasize policy stability and business environment improvements over activist development policies and compensative tax incentives.
- Risk-pooling and economic integration:
  - There is scope to strengthen risk-pooling arrangements (e.g., economic integration as a substitute for diversification) given prevalence of country-specific shocks (e.g., natural disasters) among small states.

*Source: Box 3. Diversification in Mauritius (wp1886).*

### REFERENCES

### REFERENCES

### Export diversification, export composition, and growth
- Acevedo, S., 2014, “Debt, Growth and Natural Disasters: A Caribbean Trilogy,” IMF Working Paper 14/125.
- Agosin, M. R., 2007, “Export Diversification and Growth in Emerging Markets,” Serie Documentors de Trabajo 233, Universidad de Chile.
- Al-Marhubi, F., 2000, “Export Diversification and Growth: An Empirical Investigation,” Applied Economics Letters 7:559-62.
- Berezin, P., A. Salehizadeh, and E. Santana, 2002, “The Challenge of Diversification in the Caribbean,” IMF Working Paper 02/196.
- Greenaway, D., W. Morgan and P.Wright, 1999, “Exports, export composition and growth,” Journal of International Trade and Development 8(1): 41-51.
- Gutierrez-de-Pineres, S.A. and M.Ferrantino, 2000, “Export dynamic and economic growth in Latin America: A comparative perspective,” Burlington, VT, Ashgate.
- Hesse, H., 2008, “Export Diversification and Economic Growth,” Commision on Growth and Development Working Paper 21.
- Imbs, J., and R. Wacziarg, 2003, “Stages of Diversification,” American Economic Review Vol. 93, No. 1.
- Jansen, M., 2004, “Income Volatility in Small and Developing Economies: Export Concentration Matters.” WTO Discussion Paper3, World Trade Organization, Geneva, Switzerland.
- Ledeman, D., and W. F. Maloney, 2007, “Trade Structure and Growth,” Palo Alto: Stanford University Press.
- Love, J., 1986, “Commodity Concentration and Export Earnings Instability: A Shift From Cross-Section to Time Series Analysis,” Journal of Development Economics (24) 2:239-248.
- Matthee, M. and W.A.Naude, 2007, “The determinants of regional manufactured exports from a developing country,” World Institute for Development Economics Research, Research Paper No. 2007/10.
- Michaely, M., 1977, “Exports and growth,” Journal of Development Economics (4)1 :49-53.
- Papageorgiou, C. and N. Spatafora, 2012, “Economic Diversifications in LICs: Stylized Facts and Macroeconomic Implications,” IMF Staff Discussion Note 12/13.
- Sannassee, R. V., B. Seetanah, and M. J. Lamport, 2014, “Export Diversification and Economic Growth: the Case of Mauritius,” Connecting to Global Markets, World Trade Organization.
- Yari, M., 2003, “Export Diversification in Pacific Island Countries,” Bulletin on Asia-Pacific Perspectives 2002/03.

### Export variety, quality, and product space
- Feenstra, R.C. and H.L.Kee, 2004, “Export variety and country productivity,” National Bureau for Economic Research, NBER Working Paper 10830.
- Henn, C., C. Papageorgiou, and N.Spatafora, 2013, “Export Quality in Developing Countries,” IMF Working Paper 13/108.
- Khandelwal, A., 2010, “The Long and Short of Quality Ladders,” Review of Economic Studies, Vol.77, pp 1450-1476.
- Klinger, B., and D. Lederman, 2006, “Diversification, Innovation, and Imitation inside the Global Technological Frontier,” Research Policy Working Paper 3872, World Bank, Washington, D.C.
- Schott, P., 2004, “Across-Product versus Within-Product Specialization in International Trade,” Quarterly Journal of Economics, Vol. 119, pp. 647-678.
- Hausmann, R., and D. Rodrik, 2003, “Economic Development as Self-Discovery,” Journal of Development Economics 72: 603-33.
- Hausmann, R., J. Hwang, and D. Rodrik, 2006, “What You Export Matters,” Center for International Development, Harvard University.
- Hausmann, R., and B. Klinger, 2006, “Structural Transformation and Patterns of Comparative Advantage in the Product Space,” Center for International Development, Harvard University.
- "The Atlas of Economic Complexity," Center for International Development at Harvard University, http://www.atlas.cid.harvard.edu

### Natural disasters, macroeconomic impacts, and resilience
- Cavallo E., S. Galiani, I. Noy, and J. Pantano, 2010, “Catastrophic Natural Disasters and Economic Growth,” IDB Working Paper No. 183.
- Hochrainer, S., 2009, “Assessing the Macroeconomic Impacts of Natural Disasters : Are there Any?” World Bank Policy Research Working Paper 4968.
- Loayza, N. V., E. Olaberria, J. Rigolini, and L. Christiaensen, 2012, “Natural Disasters and Growth: Going Beyond the Averages,” World Development 40/7.
- Noy, I., 2009, “The Macroeconomic Consequences of Disasters,” Journal of Development Ecoomics, Volume 88, Issue 2, March 2009, Pages 221-231.
- Acevedo, S., 2014, “Debt, Growth and Natural Disasters: A Caribbean Trilogy,” IMF Working Paper 14/125.
- International Monetary Fund, 2016, “Small States’ Resilience to Natural Disasters and Climate Change—Role for the IMF,” IMF Policy Paper.

### Trade openness, volatility, and external shocks
- Ghosh, A.R., and J. Ostry., 1994, “Export Instability and the External Balance in Developing Countries,” IMF Staff Papers 41:214-35.
- Haddad, M., J. J  . Lim, C. Pancaro, and C. Saborowski, 2012, “Trade Openness Reduces Growth Volatility when Countries Are Well Diversified,” European Central Bank Working Paper Series, No. 1491.
- Raddatz, C., 2007, "Are external shocks responsible for the instability of output in low-income countries?" Journal of Development Economics, Elsevier, vol. 84(1), pages 155-187, September.
- Loungani, P., M. Saurabh, C. Papageorgiou, and K. Wang. 2017. “World Trade in Services: Evidence from a New Dataset.” IMF Working Paper No. 17/77.

### Structural transformation, development policy, and methodology
- Chenery, H., 1979, “Structural Change and Development Policy,” New York: Oxford University Press.
- Syrquin, M., 1989, “Patterns of Structural Change,” in Handbook of Economic Development, H. Chenery and T.N. Srinavasan, eds. Amsterdam: Elsevier Science Publishers.
- McMillan, M., and D. Rodrik. and I. Verduzco-Gallo. 2014. “Globalization, Structural Change, and Productivity Growth with an Update on Africa.” World Development 63: 11–32.
- International Monetary Fund, 2014, “Sustaining Long-run Growth and Macroeconomic Stability in Low-income Countries—the Role of Structural Transformation and Diversification,” IMF Policy Paper.
- Di Salvo, M., 2015, “Non-linearity Between Export Diversification and Economic Growth: How Quality of Exports and Openness to Trade can influence the U-shaped Relationship,” Erasmus School of Economics.
- Feenstra, R.C. and H.L.Kee, 2004, “Export variety and country productivity,” National Bureau for Economic Research, NBER Working Paper 10830.

*Source: wp1886 - REFERENCES*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2018/wp1886.pdf_