## Introduction — Potential Growth and Productivity in the Caribbean (wpiea2025157-source-pdf)

## Source details

**Canonical URL:** [Introduction — Potential Growth and Productivity in the Caribbean (wpiea2025157-source-pdf)](https://www.imf.org/-/media/files/publications/wp/2025/english/wpiea2025157-source-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2025/english/wpiea2025157-source-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2025/english/wpiea2025157-source-pdf.pdf.json)

---

### Scope, approach, and research questions
- Focus: trends and prospects of potential growth in the Caribbean.
- Two complementary analyses:
  - Macroeconomic: potential growth and its components using (i) medium-term growth projections by forecasters and (ii) a growth accounting framework.
  - Microeconomic: firm-level analysis using the Innovation, Firm Performance and Gender (IFPG) survey to study resource misallocation and structural obstacles.
- Three guiding questions:
  - What is the region’s potential growth: trends and prospects?
  - What are the key components of potential growth, and which have changed over time?
  - What can be inferred from a micro perspective about causes and consequences of changes in those components?
- Macroeconomic factors analyzed: total factor productivity (TFP), human capital, population, and capital stock.

### Aggregate findings — potential growth trends and decomposition
- Caribbean-wide potential growth by decade:
  - 1981-1990: 4.2 percent
  - 1991-2000: 3.2 percent
  - 2001-2010: 2.1 percent
  - 2011-2019: 1.2 percent
- Component contributions (Caribbean-wide comparisons, 1981-1990 → 2011-2019):
  - TFP contribution: 1.1 percent → 0.1
  - Human capital contribution: 1.6 percent → 0.2
  - Capital stock contribution: 1.1 percent → 0.7
  - Population contribution: 0.3 percent → 0.3
- Counterfactual (Caribbean-wide): If TFP and human capital contributions in 2011-2019 matched 1981-1990 levels, potential growth would increase from 1.2 to 3.7 percent.
- Forecaster expectations: Five-year-ahead growth forecasts have dropped from around 3.5 percent to 2 percent; robustness checks indicate this decline is not driven by rising forecaster pessimism or income convergence.

### Firm-level framework, data, and parameterization
- Microeconomic motivation: declining aggregate TFP and human capital motivate firm-level exploration of frictions and misallocation.
- Misallocation framework: Hsieh and Klenow (2009) adapted to country-sector firm panels with Cobb-Douglas firm production:
  - Firm production: Y_csi = A_csi K_csi^{α_c} L_csi^{1−α_c}
  - Two productivity measures: physical TFP (TFPQ) and revenue TFP (TFPR)
  - Aggregate efficiency computed by equalizing firm-specific distortions and aggregating (Hsieh and Klenow ratio).
- Parameter choices and inputs:
  - Elasticity of substitution: σ = 3.3
  - Output elasticities: α_c = 0.4 for manufacturing; α_c = 0.33 for services
  - Financial parameters: r = 0.1; w normalized to 1
  - Firm measures: value added for Y, net book value of capital (capacity-adjusted) for K, total compensation for L
  - Outlier trimming: top and bottom two percent of log TFPR/TFPQ by sector
  - Dataset: IFPG survey — cross-sectional 2020 survey of 1,979 firms across 13 Caribbean countries; financials largely refer to fiscal year 2019 (pre-pandemic).

### Core firm-level results — misallocation magnitudes and patterns
- Aggregate TFP gains from eliminating misallocation (country variation, percent increases):
  - Guyana: 107 percent
  - Belize: 101 percent
  - Jamaica: 96 percent
  - Suriname: 86 percent
  - Trinidad and Tobago: 81 percent
  - Tourism-based countries: Grenada 34 percent → The Bahamas 65 percent
  - Most countries fall between 50 and 70 percent
- Sectoral patterns:
  - In eight countries greater misallocation in manufacturing; in five countries greater misallocation in services.
  - Commodity-based countries (Guyana, Suriname, Trinidad and Tobago): misallocation in services greater by around 20 to 30 percentage points.
  - Manufacturing often displays greater misallocation in tourism-based economies.
- Factor-specific dispersion (relative standard deviation ratios, value-added weighted):
  - Capital dispersion larger than labor dispersion in all countries except Belize and Dominica.
  - Average relative ratio: capital ≈ 0.6, labor ≈ 0.4
- Efficient firm-size distribution implications:
  - Efficient distribution has larger mean and variance than observed distribution.
  - Under efficient allocation: 36.1 percent of firms become smaller; 63.9 percent of firms become larger.
  - The largest 20 percent of firms are projected to become substantially larger (notably weight in 300 percent+ category).

### Structural obstacles reported by firms and links to human capital
- Top obstacles rated “major” or “very severe” (Caribbean-wide shares):
  - “Access to finance”: 71 percent
  - “Customs and trade regulations”: 66 percent
  - “Inadequately educated workforce”: 58 percent
- One-choice “most serious” ranking order:
  - “Inadequately educated workforce”
  - “Access to finance”
  - “Customs and trade regulations”
- Financing and cost indicators:
  - Only 16 percent of firms rely on medium/long-term loans on average.
  - Average interest rates (survey averages):
    - Line of credit: 11 percent
    - Overdraft facility: 10 percent
    - Credit card: 20 percent
    - Medium/long-term loans: 13 percent
  - Global median interest rate for SMEs in 2018 was 4 percent (OECD, 2020) — Caribbean costs are relatively high.
- Causes of skills shortages (rated “very important” or “critical”):
  - “Quality of the education and training offered by local educational institutions”: 62 percent
  - “Lack of necessary personal, soft skills offered by local institutions”: 51 percent
  - “Shortage in number of local professionals trained by local institutions”: 50 percent
- Worker emigration is not commonly reported as a top cause of skills shortages.

### Econometric associations — obstacle effects on firm TFP
- Productivity correlates:
  - Top decile firms are over 200 percentage points more productive relative to lowest decile across value added and employment deciles.
- Regression setup:
  - Dependent variable: log(TFPQ_c s i)
  - Controls: employment, age, capacity utilization, majority foreign-owned, research department; country-sector fixed effects; robust SEs clustered at country-sector level.
  - Obstacle dummies: major/very severe = 1 for 18 obstacles.
- Key statistically significant associations (All firms sample; coefficient interpretation in percentage points):
  - “Cost of finance”: −0.114 (−11.4 percentage points), significant at p < .01
  - “Tax administration”: −0.099 (−9.9 percentage points), significant at p < .05
  - “Business licensing and permits”: −0.085 (−8.5 percentage points), significant at p < .1
  - “Inadequately educated workforce”: −0.052 (−5.2 percentage points), significant at p < .1
- Heterogeneity:
  - Small firms (≤ 20 employees): stronger negative associations for “Cost of finance”, “Access to finance”, “Tax administration”, “Inadequately educated workforce”.
  - Small & young firms (≤ 20 employees & ≤ 20 years): largest negative associations — productivity declines around 23, 20, 16 and 13 percentage points for “Access to finance”, “Tax administration”, “Business licensing and permits”, and “Labor regulations”, respectively.
  - Large and old firms: obstacle coefficients not significant.
  - Old firms: “Cost of finance” and “Transportation” significant (transportation associated with lower productivity).

### Simulated aggregate gains from removing obstacles and income convergence implications
- Simulated country-level aggregate TFP gains by removing individual obstacles (examples):
  - Removing “Cost of finance”: country gains roughly from 1.5 percent (Antigua and Barbuda) to 6 percent (Belize).
  - Belize shows large simulated gains for several obstacles (Tax administration ~5.5 percent; Inadequately educated workforce ~5 percent; Business licensing and permits ~6 percent).
- Total gains from sets of significant obstacles (Table 6 summaries; totals are simple sums of individual obstacle effects):
  - At 5 percent significance (including “Cost of finance” and “Tax administration”): total gains by country (percent):
    - Suriname: 3
    - Antigua and Barbuda: 4
    - Barbados: 4
    - Belize: 12
    - Dominica: 7
    - Grenada: 8
    - Guyana: 7
    - Jamaica: 4
    - St. Kitts and Nevis: 5
    - St. Lucia: 9
    - St. Vincent and the Grenadines: 6
    - The Bahamas: 4
    - Trinidad and Tobago: 8
  - At 10 percent significance (including also “Inadequately educated workforce” and “Business licensing and permits”): total gains by country include:
    - St. Vincent and the Grenadines: 8
    - Belize: 23
    - (Other country totals reported in Table 6 under the 10 percent column; see source table for full list)
- Income convergence exercise (assumptions: gains to labor productivity equal gains to TFP; no change in labor participation):
  - The Bahamas could converge to around 90 percent of the United States’ GDP per capita.
  - Other countries could achieve convergence improvements between 30 and 60 percent.

### High-level conclusions and policy-relevant implications
- High-level conclusion:
  - Broad-based decline in potential growth in the Caribbean is driven mainly by declining contributions of TFP and human capital; firm-level evidence points to resource misallocation and structural obstacles as contributing factors.
- Priority policy areas to raise potential growth and TFP:
  - Reduce cost of finance and improve access to finance, especially for small and young firms.
  - Improve tax administration processes (reduce compliance burden, speed VAT refunds, digitize filings).
  - Address workforce quality gaps via investments in education and training:
    - Improve quality of education and training offered by local institutions.
    - Strengthen soft-skill training and expand the quantity of trained local professionals.
  - Streamline business licensing and permits to ease firm start-up and expansion.
  - Target capital misallocation: address firm-specific distortions related to capital allocation.
- Complementarity caveat: combined reforms may yield larger gains than the simple sum of individual reforms (e.g., cheaper finance could enable investments in human capital).

*IMF WORKING PAPERS — Potential Growth and Productivity in the Caribbean, Introduction and selected results (wpiea2025157-source-pdf).*

### Introduction ...........................................................................................................

### wpiea2025157-source-pdf - Introduction ...........................................................................................................

### Introduction
- Page references and structure:
  - Introduction .................................................................................................................................................................. 2
  - Related Literature ......................................................................................................................................................... 4
  - Potential growth ...................................................................................................................................................... 4
  - Misallocation and structural obstacles .................................................................................................................... 5

### Trends and Prospects of Potential Growth
- Sections and analytical components:
  - Trends and Prospects of Potential Growth................................................................................................................ 7
  - Insights from growth forecasts ................................................................................................................................ 7
  - Insights from a growth accounting exercise .......................................................................................................... 10

### Results (Aggregate and Accounting)
- Results .................................................................................................................................................................. 12
- Empirical presentation and summarized findings are located in:
  - Results .................................................................................................................................................................. 12

### Evidence from Firm-level Data
- Structure of firm-level analysis:
  - Evidence from Firm-level Data .................................................................................................................................. 15
  - Overview ............................................................................................................................................................... 15
  - Model .................................................................................................................................................................... 15
  - Data ...................................................................................................................................................................... 18
  - Results .................................................................................................................................................................. 19

### Conclusion and Supporting Material
- Conclusion.................................................................................................................................................................. 33
- Appendix ..................................................................................................................................................................... 34
- References .................................................................................................................................................................. 43

### Figures (listed in source)
- Figure 1a & Figure 1b ........................................................................................................................................................................ 2
- Figure 2a & Figure 2b ........................................................................................................................................................................ 8
- Figure 3a & Figure 3b ........................................................................................................................................................................ 9
- Figure 4a & Figure 4b ........................................................................................................................................................................ 9
- Figure 5 ........................................................................................................................................................................................... 12
- Figure 6 ........................................................................................................................................................................................... 13
- Figure 7 ........................................................................................................................................................................................... 20
- Figure 8 ........................................................................................................................................................................................... 21
- Figure 9 ........................................................................................................................................................................................... 22
- Figure 10 ......................................................................................................................................................................................... 23
- Figure 11 ......................................................................................................................................................................................... 23
- Figure 12 ......................................................................................................................................................................................... 24
- Figure 13a ....................................................................................................................................................................................... 25
- Figure 13b ....................................................................................................................................................................................... 26
- Figure 14 ......................................................................................................................................................................................... 27
- Figure 15 ......................................................................................................................................................................................... 32
- Figure A2 ......................................................................................................................................................................................... 35
- Figure A3 ......................................................................................................................................................................................... 36
- Figure A4 ......................................................................................................................................................................................... 37
- Figure A5 ......................................................................................................................................................................................... 39
- Figure A6 ......................................................................................................................................................................................... 41

### Tables (listed in source)
- Table 1 ............................................................................................................................................................................................ 11
- Table 2 ............................................................................................................................................................................................ 19
- Table 3 ............................................................................................................................................................................................ 24
- Table 4 ............................................................................................................................................................................................ 29
- Table 5 ............................................................................................................................................................................................ 31
- Table 6 ............................................................................................................................................................................................ 32
- Table A1 .......................................................................................................................................................................................... 34

### Document identification
- IMF WORKING PAPERS Potential Growth and Productivity in the Caribbean
- INTERNATIONAL MONETARY FUND 2

*Source: wpiea2025157-source-pdf - Introduction ...........................................................................................................*

### Introduction

### wpiea2025157-source-pdf - Introduction

### Introduction: scope and research questions
- Focus: trends and prospects of potential growth in the Caribbean.
- Two complementary analyses:
  - Macroeconomic: potential growth and its components for each country using (i) medium-term growth projections by forecasters and (ii) a growth accounting framework.
  - Microeconomic: firm-level analysis using the Innovation, Firm Performance and Gender (IFPG) survey to study resource misallocation and structural obstacles.
- Three guiding questions:
  - What is the region’s potential growth: trends and prospects?
  - What are the key components of potential growth, and which have changed over time?
  - What can be inferred from a micro perspective about causes and consequences of changes in those components?
- Macroeconomic factors analyzed: total factor productivity (TFP), human capital, population, and capital stock.

### Key findings (macroeconomic and aggregate results)
- Forecaster growth expectations on the Caribbean have declined over time; robustness checks indicate this is not driven by rising forecaster pessimism or income convergence.
- Potential growth (growth accounting estimates) has declined across all countries in the sample.
- Average potential growth in the region:
  - 4.2 percent between 1981-1990
  - 1.2 percent between 2011-2019
- The decline in potential growth is largely driven by declining contributions of TFP and human capital, rather than population or capital stock.
- Counterfactual: If TFP and human capital contributions remained at 1981-1990 levels, potential growth would increase from 1.2 to 3.7 percent in the most recent period.

### Key findings (firm-level and misallocation results)
- Using Hsieh and Klenow (2009) framework, aggregate TFP is significantly affected by firm-level resource misallocation, with large cross-country variation:
  - Potential aggregate TFP gains from removing firm-specific distortions range from 34 percent in Grenada to 107 percent in Guyana.
- In terms of GDP per capita convergence with the United States, removing misallocation could narrow the gap by between 9 and 36 percentage points (country-specific ranges provided in the paper).
- Firms report workforce education as a major or very severe obstacle, linked to quality, skills, and number of professionals produced by local institutions.
- Regression analyses identify significant firm-level obstacles associated with lower firm TFP:
  - workforce education, cost and access to finance, tax administration, and business licensing and permits.
- Heterogeneity by firm characteristics: small and young firms display the largest magnitudes (most severely affected).
- Aggregate TFP could increase between 8 and 23 percent across countries if identified structural obstacles were alleviated among firms (assuming a causal relationship).

### High-level conclusion
- Broad-based decline in potential growth in the Caribbean is driven by declining TFP and human capital contributions; evidence points to resource misallocation and firm-level obstacles as contributing factors.

*Italicized source attribution line: IMF WORKING PAPERS — Potential Growth and Productivity in the Caribbean, Introduction (wpiea2025157-source-pdf - Introduction).*

### 0.35 and 0.65, respectively, in Gollin (2002).

### wpiea2025157-source-pdf - 0.35 and 0.65, respectively, in Gollin (2002).

### Potential growth estimates — Caribbean-wide results
- Potential growth declined over time across the Caribbean:
  - 1981-1990: 4.2 percent
  - 1991-2000: 3.2 percent
  - 2001-2010: 2.1 percent
  - 2011-2019 (most recent pre-pandemic): 1.2 percent
- Main contributors to the decline:
  - Total factor productivity (TFP) and human capital contributions declined substantially:
    - TFP contribution: 1.1 percent (1981-1990) → 0.1 (2011-2019)
    - Human capital contribution: 1.6 percent (1981-1990) → 0.2 (2011-2019)
  - Capital stock and population contributions were more stable:
    - Capital stock: 1.1 percent (1981-1990) → 0.7 (2011-2019)
    - Population: 0.3 percent (1981-1990) → 0.3 (2011-2019)
- Counterfactual: if TFP and human capital contributions in 2011-2019 matched 1981-1990, potential growth would increase from 1.2 to 3.7 percent.
- Country-level note: In all sample countries potential growth (2011-2019) < potential growth (1981-1990); some countries (Antigua and Barbuda, Belize, Dominica, Jamaica, St. Lucia, St. Kitts and Nevis, St. Vincent and the Grenadines) follow Caribbean-wide dynamics; The Bahamas and Grenada show increased TFP contributions over time.

*Sources for aggregates and components: Barro and Lee (2015), EM-DAT database, IAB database, World Bank database, Penn World Table (10.1), and Authors’ calculations.*

### Firm-level microanalysis — motivation and framework
- Motivation: declining aggregate contributions of TFP and human capital motivate firm-level exploration to identify underlying frictions.
- Two channels for TFP suppression:
  - Firm-level poor business environment reduces investment and technological progress.
  - Aggregate-level poor allocative efficiency (misallocation) due to dispersion in firms’ marginal revenue products of inputs under monopolistic competition and a Cobb-Douglas production function.
- Framework: Hsieh and Klenow (2009) approach adapted to country-sector firm panels:
  - Aggregate output across S industries: Y_c = ∏_s Y_c s^{θ_c s}, with ∑_s θ_c s = 1.
  - Industry CES aggregation over M_c s differentiated firms with elasticity of substitution σ.
  - Firm production: Y_c s i = A_c s i K_c s i^{α_c} L_c s i^{1−α_c}, where α_c (capital elasticity) and 1−α_c (labor elasticity).
  - Firms face firm-specific output and capital distortions τ^Y_{c s i} and τ^K_{c s i}; marginal revenue products deviate from common levels due to distortions.
  - Two productivity measures:
    - Physical TFP (TFPQ): A_c s i = Y_c s i / (K_c s i^{α_c} L_c s i^{1−α_c})
    - Revenue TFP (TFPR): G_c s i A_c s i = G_c s i Y_c s i / (K_c s i^{α_c} L_c s i^{1−α_c})
  - Efficient/undistorted aggregates computed by equalizing distortions across firms and aggregating (Hsieh and Klenow ratio for potential increase in aggregate output).

### Model parameterization and data inputs
- Elasticity of substitution: σ = 3.3 (median value from Broda et al. (2017)).
- Output elasticities (factor-share approach, pooling across countries):
  - α_c = 0.4 for manufacturing
  - α_c = 0.33 for services
  - 1−α_c is the residual for labor elasticity.
- Financial parameters:
  - r = 0.1 (capital rental cost)
  - w normalized to 1
- Firm-level measurement:
  - Value added used for Y_c s i (sales minus intermediate inputs)
  - Net book value of capital for K_c s i (land, buildings, machinery, vehicles, equipment), adjusted by firm-reported capacity utilization
  - Total compensation for L_c s i (wages, social payments, salaries, bonuses)
- Outlier treatment: trim top and bottom two percent of distributions of log TFPR/TFPQ relative measures by sector before recalculating sector aggregates.
- Dataset: Innovation, Firm Performance and Gender (IFPG) survey — cross-sectional 2020 survey of 1,979 firms across 13 Caribbean countries; financial variables refer largely to fiscal year 2019 (pre-pandemic).

### Core firm-level findings — misallocation, firm-size, and sectoral patterns
- Aggregate TFP gains from eliminating misallocation (country variation):
  - Guyana: 107 percent
  - Belize: 101 percent
  - Jamaica: 96 percent
  - Suriname: 86 percent
  - Trinidad and Tobago: 81 percent
  - Tourism-based countries: Grenada 34 percent → The Bahamas 65 percent
  - Most countries fall between 50 and 70 percent
- Sectoral misallocation:
  - Mixed pattern: in eight countries greater misallocation in manufacturing; in five countries greater misallocation in services.
  - Commodity-based countries (Guyana, Suriname, Trinidad and Tobago): misallocation in services greater by around 20 to 30 percentage points.
  - Manufacturing often displays greater misallocation in tourism-based economies.
- Factor-specific dispersion (relative standard deviation ratios, weighted by value added shares):
  - Greater dispersion in the capital component than the labor component for both manufacturing and services in all countries except Belize and Dominica.
  - Average relative ratio: capital ≈ 0.6, labor ≈ 0.4 (implying capital misallocation typically more severe, but labor misallocation also substantial).
- Efficient firm-size distribution:
  - Efficient distribution has larger mean and variance than observed distribution — firms on average need to scale up, though some firms are inefficiently too large.
  - Aggregate movement under efficient allocation:
    - 36.1 percent of firms become smaller
    - 63.9 percent of firms become larger
  - Largest 20 percent of firms are projected to become substantially larger (notably weight in 300 percent+ category).

### Structural obstacles reported by firms — prevalence and channel to human capital
- Top obstacles rated as “major” or “very severe” (Caribbean-wide):
  - “Access to finance”: 71 percent
  - “Customs and trade regulations”: 66 percent
  - “Inadequately educated workforce”: 58 percent
- Top obstacles when firms choose a single “most serious” obstacle (one-choice ranking): order becomes:
  - “Inadequately educated workforce”
  - “Access to finance”
  - “Customs and trade regulations”
- Financing profile (survey):
  - On average, only 16 percent of firms rely on medium/long-term loans for financing.
  - Average interest rates (survey averages):
    - Line of credit: 11 percent
    - Overdraft facility: 10 percent
    - Credit card: 20 percent
    - Medium/long-term loans: 13 percent
  - Global context: global median interest rate for SMEs in 2018 was 4 percent (OECD, 2020) — Caribbean costs are relatively high.

- Reasons firms cite for skills shortages (factors rated “very important” or “critical”):
  - “Quality of the education and training offered by local educational institutions”: 62 percent
  - “Lack of necessary personal, soft skills offered by local institutions”: 51 percent
  - “Shortage in number of local professionals trained by local institutions”: 50 percent
- Worker emigration is not commonly reported as a top cause of skills shortages.

### Econometric associations — productivity regressions and obstacle impacts
- Productivity is positively associated with firm size and age:
  - Top decile (90-100 percent) firms are over 200 percentage points more productive relative to the lowest decile (0-10 percent) across value added and employment deciles.
- Regression specification for associations:
  - Dependent variable: log(TFPQ_c s i)
  - Controls: employment, age, capacity utilization, majority foreign-owned dummy, research department dummy
  - Obstacle dummies: 18 obstacles (major/very severe = 1)
  - Country-sector fixed effects included; robust SEs clustered at country-sector level.
- Key statistically significant associations (All firms sample):
  - “Cost of finance”: −0.114 (−11.4 percentage points), significant at p < .01
  - “Tax administration”: −0.099 (−9.9 percentage points), significant at p < .05
  - “Business licensing and permits”: −0.085 (−8.5 percentage points), significant at p < .1
  - “Inadequately educated workforce”: −0.052 (−5.2 percentage points), significant at p < .1
- Heterogeneity by firm type:
  - Small firms (≤ 20 employees): stronger negative associations for “Cost of finance”, “Access to finance”, “Tax administration”, “Inadequately educated workforce”.
  - Small & young firms (≤ 20 employees & ≤ 20 years): largest negative associations — productivity declines around 23, 20, 16 and 13 percentage points for “Access to finance”, “Tax administration”, “Business licensing and permits”, and “Labor regulations”, respectively.
  - Large and old firms: no significant obstacle coefficients (suggesting larger firms may have adapted or grown out of constraints).
  - Old firms: “Cost of finance” and “Transportation” are significant (transportation associated with lower productivity for older firms).

### Aggregate implications of removing obstacles
- Simulated aggregate TFP gains by removing each obstacle (apply γ_d to TFPR/TFPQ and aggregate):
  - Removing “Cost of finance” implies country gains ranging roughly from 1.5 percent (Antigua and Barbuda) to 6 percent (Belize).
  - Belize shows largest simulated gains for several obstacles (Tax administration ~5.5 percent; Inadequately educated workforce ~5 percent; Business licensing and permits ~6 percent).
  - Country-level gains vary from close to zero up to around 5 percent per obstacle depending on frequency and value-added weights.
- Total gains (simple sum of individual obstacle aggregate effects) reported in Table 6:
  - At 5 percent significance (including “Cost of finance” and “Tax administration”): total gains by country range:
    - Suriname: 3
    - Antigua and Barbuda: 4
    - Barbados: 4
    - Belize: 12
    - Dominica: 7
    - Grenada: 8
    - Guyana: 7
    - Jamaica: 4
    - St. Kitts and Nevis: 5
    - St. Lucia: 9
    - St. Vincent and the Grenadines: 6
    - The Bahamas: 4
    - Trinidad and Tobago: 8
  - At 10 percent significance (including also “Inadequately educated workforce” and “Business licensing and permits”): total gains by country range:
    - St. Vincent and the Grenadines: 8
    - Belize: 23
    - (Other country totals reported in Table 6 under the 10 percent column; see source table for full list)
- Income convergence exercise:
  - Assuming gains to labor productivity equal gains to TFP and no change in labor participation, reallocating resources could substantially improve income convergence with the United States:
    - The Bahamas could converge to around 90 percent of the United States’ GDP per capita.
    - Other countries could achieve convergence improvements between 30 and 60 percent.

### Policy-relevant implications (derived from findings)
- Primary areas for structural policy action to raise potential growth and TFP:
  - Reduce cost of finance and improve access to finance, especially for small and young firms (high interest rates and low reliance on medium/long-term loans identified).
  - Improve tax administration processes (reduce compliance burden, speed VAT refunds, digitize filings).
  - Address workforce quality gaps through investments in education and training:
    - Improve quality of education and training offered by local institutions.
    - Strengthen soft-skill training and expand the quantity of trained local professionals.
  - Streamline business licensing and permits to ease firm start-up and expansion.
  - Target capital misallocation: address firm-specific distortions related to capital allocation since capital dispersion is prominent.
- Complementarity note: Combined reforms may yield larger gains than the sum of individual reforms (e.g., cheaper finance could enable investments in human capital).

*Italic source attribution: IMF Working Papers — Potential Growth and Productivity in the Caribbean (Authors’ calculations; IFPG survey; Barro and Lee (2015); Penn World Table (10.1); Broda et al. (2017); Hsieh and Klenow (2009) — as presented in the supplied content).*

### Conclusion

### Conclusion

### Recent trends in growth and potential
- Five-year-ahead growth forecasts have dropped from around 3.5 percent to 2 percent.
- Actual growth has followed a similar downward trend.
- Potential growth is declining in the region, driven by declining contributions of TFP and human capital.

### Macroeconomic and firm-level approach and key findings
- The paper provides estimates of potential GDP growth for the Caribbean region using macroeconomic and firm-level data in a complementary way.
- From a micro perspective, there is significant scope for resource reallocation between firms, with aggregate TFP increasing between 34 and 107 percent depending on the country.
- Structural obstacles are associated with lower firm productivity, with effects on small and young firms relatively more severe.
- Assuming a causal relationship, the removal of these obstacles could increase aggregate TFP by between 8 and 23 percent across the region.

### Identified structural obstacles
- Workforce education.
- Cost and access to finance.
- Tax administration.
- Business licensing and permits.

### Appendix: data on five-year-ahead growth forecasts (WEO and SR adjustments)
- Five-year-ahead forecasts on real GDP growth for the Caribbean countries were collected from the IMF World Economic Outlook (WEO) database from 2007 onward.
- Historical real GDP growth prior to 2007 was derived from real GDP levels reported in the WEO.
- IMF Staff Reports (SRs) were used as an alternative source to adjust for potential anomalies.
- Step 1: Identify an anomaly in the historical WEO from derived GDP growth rates in years that are 2.5 percentage points higher or 2.5 points lower than the previous year, or years in which the derived real GDP growth rates are zero.
- Step 2: Replace the potential anomaly with information from the SR of the relevant year where available. The relevant year is five years before the year of the forecast (e.g., the 2006 forecast is based on the 2001 SR). If no SR was published for the relevant year, replace the potential anomaly with the average from the SR in year t-2, t-1, t+1, and t+2, where available.
- The table in the source provides the data points adjusted from the SRs.

*Source: Conclusion and Appendix (Data on Five-Year-Ahead Growth Forecasts) — Potential Growth and Productivity in the Caribbean, Working Paper No. WP/2025/157*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2025/english/wpiea2025157-source-pdf.pdf_