## Selected Growth Trends, 1980–2004

## Source details

**Canonical URL:** [Selected Growth Trends, 1980–2004](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06235.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06235.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06235.pdf.json)

---

### Introduction: puzzle and central hypothesis
- Real GDP grew, on average, by "1.6 percent" a year from 1980 to 2004, while investment rose from "15 percent of GDP" to "33 percent" over the same period.
- Two major measurement issues partially explain the “high investment–low growth” puzzle:
  - Official estimates likely understate GDP growth because the informal economy has grown rapidly and is not fully captured in statistics.
  - The capital stock appears overestimated (low capacity utilization, hurricane damage, and nonproductive investment suggest measured capital may overstate productive capital).
- Central hypothesis: high public debt adversely affects productivity and the allocation of investment by:
  - Distorting allocation of investment toward less productive areas.
  - Reducing scope for public sector investment that complements private investment.
- Cross-country evidence summary: using total public debt (external and domestic) as the key explanatory variable, a doubling of total public debt leads to a reduction in productivity growth of about "1.5 percentage points."

### The paradox: endowments, outcomes, and macro context
- Endowments and potential:
  - Natural beauty, mining resources, well-educated English-speaking workforce, proximity to North America.
- Actual outcomes and macro context:
  - GDP per capita (constant 2000 US$) over 1993–2003 remained unchanged at about "US$3,150", compared with 1970 GDP per capita of "US$3,328".
  - Public debt averaged above "110 percent of GDP" for the last decade; total public debt was "144 percent of GDP" during FY 2003/04.
  - Domestic debt was "67 percent of GDP"; external debt was "77 percent of GDP".
  - Macroeconomic uncertainty, high unemployment, emigration, and rising crime likely dampened output growth.

### Constraints to growth identified
- Size and vulnerability:
  - Small size and vulnerability to external events and natural disasters increase volatility and impair growth.
- Competitiveness and labor costs:
  - Real effective exchange rate appreciated by more than "35 percent" from 1990 to 2001.
  - World Bank (2003): a "50 percent" decline in Jamaica’s market share of world merchandise exports from 1994 to 2001.
  - Unit labor costs increased twice as fast in Jamaica as in its major trading partners from 1994 to 2001 without comparable productivity gains.
  - Planning Institute of Jamaica (PIOJ) data: salary earnings almost doubled between 2005 and 2000 when employment fell by over "2 percent".
  - Household surveys: about "80 percent" of tertiary graduates emigrated from Jamaica in the 1990s.
- Access to finance:
  - Ratio of net government debt to bank deposits was "45 percent" in June 2003 (excluding "10 percent" for public enterprise debt).
  - High public debt pushed up interest rates and crowded out credit to the private sector.
  - Structural constraints include outdated insolvency procedures, absence of credit registries, dissatisfaction with courts and long litigation delays.
- Crime:
  - Intentional homicide rate in 2001 = "44 per 100,000 inhabitants".
  - World Bank estimated annual cost of crime at around "5 percent of GDP" (including investment in fighting crime and production losses).
  - Government spending on crime control estimated at "3.1 percent of GDP".
  - Lost production and health care expenses due to violent crimes cost the economy "0.6 percent of GDP".
  - Private expenditure on security estimated at "1.3 percent of GDP".
  - World Bank survey: "42 percent" of business managers considered themselves likely to be murdered at the workplace.

### Declining productivity as the core driver of low growth
- Growth decomposition (1960–2000 and 1990–2000 comparisons):
  - Capital accumulation contributed positively, but a steady decline in productivity dominated poor performance.
  - Measured TFP performance:
    - Jamaica: decline in productivity of "0.5 percent" a year from 1960 to 2000.
    - Latin America: recorded a "0.2 percent" increase in productivity for the same period.
    - World average productivity: increased by "0.9 percent" for the same period.
  - Output growth in Jamaica was below Latin America and a sample of 84 countries by almost "2.5 percentage points" during both 1960–2000 and 1990–2000.
  - Human capital accumulation (education) was in line with Latin American and world averages; physical capital slightly lower.
  - The large differential (as high as "2.5 percent" during the 1990s) in TFP growth is the principal factor behind Jamaica’s underperformance.

### The mismeasurement story: informal economy and capital stock concerns
- Informal economy size and impact:
  - Informal economy share rose from "12.9 percent of GDP in 1991 to about 40 percent in 2000."
  - Correcting for underestimation of the informal economy, average GDP growth during 1991–2000 may be estimated as high as "3.0 percent," or "2.7 percent" higher than officially recorded.
  - Selected measures (2000–01):
    - Monetary approach: Share over registered GDP, 2000 = "39.1"
    - Monetary approach: Share over registered GDP (predicted), 2001 = "43.7"
    - Monetary approach: Share over total GDP, 2000 = "28.1"
    - Electricity Consumption Method: Share over total GDP, 2000 = "45.5"
    - Electricity Consumption Method (conservative scenario), 2000 = "40.9"
    - Method of Additions: Share over registered GDP, 2001 = "43.5"
    - Method of Additions (conservative scenario), 2001 = "42.9"
  - Table 2 highlights (1991–2000, constant prices):
    - Formal GDP: 1991 = "19,098"; 2000 = "19,603"
    - Informal GDP: 1991 = "3,344"; 2000 = "12,586"
    - Total GDP: 1991 = "21,327"; 2000 = "27,994"
- Capital stock measurement issues:
  - Capacity utilization could be as low as "50–60percent" in the manufacturing sector.
  - Hurricane damage (PIOJ estimates):
    - Hurricane Gilbert (1988) = "65 percent of GDP"
    - Hurricane Ivan (2004) = "8 percent"
    - Hurricanes Dennis and Emily (2005) = "1 percent"
    - Michelle (2001) = "1 percent"
    - Hurricane in 1980 = "2 percent"
    - Floods in 1986 = "3 percent"
    - Floods in 1991 = "6 percent"
  - High investment may partly reflect substitution of capital for labor due to high real wages and labor rigidities and real exchange rate appreciation in the 1990s.
  - Crime-prevention-related expenditure estimated by World Bank (2004) up to "4.4 percent of GDP" ("3.1 percent" public sector; "1.3 percent" private sector).

### Composition of investment and productive content
- Composition highlights (1994–2004):
  - Building construction as percent of total investment: 1995 = "45.6"; 2004 = "40.3"
  - Transport equipment total: 1995 = "12.4"; 2004 = "7.1"
    - Motor cars: 1995 = "2.1"; 2004 = "1.4"
    - Trucks and buses: 1995 = "7.1"; 2004 = "2.4"
    - Other transport equipment: 1995 = "3.2"; 2004 = "3.3"
  - Other machinery and equipment: 1995 = "37.9"; 2004 = "48.8"
  - Industrial machinery and equipment: 1995 = "12.7"; 2004 = "9.7"
  - Other machinery and other capital goods: 1995 = "24.0"; 2004 = "38.4"
  - Total composition percentages sum to "100.0" each year.

### Cross-country evidence: public debt and productivity
- Empirical specification:
  - Dependent variable: total factor productivity (TFP) growth.
  - Main explanatory variable: log of total public debt, with a spline allowing a structural break at "21 percent of GDP."
  - Controls: lagged income per capita, investment rate, human capital, initial level of debt, and fiscal balance.
- Main empirical findings:
  - Nonlinear relationship: low levels of public debt positively associated with productivity; debt above "21 percent of GDP" has a significantly negative effect.
  - Quantified effect: a doubling of total public debt is associated with an average reduction in productivity growth of about "1.5 percentage point."
  - Robust across estimation methods (OLS, Fixed Effects, Random Effects, GMM Differenced, GMM System).
- Selected regression coefficients (Table 4, 1990–2000; coefficient (t-statistic)):
  - Log (Public Debt): OLS = "5.40" (1.80)*; Fixed Effects = "5.25" (1.42); Random Effects = "5.76" (1.75)*; GMM Differenced = "1.20" (0.35); GMM System = "10.57" (2.99)***
  - Log (Public Debt) if Public Debt > "21 percent of GDP": OLS = "-8.53" (2.47)**; Fixed Effects = "-10.64" (2.27)**; Random Effects = "-9.33" (2.48)**; GMM Differenced = "-9.86" (1.98)**; GMM System = "-15.58" (3.23)***
  - Log(Investment): coefficients include "5.89" (4.36)***; "9.98" (4.61)***; "8.51" (5.11)***; "11.93" (2.53)**; "10.47" (5.72)***
  - Fiscal Balance: coefficients include "0.41" (6.78)***; "0.04" (3.28)***; "0.05" (3.87)***; "0.02" (1.10); "0.04" (6.13)***
  - Observations = "383"
  - R-Squared examples: OLS = "0.2822"; Fixed Effects = "0.2242"; Random Effects = "0.2741"
  - Memorandum: Coefficient for High Debt reported as "-3.13", "-5.39", "-3.57", "-8.66", "-5.01" across methods.

### Sectoral evidence for Jamaica
- Investment allocation and sectoral outcomes:
  - Investment concentrated in sectors providing safe and rapid returns (tourism, mining), with manufacturing declining.
  - Sectoral trends (1990–2000):
    - Manufacturing share declined from "21 percent of GDP in 1990" to "16 percent in 2000".
    - Share of tourism increased from "2.1 percent in 1990" to "9.2 percent in 2000".
  - Contributions to growth (1990s averages):
    - Mining: contributed "0.7" percentage points of GDP.
    - Tourism: contributed "2.6" percentage points of GDP.
    - Manufacturing contribution: "-2.2" percentage points of GDP (average).
    - Construction contribution: "-1.4" percentage points of GDP (average).
- Credit allocation (commercial banks, end-2004; percent of total loans):
  - Government services: "26.3 percent" (up from "5.5 percent" in 1993).
  - Personal loans: "29.5 percent".
  - Tourism: "13.9 percent" (up from "8.5 percent" in 1993).
  - Manufacturing: "3.3 percent" (down from "12.8 percent" in 1993).
  - Construction & Land Development: "5.2 percent" (down from "14.1 percent" in 1993).
  - Close to "70 percent" of total credit dedicated to government services, personal loans, and tourism.
- Public investment decline:
  - Public investment accounted for "4.7 percent" of total investment and "1.5 percent of GDP" in 2004.
  - Public investment accounted for "27 percent" of total investment and "6 percent of GDP" in 1988.

### Implications of mismeasurement and low-productivity dynamics
- Informal activity and growth prospects:
  - Informal sector concentrated in low value-added sectors with limited productivity potential:
    - Informal employment distribution (2001): agriculture, forestry and fishing = "48 percent"; wholesale, retail, hotels and restaurants = "26.4 percent"; community, social and personal services = "13.4 percent".
  - Own account and unpaid workers accounted for about "41 percent" of total employed labor force in 2001.
  - Wage differential: for every "J$1.00" earned in the informal market, a formal worker earns "J$1.5".
  - Informal firms face financing and legal constraints limiting growth prospects.
- Adjusted productivity:
  - Bartelsman (2002) estimate of average TFP growth for 1991–2000 = "-0.8 percent."
  - Adjusted productivity estimates remain low despite corrections for informal sector and capital depreciation.
- Channels by which high public debt affects productivity:
  - Debt overhang raises discount rates, favoring short-term over long-term investments.
  - Uncertainty reduces incentives for long-term investments (human capital, technology).
  - Liquidity constraints skew investments toward fast-return projects.
  - High debt absorbs public resources, reducing public capital expenditure and complementary externalities.

### Policy implications and priorities
- Main priorities:
  - Reduce the debt burden and ensure macroeconomic stability to increase the productivity of investment.
  - Improve the investment climate to channel investments into productive and nontraditional sectors.
  - State provision of adequate physical and social infrastructure and a sound regulatory framework to attract, retain, and increase the efficiency and productivity of private investment.
- Risks to sustainable development to address:
  - (i) Lack of diversification increases vulnerability to exogenous shocks.
  - (ii) Growth of “economic enclaves” with little spillovers and externalities to the rest of the economy.
  - (iii) Expansion of low-productivity sectors, notably the informal economy.

*Source: IMF working paper content unit _wp06235 - "Selected Growth Trends, 1980–2004."*

### 1.     Selected Growth Trends, 1980–2004................................................................................

### _wp06235 - 1.     Selected Growth Trends, 1980–2004................................................................................

### Introduction: puzzle and central hypothesis
- Real GDP grew, on average, by 1.6 percent a year from 1980 to 2004, while investment rose from 15 percent of GDP to 33 percent over the same period.
- The paper identifies two major measurement issues that partially explain the “high investment–low growth” puzzle:
  - Official estimates likely understate GDP growth because the informal economy has grown rapidly and is not fully captured in statistics.
  - The capital stock appears overestimated (low capacity utilization, hurricane damage, and nonproductive investment suggest measured capital may overstate productive capital).
- Central hypothesis: high public debt adversely affects productivity and the allocation of investment by:
  - Distorting allocation of investment toward less productive areas.
  - Reducing scope for public sector investment that complements private investment.
- Cross-country evidence: using total public debt (external and domestic) as the key explanatory variable, a doubling of total public debt leads to a reduction in productivity growth of about 1.5 percentage points.

### The paradox: low growth despite favorable endowments and high investment
- Jamaica’s endowments and potential:
  - Natural beauty, mining resources, well-educated English-speaking workforce, proximity to North America.
- Actual outcomes:
  - Growth underperformed other Caribbean countries and emerging economies.
  - GDP per capita (constant 2000 US$) over 1993–2003 remained unchanged at about US$3,150, compared with 1970 GDP per capita of US$3,328.
- Public debt and macroeconomic environment:
  - Public debt ballooned during the 1990s and has averaged above 110 percent of GDP for the last decade.
  - Total public debt was 144 percent of GDP during the FY 2003/04.
  - Domestic debt was 67 percent of GDP; external debt was 77 percent of GDP.
  - Macroeconomic uncertainty, high unemployment, emigration, and rising crime likely dampened output growth.

### Constraints to growth identified
- Small size and vulnerability to external events and natural disasters increase volatility and impair growth.
- Competitiveness and labor costs:
  - Real effective exchange rate appreciated by more than 35 percent from 1990 to 2001, leading to a decline in external competitiveness.
  - World Bank (2003) reports a 50 percent decline in Jamaica’s market share of world merchandise exports from 1994 to 2001.
  - From 1994 to 2001, unit labor costs increased twice as fast in Jamaica as in its major trading partners without comparable productivity gains.
  - Planning Institute of Jamaica (PIOJ) data indicate salary earnings almost doubled between 2005 and 2000 when employment fell by over 2 percent.
  - Household surveys indicate about 80 percent of tertiary graduates emigrated from Jamaica in the 1990s.
- Access to finance:
  - Ratio of net government debt to bank deposits was 45 percent in June 2003 (not including 10 percent for public enterprise debt).
  - High public debt pushed up interest rates and crowded out credit to the private sector.
  - Structural constraints: outdated insolvency procedures, absence of credit registries, dissatisfaction with courts and long litigation delays.
- Crime:
  - Jamaica had one of the highest rates of intentional homicide in the world in 2001 (44 per 100,000 inhabitants).
  - World Bank estimated annual cost of crime in Jamaica at around 5 percent of GDP (including investment in fighting crime and production losses).
  - Government spending on crime control estimated at 3.1 percent of GDP.
  - Lost production and health care expenses due to violent crimes cost the economy 0.6 percent of GDP.
  - Private expenditure on security estimated at 1.3 percent of GDP.
  - World Bank survey: 42 percent of business managers considered themselves likely to be murdered at the workplace.

### Declining productivity as the core driver of low growth
- Growth decomposition (1960–2000 and 1990–2000 comparisons) shows:
  - Capital accumulation contributed positively to growth, but a steady decline in productivity dominated poor performance.
  - TFP enters as a residual in growth accounting; mismeasurement of output and capital stock may affect it, but trends show clear productivity decline.
- Measured TFP performance:
  - Jamaica experienced a decline in productivity of 0.5 percent a year from 1960 to 2000.
  - Latin America recorded a 0.2 percent increase in productivity for the same period.
  - World average productivity increased by 0.9 percent for the same period.
  - Output growth in Jamaica was below Latin America and a sample of 84 countries by almost 2.5 percentage points during both 1960–2000 and 1990–2000.
  - Human capital accumulation (proxied by education) was in line with Latin American and world averages; physical capital slightly lower than those averages.
  - The large differential (as high as 2.5 percent during the 1990s) in TFP growth between Jamaica and the rest of the world is the principal factor behind Jamaica’s underperformance.

### The mismeasurement story: informal economy and capital stock concerns
- Measurement challenges:
  - Services account for more than 70 percent of output and are difficult to measure, especially with income from tourism and services potentially recorded abroad.
  - Discrepancies between electricity consumption growth and reported real GDP growth suggest underestimation of real GDP growth.
- Informal economy:
  - The informal sector accounted for about 35 to 40 percent of total GDP in 2000–01 (IADB, 2002 using 2001 Survey of Living Conditions).
  - The share of the informal economy in total output grew continuously from 1991 to 2000.
- Capital stock measurement issues:
  - Low capacity utilization, hurricane damages, and nonproductive investment imply the measured capital stock may be overstated relative to productive capital.

### Cross-country evidence and channels linking debt to productivity
- Key empirical finding:
  - Using a cross-country database and total public debt (external and domestic), the paper finds a significant and negative relationship between total public debt and productivity.
  - Quantified effect: a doubling of total public debt leads to a reduction in productivity growth of about 1.5 percentage points.
- Proposed channels in the Jamaican context:
  - High public debt associated with macroeconomic uncertainty and an output structure concentrated in maturing sectors (tourism and mining), with a declining manufacturing sector and an expanding informal economy.
  - Public investment has been crowded out by debt service, adversely affecting productivity through reduced complementary public inputs.

### Structure of the paper (organization)
- Section II: summarizes key stylized facts.
- Section III: addresses mismeasurement issues.
- Section IV: estimates, using cross-country data, the impact of public debt on productivity.
- Section V: presents sectoral evidence from Jamaica for the high debt–low productivity growth hypothesis.
- Section VI: concludes and takes stock.

*Source: IMF working paper content unit _wp06235 - "Selected Growth Trends, 1980–2004."*

### 12.9 percent of GDP in 1991 to about 40 percent in 2000. This increase in the size of the

### _wp06235 - 12.9 percent of GDP in 1991 to about 40 percent in 2000. This increase in the size of the

### Informal economy: size, trends, and corrected GDP growth
- Informal economy share rose from "12.9 percent of GDP in 1991 to about 40 percent in 2000."
- Correcting for underestimation of the informal economy, average GDP growth during the 1991-2000 period may be estimated as high as "3.0 percent," or "2.7 percent" higher than officially recorded.
- Jamaica: Size of the Informal Sector, 2000-01 (selected measures reported):
  - Monetary approach: Share over registered GDP, 2000 = "39.1"
  - Monetary approach: Share over registered GDP (predicted), 2001 = "43.7"
  - Monetary approach: Share over total GDP, 2000 = "28.1"
  - Electricity Consumption Method: Share over total GDP, 2000 = "45.5"
  - Electricity Consumption Method: Share over total GDP (conservative scenario), 2000 = "40.9"
  - Method of Additions: Share over registered GDP, 2001 = "43.5"
  - Method of Additions: Share over registered GDP (conservative scenario), 2001 = "42.9"
- Table 2 (1991–2000, constant prices) highlights:
  - Formal GDP (selected years): 1991 = "19,098"; 2000 = "19,603"
  - Informal GDP (selected years): 1991 = "3,344"; 2000 = "12,586"
  - Total GDP (selected years): 1991 = "21,327"; 2000 = "27,994"
  - Corrected/Adjusted formal GDP (accounting for informal economy capture assumptions) shown for each year in table.

### Measurement issues and productive investment
- Evidence points toward an overestimation of the capital stock, adding to mismeasurement:
  - Capacity utilization informal indications: could be as low as "50–60percent" in the manufacturing sector.
  - Capital stock depreciation understated because of hurricane damage. PIOJ estimates of hurricane damages:
    - Hurricane Gilbert (1988) = "65 percent of GDP"
    - Hurricane Ivan (2004) = "8 percent"
    - Hurricanes Dennis and Emily (2005) = "1 percent"
    - Other events: Michelle (2001) = "1 percent"; hurricane in 1980 = "2 percent"; floods in 1986 = "3 percent"; floods in 1991 = "6 percent"
  - High investment may partly reflect substitution of capital for labor due to high real wages and labor rigidities and real exchange rate appreciation in the 1990s.
  - Crime-prevention-related expenditure estimated by World Bank (2004) up to "4.4 percent of GDP" ("3.1 percent" public sector; "1.3 percent" private sector).

### Composition of fixed capital formation (Table 3, 1994-2004 highlights)
- Building construction share examples:
  - Building construction (as percent of total investment): 1995 = "45.6"; 2004 = "40.3"
- Transport equipment components (selected):
  - Transport equipment total: 1995 = "12.4"; 2004 = "7.1"
  - Motor cars: 1995 = "2.1"; 2004 = "1.4"
  - Trucks and buses: 1995 = "7.1"; 2004 = "2.4"
  - Other transport equipment: 1995 = "3.2"; 2004 = "3.3"
- Other machinery and equipment: 1995 = "37.9"; 2004 = "48.8"
- Industrial machinery and equipment: 1995 = "12.7"; 2004 = "9.7"
- Other machinery and other capital goods: 1995 = "24.0"; 2004 = "38.4"
- Total (each year) = "100.0" (composition percentages sum to 100.0 across years in table).

### Implications of mismeasurement for growth prospects
- Informal economy growth does not substitute for formal-sector solutions:
  - Financing and legal constraints limit growth prospects of informal firms (limited access to finance and legal enforcement).
  - Sizeable activity outside the tax net complicates fiscal policy.
  - Informal activities concentrated in low value-added sectors with limited productivity potential:
    - Informal employment distribution (2001): agriculture, forestry and fishing = "48 percent"; wholesale, retail, hotels and restaurants = "26.4 percent"; community, social and personal services = "13.4 percent".
    - Formal employment more concentrated in manufacturing, construction, and financial services.
  - Informal sector affects human capital quality: own account and unpaid workers accounted for about "41 percent" of total employed labor force in 2001.
  - Wage differential: for every "J$1.00" earned in the informal market, a formal worker earns "J$1.5".

### Low productivity: role of debt
- Adjusted productivity estimates remain low despite correcting for informal sector and capital depreciation:
  - Bartelsman (2002) estimate of average TFP growth for 1991–2000 = "-0.8 percent."
- Debt literature summary:
  - Nonlinear relationship between external debt and growth; low debt can be beneficial, high debt reduces growth.
  - Pattillo, Poirson and Ricci (2004) find average negative impact for net present value of debt above "160–170 percent of exports" and "35–40 percent of GDP."
  - A doubling of external debt reduces by almost "1 percentage point" both growth in per capita physical capital and growth in total factor productivity (Pattillo et al., 2004).

### Channels by which high public debt affects productivity
- Mechanisms listed:
  - Debt overhang raises discount rates, favoring short-term over long-term (potentially more productive) investments.
  - Uncertainty reduces incentives for long-term investments (human capital, technology).
  - Liquidity constraints skew investments toward fast-return projects.
  - High debt absorbs public resources, reducing public capital expenditure and lost externalities from public investment.
  - Governments may avoid difficult reforms if benefits partly accrue to creditors, reducing investment efficiency.

### Cross-country empirical findings and relevance for Jamaica
- Total public debt (including domestic) is the relevant variable for Jamaica; Jamaica appears as an outlier if only external debt is considered but fits the relationship when total public debt is used.
- Empirical model specification:
  - Dependent variable: total factor productivity (TFP) growth.
  - Main explanatory variable: log of total public debt, with a spline allowing structural break at a debt threshold.
  - Controls: lagged income per capita, investment rate, human capital, initial level of debt (all in logs), and fiscal balance.
  - Public debt threshold determined at "21 percent of GDP."
- Main empirical result:
  - Nonlinear relationship confirmed: low levels of public debt positively associated with productivity; debt above "21 percent of GDP" has a significantly negative effect.
  - A doubling of total public debt is associated with an average reduction in productivity growth of about "1.5 percentage point." Result robust across estimation methods.

### Regression/table highlights (Table 4: Nonlinear Effects of Public Debt on Productivity, 1990-2000)
- Estimated coefficients (with t-statistics in parentheses):
  - Log (Public Debt): OLS = "5.40" (1.80)*; Fixed Effects = "5.25" (1.42); Random Effects = "5.76" (1.75)*; GMM Differenced = "1.20" (0.35); GMM System = "10.57" (2.99)***
  - Log (Public Debt) if Public Debt > "21 percent of GDP": OLS = "-8.53" (2.47)**; Fixed Effects = "-10.64" (2.27)**; Random Effects = "-9.33" (2.48)**; GMM Differenced = "-9.86" (1.98)**; GMM System = "-15.58" (3.23)***
  - Log(Per Capita GDP): range across methods includes "-0.16" (0.64) to "0.53" (1.77)*
  - Population Growth: coefficients include "-0.59" (2.70)**; "-0.58" (2.18)**; "-0.58" (2.53)**; "-0.40" (4.41)***; "-0.34" (1.57)
  - Log (Investment): coefficients include "5.89" (4.36)***; "9.98" (4.61)***; "8.51" (5.11)***; "11.93" (2.53)**; "10.47" (5.72)***
  - Fiscal Balance: coefficients include "0.41" (6.78)***; "0.04" (3.28)***; "0.05" (3.87)***; "0.02" (1.10); "0.04" (6.13)***
- Observations = "383" (across main specifications)
- R-Squared examples: OLS = "0.2822"; Fixed Effects = "0.2242"; Random Effects = "0.2741"
- Memorandum item: Coefficient for High Debt reported as "-3.13", "-5.39", "-3.57", "-8.66", "-5.01" across estimation methods.

*Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06235.pdf*

### Section IV) might have been fully at play. Investments in Jamaica have been concentrated in

### _wp06235 - Section IV) might have been fully at play. Investments in Jamaica have been concentrated in

### Sectoral Evidence
- High public debt can increase uncertainty and reduce externalities via low public investment, distorting investment allocation and reducing efficiency.
- Investment concentrated in sectors providing safe and rapid returns, at the expense of diversification and development of higher-yielding, higher-risk, longer-term sectors.
- Evidence suggests continued concentration of the formal economy in a limited number of sectors—the “enclave” tourism and mining industries—and fast growth of the informal sector.
- Manufacturing declined dramatically, notably due to the exit of some textile and tire manufacturing companies.
- Growth positive in a limited number of sectors:
  - Nontradable sectors: communications and power.
  - Location- and natural-resource based activities: transport, mining, and tourism.
  - Tourism and bauxite mining are shielded from Jamaica-specific risks and have earnings in foreign exchange.
- Data limitations: difficult to isolate public debt impact from other factors (policy choices, small island economy constraints); sectoral data sparse; need improved data collection and breakdown of investment and productivity by sector.
- Sectoral trends, 1990–2000:
  - Manufacturing share declined from 21 percent of GDP in 1990 to 16 percent in 2000.
  - Share of tourism increased from 2.1 percent in 1990 to 9.2 percent in 2000.
  - Construction declined slightly; other sectors broadly constant.
- Contributions to growth (1990s averages):
  - Mining and tourism contributed positively: 0.7 and 2.6 percentage points of GDP, respectively.
  - Manufacturing contribution: -2.2 percentage points of GDP (average).
  - Construction contribution: -1.4 percentage points of GDP (average).
- Commercial banks’ loans and advances by sector (end-2004; in percent of total loans):
  - Government services: 26.3 percent (up from 5.5 percent in 1993).
  - Personal loans: 29.5 percent.
  - Tourism: 13.9 percent (up from 8.5 percent in 1993).
  - Manufacturing: 3.3 percent (down from 12.8 percent in 1993).
  - Construction & Land Development: 5.2 percent (down from 14.1 percent in 1993).
  - Close to 70 percent of total credit dedicated to government services, personal loans, and tourism.
- Interpretation:
  - Scarce availability of bank credit for the private sector.
  - High levels of investment likely financed through retained earnings and foreign financing; retained earnings available mainly to well-established, cash-flow generating firms; foreign financing directed toward activities shielded from Jamaica-specific risks.

### Low Public Investment and Low Productivity
- High public debt has crowded out public investment in the government budget, which declined to negligible levels of GDP and total investment.
- Public investment metrics:
  - Public investment accounted for 4.7 percent of total investment and 1.5 percent of GDP in 2004.
  - Public investment accounted for 27 percent of total investment and 6 percent of GDP in 1988.
- Jamaica exhibits high overall investment rates but low public investment.
- Role of public investment:
  - Provides complementary public infrastructure and public and social services that raise the efficiency of private investment and overall productivity.
- Comparative evidence (1990–2000 averages):
  - High overall investment rates are insufficient for high economic growth if public investment is lacking.
  - Jamaica underperformed other emerging economies with similar levels of total investment and domestic private investment.
  - Jamaica’s growth performance is only slightly inferior to the average growth for countries with similar levels of public investment.
- Figures and sources:
  - Public investment in percent of total investment, 1988–2005 (WEO Database, February 2006) shows a substantial decline.
  - Charts referenced: Total Domestic Investment (in percent of GDP) and Real GDP Growth, 1990–2000; Private investment (exc. FDI) and Real GDP Growth, 1990–2000; Public Investment (in percent of GDP) and real GDP growth, 1990–2000.

### Taking Stock: Summary and Conclusions
- Measurement caveats:
  - Jamaica’s ‘high investment-low growth’ puzzle partly due to measurement problems; informal economy share likely grew substantially during the 1990s.
  - Investment figures may overestimate productive investments; capital depreciation may be understated (e.g., hurricane-related damage not fully accounted for).
  - Considerable investment may go to nonproductive activities, such as crime prevention and residential housing.
- Empirical association:
  - Panel data analysis for a sample of 35 emerging economies shows low productivity robustly associated with high levels of public debt.
  - A doubling in total public debt is estimated to result in a 1.5 percentage point reduction in productivity growth.
- Mechanisms:
  - High public debt distorts capital allocation by increasing uncertainty and reducing externalities from public investment.
- Observed patterns in Jamaica:
  - Continued concentration of the formal economy in a limited number of sectors consistent with debt affecting productivity and growth.
  - Growing “enclave” tourism and mining industries shielded from many Jamaica-specific risks.
  - Rapid growth of the informal sector, which is partially out of reach of the state and less affected by formal sector developments.
- Risks to sustainable development:
  - (i) Lack of diversification increases vulnerability to exogenous shocks.
  - (ii) Growth of “economic enclaves” with little spillovers and externalities to the rest of the economy.
  - (iii) Expansion of low-productivity sectors, notably the informal economy.
- Policy implications and priorities:
  - Increase productivity of investment by addressing the debt burden and ensuring macroeconomic stability.
  - Improve the investment climate to channel investments into productive and nontraditional sectors.
  - State provision of adequate physical and social infrastructure complements a sound regulatory framework to attract, retain, and increase the efficiency and productivity of private investment.

*Source: IMF working paper content provided in the supplied PDF extract.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp06235.pdf_