Abstract
The objective of this paper is to analyze the growth performance of the ECCU countries since independence and the policy challenges they face to ensure sustained growth in the period ahead. Although tourism specialization may bring about higher growth, it could also increase volatility in growth by amplifying the impact of business cycles in source countries on the tourism sector. Low productivity growth is principally the reason for the slowdown in growth. High debt levels have been a major drag on growth.
I. Growth in the ECCU: What Went Wrong and Can it be Fixed1
A. Introduction
1. Growth in the ECCU countries has been on a declining trend since the 1990s, with the current global slowdown exacerbating this trend. The ECCU region has been buffeted by a series of adverse exogenous shocks over time, including the erosion of trade preferences; the decline in official foreign assistance; recessions in the developed countries, the main source of tourism and FDI for the region; and frequent natural disasters. The recent global slowdown has exacerbated the already declining trend in growth. Growth has declined from an average of 6 percent in the 1980s to just over 2 percent since 2000, with most ECCU countries reporting negative growth in 2008-09. At the same time, the relaxation of the fiscal stance, partly reflecting accommodation to these shocks, has led to a rapid build-up of public debt in the region.
2. The objective of this chapter is to analyze the growth performance of the ECCU countries since independence and the policy challenges they face to ensure sustained growth in the period ahead.2 Using both a growth accounting framework and regression analysis, we investigate the role tourism has played and the extent to which tourism-led growth remains a viable strategy for future. We also focus on the extent to which high and rising debt levels have hindered growth.
3. The chapter is organized as follows. Section B presents some stylized facts about growth in the ECCU since the 1970s. Section C uses the growth accounting framework to assess the extent to which growth has been driven by factor accumulation versus gains in total factor productivity (TFP). Section D reviews the determinants of long-term growth using regression analysis. In addition to the standard variables considered in the growth literature, we augment an otherwise empirical growth equation by adding tourism proxies to determine its importance in explaining growth. Section E looks at the impact of high debt levels on growth in the ECCU countries. Section F draws some policy conclusions based on the results of the study.
B. ECCU’s Growth Performance in an International Perspective
4. After growing faster than the rest of the world in the early years after independence, the ECCU countries have experienced a significant decline in growth since the 1990s. Growth averaged about 4¼ percent in the ECCU countries during 1970-2009, compared with 3.8 percent for small island (SI) countries and 3.4 percent in the other Caribbean countries in the region.3 However, there is a marked slowdown since the 1990s, reflecting largely the structural shifts in production caused by the dismantling of trade preferences with Europe (largely for bananas and sugar), a decline in aid inflow and exogenous shocks. At the same time, countries which made an early switch to tourism, for example, Antigua and Barbuda and St. Kitts and Nevis, were able to offset, in part, the impact of the decline in the agriculture sector. In addition, the development of the offshore financial sector in the late 1990s seems to have contributed positively to growth, most notably in St. Kitts and Nevis, Antigua and Barbuda, and St Vincent and the Grenadines. More recently, the global financial and economic crisis has led to a sharp reduction in tourism, remittances and FDI, causing a severe negative effect on growth in the region.
5. The pace of growth in real per capita GDP has also slowed since the 1990s, falling below the world average. While the per capita income level has increased almost four-fold in the ECCU countries, and the average per capita GDP of the ECCU countries is higher than the average for all Small Islands (SI) and above the average for emerging and developing countries, the drop in the growth rate of per capita GDP since the 1990s has been especially sharp compared to developing and emerging market economies. Within the ECCU group, Grenada and Dominica are at the lower end while Antigua and Barbuda and St. Kitts and Nevis have the highest per capita GDP.
ECCU: Per capita GDP (PPP), 1970-2009
(In percent)
Citation: IMF Staff Country Reports 2011, 032; 10.5089/9781455213894.002.A001
Sources: WEO, Fund staff calculations.ECCU: Per capita GDP (PPP) Relative to the World, 1970-2009
(Ratios)
Citation: IMF Staff Country Reports 2011, 032; 10.5089/9781455213894.002.A001
Sources: WEO, Fund staff calculations.6. Volatility of output growth in the ECCU has been considerably lower than for the comparator group of other SI countries and for emerging and developing countries. This is perhaps surprising because of the openness of their economies and their exposure to natural disasters. The same picture holds irrespective of whether volatility is measured as the standard deviation of GDP growth, or is based on the frequency of growth crashes (events of large drops in activity). With these stylized facts in mind, we analyze in the rest of the chapter the role of productivity and factor accumulation in the growth performance of the region, whether tourism is an activity that has traction for growth and the implications of high indebtedness for growth.
Volatility of GDP Growth
(Based on real GDP growth rates (PPP) 1971-2009)
Share of years (in percent) with growth lower than -5.1 percent (which corresponds to the 5th percentile of all country/years growth).
Bahamas, Barbados, Belize, Guyana, Jamaica, Suriname, Trinidad and Tobago.
Volatility of GDP Growth
(Based on real GDP growth rates (PPP) 1971-2009)
| Countries | Countries | Average Growth | Average Std. Dev. | Frequency of Growth Crashes 1/ | ||
|---|---|---|---|---|---|---|
| Caribbean (simple average) | 13 | 3.4 | 4.7 | 4.5 | ||
| Caribbean | 13 | 2.2 | 4.7 | 4.5 | ||
| ECCU | 6 | 4.3 | 4.2 | 3.0 | ||
| Antigua and Barbuda | 1 | 5.0 | 3.7 | 2.6 | ||
| Dominica | 1 | 3.2 | 5.1 | 5.1 | ||
| Grenada | 1 | 3.9 | 4.5 | 5.1 | ||
| St. Kitts and Nevis | 1 | 4.3 | 2.9 | 2.6 | ||
| St. Lucia | 1 | 4.4 | 4.8 | 2.6 | ||
| St. Vincent & the Grens. | 1 | 4.4 | 3.5 | 0.0 | ||
| Non ECCU Caribbean 2/ | 7 | 2.1 | 5.0 | 5.9 | ||
| Non Caribbean Emerging and Developing Economies | 136 | 5.1 | 6.8 | 6.2 | ||
| Non Caribbean Small Islands | 18 | 4.3 | 5.6 | 4.2 | ||
Share of years (in percent) with growth lower than -5.1 percent (which corresponds to the 5th percentile of all country/years growth).
Bahamas, Barbados, Belize, Guyana, Jamaica, Suriname, Trinidad and Tobago.
Volatility of GDP Growth
(Based on real GDP growth rates (PPP) 1971-2009)
| Countries | Countries | Average Growth | Average Std. Dev. | Frequency of Growth Crashes 1/ | ||
|---|---|---|---|---|---|---|
| Caribbean (simple average) | 13 | 3.4 | 4.7 | 4.5 | ||
| Caribbean | 13 | 2.2 | 4.7 | 4.5 | ||
| ECCU | 6 | 4.3 | 4.2 | 3.0 | ||
| Antigua and Barbuda | 1 | 5.0 | 3.7 | 2.6 | ||
| Dominica | 1 | 3.2 | 5.1 | 5.1 | ||
| Grenada | 1 | 3.9 | 4.5 | 5.1 | ||
| St. Kitts and Nevis | 1 | 4.3 | 2.9 | 2.6 | ||
| St. Lucia | 1 | 4.4 | 4.8 | 2.6 | ||
| St. Vincent & the Grens. | 1 | 4.4 | 3.5 | 0.0 | ||
| Non ECCU Caribbean 2/ | 7 | 2.1 | 5.0 | 5.9 | ||
| Non Caribbean Emerging and Developing Economies | 136 | 5.1 | 6.8 | 6.2 | ||
| Non Caribbean Small Islands | 18 | 4.3 | 5.6 | 4.2 | ||
Share of years (in percent) with growth lower than -5.1 percent (which corresponds to the 5th percentile of all country/years growth).
Bahamas, Barbados, Belize, Guyana, Jamaica, Suriname, Trinidad and Tobago.
C. Explaining the Declining Trend in Growth in the ECCU
7. A growth accounting exercise reveals that total factor productivity (TFP) explains the bulk of the variation in economic growth in the ECCU during the last forty years.4 Capital accumulation and TFP gains account for 80-90 percent of the growth in ECCU countries (Figure 1 and Table 1 in Appendix 2). TFP growth was the strongest in the 1970s in Dominica, St Kitts and Nevis and St. Lucia. In Dominica, TFP growth has since declined and in recent years, growth in TFP has been negative. Grenada has also experienced declining productivity levels in recent years. As Barro (1998) suggests, negative TFP growth is hard to interpret because it implies “a technical regress”, reflecting a drop in the efficiency with which the other factors of production are used either because other complementary factors have changed or due to bad policies and weak institutions. That said, some countries in the ECCU - Antigua and Barbuda, St. Kitts and Nevis and St Vincent and the Grenadines have been able to reverse the declining trend in productivity, and TFP has risen since 2000. While difficult to pinpoint what explains this improvement in productivity, all three countries have tried to find a niche for their products, mainly in high-end tourism services, which may explain the improvement in productivity (see below).
ECCU: Contibutions to Growth, 1970-2007
(In percent, adjusted for the effect of hurricanes on capital)
Citation: IMF Staff Country Reports 2011, 032; 10.5089/9781455213894.002.A001
Source: Heston, Summers and Aten (2009). Emergency Disaster Database (EM-DAT), CRED (2010). Fund staff calculations.Note: The data for St. Lucia omits EM-DAT’s entry of US$1 billion damages in 1988, which is erroneous.8. Compared to Barbados5, the ECCU countries have been catching up in terms of output per worker and productivity since the seventies. As revealed by the figure, output per worker would increase in all the ECCU countries assuming they achieved Barbados productivity level or capital per worker level. Most of the ECCU countries should concentrate on increasing productivity as this would yield a higher output per worker; especially for Grenada and Dominica. The only exception is Antigua and Barbuda, whose productivity was slightly above Barbados in 2007, implying that it could have grown faster than Barbados if it had had the same level of capital accumulation. For the rest of the ECCU countries output would also increase if capital per worker were to reach Barbados levels, although the gains would be lower than if they focused toward raising productivity.
ECCU: Output per worker and its components (ratios to Barbados), 2007
(In percent, adjusted for the effects of hurricanes on capital)
Citation: IMF Staff Country Reports 2011, 032; 10.5089/9781455213894.002.A001
Source: Heston, Summers and Aten (2009). Emergency Disaster Database (EM-DAT), CRED (2010). Fund staff calculations.Note: The data for St. Lucia omits EM-DAT’s entry of US$1 billion damages in 1988, which is erroneous.D. Tourism’s Contribution to Growth in the ECCU
9. This section uses the standard growth model, augmented by various measures of tourism, to investigate if the latter played a statistically significant role in determining growth. Panel regression using cross-country data from 154 countries covering a period of 29 years is used to investigate the determinants of long-term growth.6 The basic regression model is specified as follows:
Δyit = α + β1yit-1 + β2xit + β3tourismit + ηi + εit
where, Δyit is the average growth rate of country i ‘s GDP per capita between time t -1 and t (five-year averages are used7), α is a constant, yit-1 is the logarithm of initial GDP per capita, which controls for income convergence, xit is a vector of standard determinant of economic growth, tourismit is a measure of tourism, ηi are unobserved country specific effects, and εit is a time and country specific disturbance. Following the growth literature (Arezki et al 2009), xit8 includes human capital measured by primary education, openness of the economy (measured as exports and imports as a ratio of GDP), size of government consumption (the ratio of government spending on goods and services to GDP), the investment-output ratio, the inflation rate, terms of trade, and life expectancy. A set of dummy variables are also used to capture the effect of being a small island.9
10. The results confirm that tourism has been a positive contributor to economic growth (Appendix 2 Table 3).10 An expansion of the tourism sector, or more generally the service sector, can theoretically have a negative impact on growth by crowding out the production of the tradable goods sector through a shift of resources to a less productive sector or increase of relative price of nontradable goods (Dutch disease effect). However, our empirical analysis shows otherwise. As can be seen in Column 2, in which we augmented the standard growth model with tourist arrivals per population, tourism has been a significant factor in long-term economic growth. It implies that a 10 percent increase in tourist arrivals per capita raises economic growth by about 0.2 percent. Column 6 shows that not only the volume of tourism but also the quality and value added of tourism, proxied by receipts per tourist, are significant factors driving economic growth, justifying some Caribbean countries strategy of focusing on high-end tourism demand. Another interesting result is that there seems to be no adverse effect from attracting too many tourists, as reflected by the positive sign of the square of the tourism variable in Column 7.
11. Moreover, tourism has been the most significant contributor of long-term economic growth for the ECCU and helped to more than offset the negative impacts of geography and “being small”. To quantify the contribution of each factor to economic growth of the ECCU, we calculate the predicted growth rate of the model using the coefficients in Column 7 of Table 3 (see Appendix 2) and the average values of each explanatory variable for the ECCU. We then calculate the contribution of each factor by replacing its value with the average value for the World. The accompanying figure presents the results, where the green arrows represent the variables that contributed positively to long-term growth, while the red arrows indicate the variables that negatively affected long-term growth. It is evident that tourism played the most significant role in the growth of the ECCU over the period 1970-2007. Tourism arrivals to the ECCU countries have been higher than the world average; this has added 5.0 percentage points to growth in the ECCU. At the same time, growth in the ECCU has been lower by 0.3 percentage points and 3.1 percentage points on average because of being “small” (the average size of an ECCU country is 448.5 km2 compared to the world average of 825,644 km2) and given the island geographical nature of the ECCU, respectively. The relatively high investment ratios (about 15–20 percent on average) and low inflation in the ECCU have also been positive contributors to growth. But, specialization in tourism has been the most advantageous and indeed has offset some of the limitations that come from being a small island economy (for example, remoteness, higher transportation costs, diseconomies of scale).
ECCU: Relative Performance vis-à-vis the World
(In percent)
Citation: IMF Staff Country Reports 2011, 032; 10.5089/9781455213894.002.A001
Sources: Fund staff calculations.12. An early transition to the tourism industry also appears to have contributed to higher long-term economic growth for countries making an early switch from agriculture to tourism. Figure 2 in Appendix 2 shows that there is a positive relationship between average real GDP growth rate during 1981–2007 and tourism concentration in 1981, where tourism concentration is measured as tourist arrivals or tourism receipts normalized by population or land size. This implies that the tourism sector has on average brought higher economic growth for those countries which developed this industry at an early stage. For example, Antigua and Barbuda and St. Kitts and Nevis, which switched from an agriculture-driven economy to a tourism-based economy early on, have on average grown at about 4½ percent during this period, compared to an average of under 3 percent for Dominica which has remained mostly an agriculture oriented economy
E. Tourism and Growth Volatility
13. While tourism specialization may bring about higher growth, it could also increase volatility in growth by amplifying the impact of business cycles in source countries on the tourism sector. The impact of the current global crisis on ECCU countries has brought to the fore the negative side effects of tourism-led growth. However, it is not clear whether this also increases the volatility of growth in the long run. In order to determine this, we try to analyze the standard deviation of growth (the standard measure of volatility) in terms of the same explanatory variables included in the models of Table 3 in Appendix 2.11
14. Tourism not only raises per capita GDP growth but also helps to reduce its volatility. Specializing in the supply of tourism services has an unambiguously positive effect for the countries that have adopted this strategy for their economic development. This is particularly so for SI countries. In fact, for these countries, tourism helps to reduce the volatility of growth by counter balancing other factors that would otherwise increase volatility. This is in line with the finding that volatility in the ECCU has not been unusually high relative to other regions, as noted earlier. For example, inflation (by injecting uncertainty in decision making) and the ratio of government spending to GDP (through its impact on long term interest rates among other things) both have a negative impact on growth and increase volatility. Also being a “small island” affects growth negatively and increases volatility. Other variables show some trade-off in their overall impacts on growth. The ratio of investment to GDP contributed positively to growth but at the same time it increases the volatility of growth, implying that any shock to investment is difficult to offset. The estimation results also indicate that countries that started with a lower level of income per capita grew faster but the higher growth was accompanied by greater volatility. Finally, as captured by the coefficient estimates on the openness variable, the beneficial effect from an open economy is outweighed by the increased volatility resulting from increased vulnerabilities to external shocks.
F. ECCU Tourism Competitiveness vis-á-vis the Caribbean
15. The Caribbean countries appear to have lost some market share in tourism arrivals. The share of stay-over visitor arrivals to the Caribbean (as percent of total visitors to the World) fell from 2.7 percent during 1990-95 to 2.2 percent during 2005–08. ECCU countries share within the Caribbean fell from 5.7 percent to 5.1 percent over the same period because of losses by Antigua and Barbuda and Grenada. The rest of the ECCU countries have either gained by small margins or maintained their shares.
ECCU: Average Market Shares in Stay-over Arrivals
(In percent)
ECCU: Average Market Shares in Stay-over Arrivals
(In percent)
| 1990-95 | 2005-08 | %change | |
|---|---|---|---|
| Share to Caribbean total: | |||
| Anguilla | 0.30 | 0.36 | 19.46 |
| Antigua and Barbuda | 1.85 | 1.33 | -28.05 |
| Dominica | 0.41 | 0.42 | 1.46 |
| Grenada | 0.76 | 0.61 | -19.33 |
| St Kitts | 0.68 | 0.70 | 2.62 |
| St Lucia | 151 | 156 | 3.10 |
| St Vincent and the Grenadines | 0.45 | 0.48 | 5.93 |
| ECCU | 5.66 | 5.09 | -10.01 |
| Share to the World: | |||
| ECCU | 0.14 | 0.11 | -21.95 |
| Caribbean | 2.67 | 2.24 | -16.30 |
ECCU: Average Market Shares in Stay-over Arrivals
(In percent)
| 1990-95 | 2005-08 | %change | |
|---|---|---|---|
| Share to Caribbean total: | |||
| Anguilla | 0.30 | 0.36 | 19.46 |
| Antigua and Barbuda | 1.85 | 1.33 | -28.05 |
| Dominica | 0.41 | 0.42 | 1.46 |
| Grenada | 0.76 | 0.61 | -19.33 |
| St Kitts | 0.68 | 0.70 | 2.62 |
| St Lucia | 151 | 156 | 3.10 |
| St Vincent and the Grenadines | 0.45 | 0.48 | 5.93 |
| ECCU | 5.66 | 5.09 | -10.01 |
| Share to the World: | |||
| ECCU | 0.14 | 0.11 | -21.95 |
| Caribbean | 2.67 | 2.24 | -16.30 |
16. Simultaneous examination of the changes in shares of tourist arrivals and shares of tourism receipt suggests that some countries may have lost market shares in arrivals because of a shift in strategy. While loss of market share in arrivals is often interpreted as a loss of competitiveness in tourism, it may also reflect a shift in strategy from mass tourism to upscale tourism, resulting in higher tourism receipts. Hence, changes in market shares in arrivals are assessed in conjunction with changes in shares of tourism receipts from 1990–95 to 2005–08. The results suggest:
Antigua and Barbuda, and Grenada have seen declines in both arrival and receipt shares—a clear indication of loss in competitiveness.
St. Kitts and Nevis and St. Lucia have recorded increases in shares of arrivals but declines in shares of receipts, which may reflect a shift towards mass tourism achieved through cost and price cutting, provided that the loss in share of receipts is temporary. Otherwise, it may reflect a loss of competitiveness as the countries are not earning income proportional to the number of visitors they host.
Anguilla, Dominica, and St. Vincent and the Grenadines have recorded increases in shares of both arrivals and receipts, indicating improvements in competitiveness.
17. Higher prices and low quality—instead of limited room supply—seem to be the main reasons for the loss in market share by ECCU countries within the Caribbean. Between the periods 1990–95 and 2005–08, the ECCU has gained shares in room supply within the Caribbean (Appendix 2 Figure 3). Nevertheless, the price charged per room is relatively high in some ECCU countries, with Antigua and Barbuda having the highest hotel prices in the region. This may explain the loss in the shares in both stay-over arrivals and tourism receipts. On the other hand, St. Kitts and Nevis and St. Lucia have lost shares in tourism receipts, even though they have gained some market share in stay-over arrivals, reflecting perhaps the lower average hotel costs. While difficult to document, anecdotal evidence indicates that enhancing the quality of services and diversifying the kinds of services would give a boost to the tourism sector.
G. Debt and Its Impact on Growth
18. High levels of public debt12 undermine economic performance by crowding out private investment and acting as a tax on future investment projects (Krugman, 1988; Sachs, 1989). Results from the Fiscal Monitor (FAD 2010) suggest that on average a 10 percent increase in initial debt reduces real per capita GDP growth by 0.2 percent per year. The accompanying figure shows that there is a negative relationship between government debt and economic growth. In a recent study on 44 advanced and emerging countries covering a period of almost 200 years, Reinhart and Rogoff (2010) find that above a threshold of debt to GDP ratio of 90 percent, median growth rates fall 1 percent while average growth falls even more. They also find that emerging markets have a lower threshold of external debt (60 percent) above which growth rates decrease by two percent or more.
World: Debt and Growth, 2000-2007
(In percent)
Citation: IMF Staff Country Reports 2011, 032; 10.5089/9781455213894.002.A001
Sources: Heston, Summers and Aten (2009). World Economic Outlook. Fund staff calculations.19. Debt has negatively and significantly affected growth in the ECCU, if one considers the cross country experience as a guide. Table 4 (in Appendix 2)13 shows the results from an extended regression model which adds the debt-to-GDP ratio to the growth model, including tourism and size, as an explanatory variable. The previous results, particularly those related to the importance of tourism, are largely the same even when controlling for government debt. The evidence shows that for debt-to-GDP ratios above 30 percent, debt reduces growth. The negative impact of debt increases when the debt-to-GDP ratio crosses the 60 percent threshold.14 Moreover, once debt is considered, investment no longer appears to have a positive impact on growth suggesting that government spending is crowding out private investment (Appendix 2 Table 4).
20. Empirical evidence suggests that, when public debt is above a certain threshold, debt reduction will help improve the economic performance of a country. The accompanying figure shows that, in general, when debt is high and countries implement policies directed to reduce its levels, GDP growth is higher than when governments allow the debt to keep growing. This seems to be more important for emerging and developing countries where the growth rate in periods of falling debt was more than 3 times the one in periods of rising debt. For the ECCU in particular the difference between business as usual or targeted policies to reduce debt could entail almost a doubling of the growth rate when government commits to cut debt.
Real per capita GDP Growth During Periods of Rising and Falling Debt
(Initial debt > 60 percent of GDP)
Citation: IMF Staff Country Reports 2011, 032; 10.5089/9781455213894.002.A001
Sources: Heston, Summers and Aten (2009). World Economic Outlook. Fund staff calculations.H. Conclusions
21. After an initial strong growth performance following independence, since the early nineties, the growth performance in most ECCU countries has been disappointing. Although the ECCU countries recorded the best growth performance overall in the Caribbean region, over the last twenty years they have lost ground in terms of per capita GDP to their small island peer countries and the fast-growing emerging and developing countries. Rather than converging toward advanced country levels, income has fallen further behind.
22. Low productivity growth is principally the reason for the slowdown in growth. The decline in TFP growth despite high levels of capital accumulation accounts for the relatively poor performance of the ECCU countries in recent years. While difficult to pinpoint the factors behind low productivity, structural reforms that encourage innovation and adoption of productivity-enhancing technology could help to reverse this trend.
23. Tourism has been a significant contributor to growth and there remains scope for further expansion. Tourism has been an important contributor to growth and there is significant scope in all of the ECCU countries to boost growth by enhancing the performance of this sector. Moreover, the ECCU has lost competitiveness among the Caribbean countries, indicating the need for structural reform measures to improve the competitiveness of this sector, both in terms of improved product quality and lower costs.
24. High debt levels have been a major drag on growth. Therefore, a key issue that needs to be addressed is the rising debt level in the region. To the extent that governments are proactive in reducing debt, growth will improve not just by reducing budgetary interest costs and creating fiscal space but more importantly through their effect on reducing long term interest rates and building confidence and investment.
Appendix 1
This appendix details the growth and level accounting framework used to assess the extent to which observed output growth has been driven by factor accumulation or TFP gains.15 The analysis of the sources of growth dates back to the 1950s with the seminal work of Solow (1957). Solow first decomposed output growth into the growth of labor, capital and a residual—referred to in the literature as the “Solow residual”—and interpreted it as a measure of the contribution of technological change to growth. In the 1960s and 1970s, Denison (1962), Jorgenson and Griliches (1967) and Denison, Jorgenson and Griliches (1972) further extended this work to include both the quantity and quality of labor and capital. More recently, Barro and Lee (1994), Lee (2005) and Loayza, Fajnzylber and Calderon (2005) extended this analysis to a cross-section of countries. However, none of these studies look at the ECCU countries.
Analytical framework. The conventional Cobb-Douglas production function with constant returns to scale is used to calculate the contribution of each factor16:
where, Y is aggregate output, A is total factor productivity, K is the physical capital stock, L is unit of labor, and a is the elasticity of output with respect to the physical capital stock. Taking logs and differentiating with respect to time gives17
The contribution of each factor is calculated as its growth rate multiplied by its share, with TFP as the residual. The Penn World Tables (PWT 6.3 of Heston et al. (2009)) are used to ensure comparability of data across countries.18 As in Lee (2005), the working age population, i.e., population between the ages of 15–64, is used as a measure of the size of the labor force. The physical capital stock is constructed using investment data from the PWT and applying the perpetual inventory method. Following the literature in this area, it is assumed that the share of capital is α = 0.35, and that it is constant and equal across countries.19 Average depreciation is assumed to be about 6 percent per year over the period. However, given that the region suffers from the recurrence of natural disasters that destroy the capital stock, data has been adjusted in years in which there were major hurricanes that inflicted considerable damage to the islands. The impact of hurricanes is obtained from the Emergency Disaster Database (EM-DAT). The current price US dollar estimates from this database are converted to 2005 constant prices to use the same units as investment and the capital stock. This information in then used in the perpetual inventory method to adjust the capital stock in every year where there is a hurricane with recorded damages.20
Another way to analyze TFP is by comparing the ratio of output per worker between two countries. For this purpose, we use the level accounting methodology of Hall and Jones (1999) which transforms equation (l) into per worker units of output and capital (denoted by small letters). For a pair of countries i and j, the ratio of output per worker is given by
Therefore, the difference in output per worker between countries can be decomposed as the difference between their capital-labor ratios and the TFP ratios. Table 2 in the Appendix and the accompanying figure present the output and capital per worker and the productivity levels of ECCU countries with respect to Barbados-the benchmark country for this exercise.
Appendix 2
ECCU: Contributions to Growth, 1970-2007
(In percent, adjusted for the effect of hurricanes on capital)
Citation: IMF Staff Country Reports 2011, 032; 10.5089/9781455213894.002.A001
Source: Heston, Summers and Aten (2009). Emergency Disaster Database (EM-DAT), CRED (2010). Fund staff calculations.Note: The data for St. Lucia omits EM-DAT’s entry of US$1 billion damages in 1988, which is erroneous.Small Islands: Tourism and Growth, 1981-2007
Citation: IMF Staff Country Reports 2011, 032; 10.5089/9781455213894.002.A001
Source: World Bank (2010); IMF; WTO; Heston, Summers and Aten (2009). Fund staff calculations.Caribbean: Tourism Competitiveness.
Citation: IMF Staff Country Reports 2011, 032; 10.5089/9781455213894.002.A001
ECCU: Growth Accounting, 1970-2007
(In percent, adjusted for the effects of hurricanes on capital)
Data available only from 1996.
Data available only from 1977.
ECCU: Growth Accounting, 1970-2007
(In percent, adjusted for the effects of hurricanes on capital)
| Average annual growth | Contribution to growth (% point) | |||||||
|---|---|---|---|---|---|---|---|---|
| Output | Capital | Labor | TFP | K | L | TFP | ||
| Antigua and Barbuda /a | 4.52 | 6.95 | 1.84 | 0.89 | 2.43 | 1.20 | 0.89 | |
| 1971-1980 | ||||||||
| 1981-1990 | ||||||||
| 1991-2000 | 3.73 | 4.90 | 1.89 | 0.79 | 1.72 | 1.23 | 0.79 | |
| 2001-2007 | 4.97 | 8.13 | 1.81 | 0.95 | 2.84 | 1.18 | 0.95 | |
| Dominica /b | 2.55 | 3.32 | 0.74 | 0.91 | 1.16 | 0.48 | 0.91 | |
| 1971-1980 | 2.93 | -15.73 | -0.20 | 8.56 | -5.50 | -0.13 | 8.56 | |
| 1981-1990 | 5.22 | 9.97 | 0.38 | 1.48 | 3.49 | 0.25 | 1.48 | |
| 1991-2000 | 1.58 | 2.23 | 1.09 | 0.09 | 0.78 | 0.71 | 0.09 | |
| 2001-2007 | -0.05 | 3.54 | 1.14 | -2.03 | 1.24 | 0.74 | -2.03 | |
| Grenada | 4.13 | 5.47 | 0.78 | 1.71 | 1.92 | 0.51 | 1.71 | |
| 1971-1980 | 3.98 | 4.56 | 1.19 | 1.61 | 1.59 | 0.77 | 1.61 | |
| 1981-1990 | 7.39 | 8.28 | 0.43 | 4.21 | 2.90 | 0.28 | 4.21 | |
| 1991-2000 | 3.70 | 5.65 | 0.37 | 1.48 | 1.98 | 0.24 | 1.48 | |
| 2001-2007 | 0.32 | 2.50 | 1.28 | -1.39 | 0.88 | 0.83 | -1.39 | |
| St. Kitts & Nevis /b | 4.88 | 5.14 | -0.18 | 3.20 | 1.80 | -0.12 | 3.20 | |
| 1971-1980 | 7.61 | 5.21 | -0.41 | 6.05 | 1.82 | -0.26 | 6.05 | |
| 1981-1990 | 5.43 | 5.60 | -0.65 | 3.90 | 1.96 | -0.42 | 3.90 | |
| 1991-2000 | 3.22 | -0.05 | -0.33 | 3.45 | -0.02 | -0.21 | 3.45 | |
| 2001-2007 | 5.28 | 11.86 | 0.78 | 0.62 | 4.15 | 0.50 | 0.62 | |
| St. Lucia | 3.99 | 6.12 | 2.28 | 0.37 | 2.14 | 1.48 | 0.37 | |
| 1971-1980 | 5.40 | 2.44 | 2.96 | 2.62 | 0.85 | 1.92 | 2.62 | |
| 1981-1990 | 4.85 | 9.93 | 2.36 | -0.16 | 3.47 | 1.53 | -0.16 | |
| 1991-2000 | 2.62 | 7.87 | 2.00 | -1.44 | 2.75 | 1.30 | -1.44 | |
| 2001-2007 | 2.72 | 3.46 | 1.60 | 0.47 | 1.21 | 1.04 | 0.47 | |
| St. Vincent & Grenadines | 4.15 | 4.09 | 1.43 | 1.79 | 1.43 | 0.93 | 1.79 | |
| 1971-1980 | 2.99 | 1.76 | 2.28 | 0.89 | 0.62 | 1.48 | 0.89 | |
| 1981-1990 | 6.85 | 5.36 | 1.82 | 3.80 | 1.88 | 1.18 | 3.80 | |
| 1991-2000 | 2.22 | 4.98 | 0.77 | -0.02 | 1.74 | 0.50 | -0.02 | |
| 2001-2007 | 4.73 | 4.36 | 0.63 | 2.79 | 1.53 | 0.41 | 2.79 | |
| Average | 3.99 | 5.00 | 1.12 | 1.52 | 1.75 | 0.73 | 1.52 | |
Data available only from 1996.
Data available only from 1977.
ECCU: Growth Accounting, 1970-2007
(In percent, adjusted for the effects of hurricanes on capital)
| Average annual growth | Contribution to growth (% point) | |||||||
|---|---|---|---|---|---|---|---|---|
| Output | Capital | Labor | TFP | K | L | TFP | ||
| Antigua and Barbuda /a | 4.52 | 6.95 | 1.84 | 0.89 | 2.43 | 1.20 | 0.89 | |
| 1971-1980 | ||||||||
| 1981-1990 | ||||||||
| 1991-2000 | 3.73 | 4.90 | 1.89 | 0.79 | 1.72 | 1.23 | 0.79 | |
| 2001-2007 | 4.97 | 8.13 | 1.81 | 0.95 | 2.84 | 1.18 | 0.95 | |
| Dominica /b | 2.55 | 3.32 | 0.74 | 0.91 | 1.16 | 0.48 | 0.91 | |
| 1971-1980 | 2.93 | -15.73 | -0.20 | 8.56 | -5.50 | -0.13 | 8.56 | |
| 1981-1990 | 5.22 | 9.97 | 0.38 | 1.48 | 3.49 | 0.25 | 1.48 | |
| 1991-2000 | 1.58 | 2.23 | 1.09 | 0.09 | 0.78 | 0.71 | 0.09 | |
| 2001-2007 | -0.05 | 3.54 | 1.14 | -2.03 | 1.24 | 0.74 | -2.03 | |
| Grenada | 4.13 | 5.47 | 0.78 | 1.71 | 1.92 | 0.51 | 1.71 | |
| 1971-1980 | 3.98 | 4.56 | 1.19 | 1.61 | 1.59 | 0.77 | 1.61 | |
| 1981-1990 | 7.39 | 8.28 | 0.43 | 4.21 | 2.90 | 0.28 | 4.21 | |
| 1991-2000 | 3.70 | 5.65 | 0.37 | 1.48 | 1.98 | 0.24 | 1.48 | |
| 2001-2007 | 0.32 | 2.50 | 1.28 | -1.39 | 0.88 | 0.83 | -1.39 | |
| St. Kitts & Nevis /b | 4.88 | 5.14 | -0.18 | 3.20 | 1.80 | -0.12 | 3.20 | |
| 1971-1980 | 7.61 | 5.21 | -0.41 | 6.05 | 1.82 | -0.26 | 6.05 | |
| 1981-1990 | 5.43 | 5.60 | -0.65 | 3.90 | 1.96 | -0.42 | 3.90 | |
| 1991-2000 | 3.22 | -0.05 | -0.33 | 3.45 | -0.02 | -0.21 | 3.45 | |
| 2001-2007 | 5.28 | 11.86 | 0.78 | 0.62 | 4.15 | 0.50 | 0.62 | |
| St. Lucia | 3.99 | 6.12 | 2.28 | 0.37 | 2.14 | 1.48 | 0.37 | |
| 1971-1980 | 5.40 | 2.44 | 2.96 | 2.62 | 0.85 | 1.92 | 2.62 | |
| 1981-1990 | 4.85 | 9.93 | 2.36 | -0.16 | 3.47 | 1.53 | -0.16 | |
| 1991-2000 | 2.62 | 7.87 | 2.00 | -1.44 | 2.75 | 1.30 | -1.44 | |
| 2001-2007 | 2.72 | 3.46 | 1.60 | 0.47 | 1.21 | 1.04 | 0.47 | |
| St. Vincent & Grenadines | 4.15 | 4.09 | 1.43 | 1.79 | 1.43 | 0.93 | 1.79 | |
| 1971-1980 | 2.99 | 1.76 | 2.28 | 0.89 | 0.62 | 1.48 | 0.89 | |
| 1981-1990 | 6.85 | 5.36 | 1.82 | 3.80 | 1.88 | 1.18 | 3.80 | |
| 1991-2000 | 2.22 | 4.98 | 0.77 | -0.02 | 1.74 | 0.50 | -0.02 | |
| 2001-2007 | 4.73 | 4.36 | 0.63 | 2.79 | 1.53 | 0.41 | 2.79 | |
| Average | 3.99 | 5.00 | 1.12 | 1.52 | 1.75 | 0.73 | 1.52 | |
Data available only from 1996.
Data available only from 1977.
ECCU: Output Growth and its Components: Ratio to Barbados Values, 1970–2007
(In percent, adjusted for the effects of hurricanes on capital)
ECCU: Output Growth and its Components: Ratio to Barbados Values, 1970–2007
(In percent, adjusted for the effects of hurricanes on capital)
| Country | Year | Output per worker | Capital per worker | Productivity |
|---|---|---|---|---|
| Antigua and Barbuda | 1970 | |||
| 1980 | ||||
| 1990 | ||||
| 2000 | 0.70 | 0.34 | 1.01 | |
| 2007 | 0.80 | 0.52 | 1.01 | |
| Dominica | 1970 | |||
| 1980 | 0.14 | 0.06 | 0.37 | |
| 1990 | 0.24 | 0.14 | 0.47 | |
| 2000 | 0.24 | 0.17 | 0.46 | |
| 2007 | 0.21 | 0.19 | 0.37 | |
| Grenada | 1970 | 0.26 | 0.25 | 0.43 |
| 1980 | 0.27 | 0.29 | 0.41 | |
| 1990 | 0.57 | 0.62 | 0.68 | |
| 2000 | 0.77 | 1.10 | 0.75 | |
| 2007 | 0.68 | 1.17 | 0.64 | |
| St. Kitts & Nevis | 1970 | |||
| 1980 | 0.19 | 0.15 | 0.36 | |
| 1990 | 0.36 | 0.28 | 0.57 | |
| 2000 | 0.50 | 0.30 | 0.77 | |
| 2007 | 0.64 | 0.63 | 0.75 | |
| St. Lucia | 1970 | 0.39 | 0.14 | 0.78 |
| 1980 | 0.38 | 0.11 | 0.82 | |
| 1990 | 0.52 | 0.22 | 0.88 | |
| 2000 | 0.53 | 0.41 | 0.73 | |
| 2007 | 0.54 | 0.46 | 0.71 | |
| St. Vincent & Grenadines | 1970 | 0.14 | 0.11 | 0.29 |
| 1980 | 0.11 | 0.09 | 0.26 | |
| 1990 | 0.19 | 0.12 | 0.41 | |
| 2000 | 0.22 | 0.19 | 0.39 | |
| 2007 | 0.27 | 0.24 | 0.45 |
ECCU: Output Growth and its Components: Ratio to Barbados Values, 1970–2007
(In percent, adjusted for the effects of hurricanes on capital)
| Country | Year | Output per worker | Capital per worker | Productivity |
|---|---|---|---|---|
| Antigua and Barbuda | 1970 | |||
| 1980 | ||||
| 1990 | ||||
| 2000 | 0.70 | 0.34 | 1.01 | |
| 2007 | 0.80 | 0.52 | 1.01 | |
| Dominica | 1970 | |||
| 1980 | 0.14 | 0.06 | 0.37 | |
| 1990 | 0.24 | 0.14 | 0.47 | |
| 2000 | 0.24 | 0.17 | 0.46 | |
| 2007 | 0.21 | 0.19 | 0.37 | |
| Grenada | 1970 | 0.26 | 0.25 | 0.43 |
| 1980 | 0.27 | 0.29 | 0.41 | |
| 1990 | 0.57 | 0.62 | 0.68 | |
| 2000 | 0.77 | 1.10 | 0.75 | |
| 2007 | 0.68 | 1.17 | 0.64 | |
| St. Kitts & Nevis | 1970 | |||
| 1980 | 0.19 | 0.15 | 0.36 | |
| 1990 | 0.36 | 0.28 | 0.57 | |
| 2000 | 0.50 | 0.30 | 0.77 | |
| 2007 | 0.64 | 0.63 | 0.75 | |
| St. Lucia | 1970 | 0.39 | 0.14 | 0.78 |
| 1980 | 0.38 | 0.11 | 0.82 | |
| 1990 | 0.52 | 0.22 | 0.88 | |
| 2000 | 0.53 | 0.41 | 0.73 | |
| 2007 | 0.54 | 0.46 | 0.71 | |
| St. Vincent & Grenadines | 1970 | 0.14 | 0.11 | 0.29 |
| 1980 | 0.11 | 0.09 | 0.26 | |
| 1990 | 0.19 | 0.12 | 0.41 | |
| 2000 | 0.22 | 0.19 | 0.39 | |
| 2007 | 0.27 | 0.24 | 0.45 |
Tourism and Growth Estimations
Tourism and Growth Estimations
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | |
|---|---|---|---|---|---|---|---|
| Variables | GDP growth | GDP growth | GDP growth | GDP growth | GDP growth | GDP growth | GDP growth |
| Convergence | -2.055*** | -3.582*** | -3.692*** | -3.617*** | -1.911*** | -4.019*** | -3.888*** |
| (0.324) | (0.406) | (0.407) | (0.404) | (0.335) | (0.426) | (0.421) | |
| Government spending/GDP | -1.000** | -0.926* | -0.773 | -0.935** | -1.163** | -0.551 | -0.722 |
| (0.455) | (0.476) | (0.470) | (0.473) | (0.466) | (0.486) | (0.477) | |
| Primary education | 1.985*** | 1.270* | 1.201 | 1.260* | 1.835** | 0.805 | 1.510** |
| (0.730) | (0.743) | (0.737) | (0.741) | (0.762) | (0.766) | (0.760) | |
| Openness | 1.354*** | 0.182 | 0.251 | 0.203 | 1.059*** | 0.036 | 0.160 |
| (0.377) | (0.414) | (0.407) | (0.412) | (0.396) | (0.420) | (0.415) | |
| Inflation | -0.522*** | -0.534*** | -0.518*** | -0.525*** | -0.591*** | -0.497*** | -0.506*** |
| (0.134) | (0.126) | (0.126) | (0.126) | (0.133) | (0.129) | (0.126) | |
| Terms of trade | 0.043** | 0.081*** | 0.080*** | 0.080*** | 0.045** | 0.086*** | 0.082*** |
| (0.020) | (0.019) | (0.019) | (0.019) | (0.020) | (0.019) | (0.019) | |
| Life expectancy | 7.272*** | 5.097** | 5.083** | 5.220** | 5.586** | 3.874* | 5.476** |
| (2.390) | (2.272) | (2.267) | (2.269) | (2.342) | (2.243) | (2.283) | |
| Size | 0.000* | 0.000*** | 0.000*** | 0.000*** | 0.000* | 0.000*** | 0.000*** |
| (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | |
| Investment / GDP | 2.824*** | 1.221*** | 1.311*** | 1.302*** | 2.038*** | 0.897** | 1.364*** |
| (0.473) | (0.455) | (0.456) | (0.456) | (0.484) | (0.451) | (0.457) | |
| Arrivals / population | 1.795*** | 1.915*** | 1.875*** | 2.159*** | 2.516*** | ||
| (0.265) | (0.271) | (0.268) | (0.271) | (0.440) | |||
| SI | -2.991*** | -1.820** | -3.099*** | -3.339*** | |||
| (0.906) | (0.825) | (0.971) | (0.949) | ||||
| Islands | -1.870** | ||||||
| (0.756) | |||||||
| Receipts as % of GDP | 0.420** | ||||||
| (0.197) | |||||||
| Receipts per tourist | 0.426** | ||||||
| (0.212) | |||||||
| (Arrivals / population)2 | 0.096* | ||||||
| (0.057) | |||||||
| Constant | -29.730*** | 9.310 | 10.390 | 9.477 | -17.246* | 25.345** | 9.665 |
| (8.740) | (9.917) | (9.919) | (9.910) | (9.025) | (10.648) | (9.943) | |
| Observations | 791 | 760 | 760 | 760 | 772 | 749 | 760 |
| Number of countries | 154 | 152 | 152 | 152 | 153 | 151 | 152 |
Tourism and Growth Estimations
| (1) | (2) | (3) | (4) | (5) | (6) | (7) | |
|---|---|---|---|---|---|---|---|
| Variables | GDP growth | GDP growth | GDP growth | GDP growth | GDP growth | GDP growth | GDP growth |
| Convergence | -2.055*** | -3.582*** | -3.692*** | -3.617*** | -1.911*** | -4.019*** | -3.888*** |
| (0.324) | (0.406) | (0.407) | (0.404) | (0.335) | (0.426) | (0.421) | |
| Government spending/GDP | -1.000** | -0.926* | -0.773 | -0.935** | -1.163** | -0.551 | -0.722 |
| (0.455) | (0.476) | (0.470) | (0.473) | (0.466) | (0.486) | (0.477) | |
| Primary education | 1.985*** | 1.270* | 1.201 | 1.260* | 1.835** | 0.805 | 1.510** |
| (0.730) | (0.743) | (0.737) | (0.741) | (0.762) | (0.766) | (0.760) | |
| Openness | 1.354*** | 0.182 | 0.251 | 0.203 | 1.059*** | 0.036 | 0.160 |
| (0.377) | (0.414) | (0.407) | (0.412) | (0.396) | (0.420) | (0.415) | |
| Inflation | -0.522*** | -0.534*** | -0.518*** | -0.525*** | -0.591*** | -0.497*** | -0.506*** |
| (0.134) | (0.126) | (0.126) | (0.126) | (0.133) | (0.129) | (0.126) | |
| Terms of trade | 0.043** | 0.081*** | 0.080*** | 0.080*** | 0.045** | 0.086*** | 0.082*** |
| (0.020) | (0.019) | (0.019) | (0.019) | (0.020) | (0.019) | (0.019) | |
| Life expectancy | 7.272*** | 5.097** | 5.083** | 5.220** | 5.586** | 3.874* | 5.476** |
| (2.390) | (2.272) | (2.267) | (2.269) | (2.342) | (2.243) | (2.283) | |
| Size | 0.000* | 0.000*** | 0.000*** | 0.000*** | 0.000* | 0.000*** | 0.000*** |
| (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | |
| Investment / GDP | 2.824*** | 1.221*** | 1.311*** | 1.302*** | 2.038*** | 0.897** | 1.364*** |
| (0.473) | (0.455) | (0.456) | (0.456) | (0.484) | (0.451) | (0.457) | |
| Arrivals / population | 1.795*** | 1.915*** | 1.875*** | 2.159*** | 2.516*** | ||
| (0.265) | (0.271) | (0.268) | (0.271) | (0.440) | |||
| SI | -2.991*** | -1.820** | -3.099*** | -3.339*** | |||
| (0.906) | (0.825) | (0.971) | (0.949) | ||||
| Islands | -1.870** | ||||||
| (0.756) | |||||||
| Receipts as % of GDP | 0.420** | ||||||
| (0.197) | |||||||
| Receipts per tourist | 0.426** | ||||||
| (0.212) | |||||||
| (Arrivals / population)2 | 0.096* | ||||||
| (0.057) | |||||||
| Constant | -29.730*** | 9.310 | 10.390 | 9.477 | -17.246* | 25.345** | 9.665 |
| (8.740) | (9.917) | (9.919) | (9.910) | (9.025) | (10.648) | (9.943) | |
| Observations | 791 | 760 | 760 | 760 | 772 | 749 | 760 |
| Number of countries | 154 | 152 | 152 | 152 | 153 | 151 | 152 |
Debt and Growth Estimations
Debt and Growth Estimations
| (1) | (2) | (3) | (4) | (5) | |
|---|---|---|---|---|---|
| Variables | GDP growth | GDP growth | GDP growth | GDP growth | GDP growth |
| Convergence | -3.692*** | -2.343*** | -2.424*** | -2.421*** | -2.376*** |
| (0.407) | (0.613) | (0.620) | (0.621) | (0.616) | |
| Government / GDP | -0.773 | -1.014 | -1.036 | -1.016 | -1.014 |
| (0.470) | (0.657) | (0.661) | (0.662) | (0.661) | |
| Primary education | 1.201 | 0.404 | 0.245 | 0.220 | 0.418 |
| (0.737) | (1.352) | (1.353) | (1.367) | (1.356) | |
| Openness | 0.251 | 0.693 | 0.621 | 0.619 | 0.695 |
| (0.407) | (0.592) | (0.594) | (0.594) | (0.594) | |
| Inflation | -0.518*** | -0.468** | -0.498*** | -0.498*** | -0.465** |
| (0.126) | (0.184) | (0.186) | (0.186) | (0.184) | |
| Terms of trade | 0.080*** | 0.111*** | 0.110*** | 0.110*** | 0.110*** |
| (0.019) | (0.030) | (0.030) | (0.030) | (0.030) | |
| Life expectancy | 5.083** | -3.337 | -3.542 | -3.499 | -3.057 |
| (2.267) | (3.806) | (3.784) | (3.808) | (3.848) | |
| Investment / GDP | 1.311*** | 1.005 | 0.933 | 0.967 | 1.023 |
| (0.456) | (0.761) | (0.767) | (0.766) | (0.765) | |
| Arrivals / population | 1.915*** | 1.763*** | 1.833*** | 1.830*** | 1.793*** |
| (0.271) | (0.405) | (0.404) | (0.404) | (0.405) | |
| Size | 0.000*** | 0.000** | 0.000** | 0.000** | 0.000** |
| (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | |
| Small Islands | -2.991*** | -2.595** | -2.575** | -2.608** | -2.670** |
| (0.906) | (1.263) | (1.278) | (1.275) | (1.276) | |
| Debt / GDP | -0.509** | -0.667* | |||
| (0.230) | (0.373) | ||||
| (Debt / GDP)*Dummy0-30 | -0.270 | -0.269 | |||
| (0.335) | (0.330) | ||||
| (Debt / GDP)*Dummy30-60 | -0.418 | ||||
| (0.268) | |||||
| (Debt/GDP)*Dummy>60 | -0.444* | ||||
| (0.241) | |||||
| (Debt / GDP)*Dummy30-90 | -0.422* | ||||
| (0.256) | |||||
| (Debt/GDP)*Dummy>90 | -0.453* | ||||
| (0.240) | |||||
| (Debt / GDP)2 | 0.026 | 0.026 | |||
| (0.047) | (0.047) | ||||
| Constant | 10.390 | 38.936** | 38.183** | 41.364** | 38.183** |
| (9.919) | (16.455) | (16.585) | (16.272) | (16.585) | |
| Observations | 760 | 467 | 467 | 467 | 467 |
| Number of countries | 152 | 136 | 136 | 136 | 136 |
Debt and Growth Estimations
| (1) | (2) | (3) | (4) | (5) | |
|---|---|---|---|---|---|
| Variables | GDP growth | GDP growth | GDP growth | GDP growth | GDP growth |
| Convergence | -3.692*** | -2.343*** | -2.424*** | -2.421*** | -2.376*** |
| (0.407) | (0.613) | (0.620) | (0.621) | (0.616) | |
| Government / GDP | -0.773 | -1.014 | -1.036 | -1.016 | -1.014 |
| (0.470) | (0.657) | (0.661) | (0.662) | (0.661) | |
| Primary education | 1.201 | 0.404 | 0.245 | 0.220 | 0.418 |
| (0.737) | (1.352) | (1.353) | (1.367) | (1.356) | |
| Openness | 0.251 | 0.693 | 0.621 | 0.619 | 0.695 |
| (0.407) | (0.592) | (0.594) | (0.594) | (0.594) | |
| Inflation | -0.518*** | -0.468** | -0.498*** | -0.498*** | -0.465** |
| (0.126) | (0.184) | (0.186) | (0.186) | (0.184) | |
| Terms of trade | 0.080*** | 0.111*** | 0.110*** | 0.110*** | 0.110*** |
| (0.019) | (0.030) | (0.030) | (0.030) | (0.030) | |
| Life expectancy | 5.083** | -3.337 | -3.542 | -3.499 | -3.057 |
| (2.267) | (3.806) | (3.784) | (3.808) | (3.848) | |
| Investment / GDP | 1.311*** | 1.005 | 0.933 | 0.967 | 1.023 |
| (0.456) | (0.761) | (0.767) | (0.766) | (0.765) | |
| Arrivals / population | 1.915*** | 1.763*** | 1.833*** | 1.830*** | 1.793*** |
| (0.271) | (0.405) | (0.404) | (0.404) | (0.405) | |
| Size | 0.000*** | 0.000** | 0.000** | 0.000** | 0.000** |
| (0.000) | (0.000) | (0.000) | (0.000) | (0.000) | |
| Small Islands | -2.991*** | -2.595** | -2.575** | -2.608** | -2.670** |
| (0.906) | (1.263) | (1.278) | (1.275) | (1.276) | |
| Debt / GDP | -0.509** | -0.667* | |||
| (0.230) | (0.373) | ||||
| (Debt / GDP)*Dummy0-30 | -0.270 | -0.269 | |||
| (0.335) | (0.330) | ||||
| (Debt / GDP)*Dummy30-60 | -0.418 | ||||
| (0.268) | |||||
| (Debt/GDP)*Dummy>60 | -0.444* | ||||
| (0.241) | |||||
| (Debt / GDP)*Dummy30-90 | -0.422* | ||||
| (0.256) | |||||
| (Debt/GDP)*Dummy>90 | -0.453* | ||||
| (0.240) | |||||
| (Debt / GDP)2 | 0.026 | 0.026 | |||
| (0.047) | (0.047) | ||||
| Constant | 10.390 | 38.936** | 38.183** | 41.364** | 38.183** |
| (9.919) | (16.455) | (16.585) | (16.272) | (16.585) | |
| Observations | 760 | 467 | 467 | 467 | 467 |
| Number of countries | 152 | 136 | 136 | 136 | 136 |
Countries Included in the Samples
Countries Included in the Samples
| Algeria * | Egypt | Lithuania |
| Angola * | El Salvador | Macedonia, Republic of |
| Argentina | Equatorial Guinea§† | Madagascar |
| Australia | Estonia | Malawi |
| Austria | Ethiopia | Malaysia |
| Azerbaijan | Finland | Maldives * |
| Bahamas, The | France | Mali |
| Bahrain | Gabon† | Malta |
| Bangladesh * | Gambia, The | Mauritania |
| Barbados | Georgia | Mauritius |
| Belarus | Germany | Mexico |
| Belgium | Ghana | Moldova§† |
| Belize | Greece | Mongolia |
| Benin | Grenada | Morocco |
| Bhutan | Guatemala | Mozambique |
| Bolivia | Guinea-Bissau | Namibia |
| Botswana | Guyana | Nepal * |
| Brazil | Haiti * | Netherlands |
| Bulgaria | Honduras | New Zealand |
| Burkina Faso | Hong Kong SAR | Nicaragua |
| Burundi | Hungary | Niger |
| Cambodia | Iceland | Nigeria |
| Cameroon | India | Norway |
| Canada | Indonesia | Oman |
| Cape Verde | Iran, Islamic Republic of | Pakistan |
| Central African Republic * | Ireland | Panama |
| Chad | Israel | Papua New Guinea * |
| Chile | Italy | Paraguay |
| China | Jamaica | Peru |
| Colombia | Japan | Philippines |
| Congo, Dem. Rep. of * | Jordan | Poland |
| Congo, Rep. of | Kazakhstan | Portugal |
| Costa Rica | Kenya | Qatar§† |
| Côte d’Ivoire | Korea, Republic of | República Bolivariana de Venezuela |
| Croatia | Kuwait | Romania |
| Cyprus | Kyrgyz Republic | Russian Federation |
| Czech Republic | Lao People’s Democratic Republic | Rwanda |
| Denmark | Latvia | Samoa * |
| Dominica | Lebanon | Saudi Arabia |
| Dominican Republic * | Lesotho | Senegal |
| Ecuador | Libya | Seychelles |
| Sierra Leone * | Sudan | Turkey |
| Singapore | Suriname | Uganda |
| Slovak Republic | Swaziland | Ukraine |
| Slovenia | Sweden | United Kingdom |
| Solomon Islands * | Switzerland | United States |
| South Africa | Syrian Arab Republic | Uruguay |
| Spain | Tanzania * | Vanuatu * |
| Sri Lanka | Thailand | Vietnam |
| St. Kitts and Nevis | Togo | Zambia |
| St. Lucia | Trinidad and Tobago | |
| St. Vincent and the Grenadines | Tunisia |
Countries Included in the Samples
| Algeria * | Egypt | Lithuania |
| Angola * | El Salvador | Macedonia, Republic of |
| Argentina | Equatorial Guinea§† | Madagascar |
| Australia | Estonia | Malawi |
| Austria | Ethiopia | Malaysia |
| Azerbaijan | Finland | Maldives * |
| Bahamas, The | France | Mali |
| Bahrain | Gabon† | Malta |
| Bangladesh * | Gambia, The | Mauritania |
| Barbados | Georgia | Mauritius |
| Belarus | Germany | Mexico |
| Belgium | Ghana | Moldova§† |
| Belize | Greece | Mongolia |
| Benin | Grenada | Morocco |
| Bhutan | Guatemala | Mozambique |
| Bolivia | Guinea-Bissau | Namibia |
| Botswana | Guyana | Nepal * |
| Brazil | Haiti * | Netherlands |
| Bulgaria | Honduras | New Zealand |
| Burkina Faso | Hong Kong SAR | Nicaragua |
| Burundi | Hungary | Niger |
| Cambodia | Iceland | Nigeria |
| Cameroon | India | Norway |
| Canada | Indonesia | Oman |
| Cape Verde | Iran, Islamic Republic of | Pakistan |
| Central African Republic * | Ireland | Panama |
| Chad | Israel | Papua New Guinea * |
| Chile | Italy | Paraguay |
| China | Jamaica | Peru |
| Colombia | Japan | Philippines |
| Congo, Dem. Rep. of * | Jordan | Poland |
| Congo, Rep. of | Kazakhstan | Portugal |
| Costa Rica | Kenya | Qatar§† |
| Côte d’Ivoire | Korea, Republic of | República Bolivariana de Venezuela |
| Croatia | Kuwait | Romania |
| Cyprus | Kyrgyz Republic | Russian Federation |
| Czech Republic | Lao People’s Democratic Republic | Rwanda |
| Denmark | Latvia | Samoa * |
| Dominica | Lebanon | Saudi Arabia |
| Dominican Republic * | Lesotho | Senegal |
| Ecuador | Libya | Seychelles |
| Sierra Leone * | Sudan | Turkey |
| Singapore | Suriname | Uganda |
| Slovak Republic | Swaziland | Ukraine |
| Slovenia | Sweden | United Kingdom |
| Solomon Islands * | Switzerland | United States |
| South Africa | Syrian Arab Republic | Uruguay |
| Spain | Tanzania * | Vanuatu * |
| Sri Lanka | Thailand | Vietnam |
| St. Kitts and Nevis | Togo | Zambia |
| St. Lucia | Trinidad and Tobago | |
| St. Vincent and the Grenadines | Tunisia |
References
Arezki, Rabah, Reda Cherif, and John Piotrowski, 2009, “Tourism Specialization and Economic Development: Evidence from the UNESCO World Heritage List,” IMF Working Paper 09/176 (Washington: International Monetary Fund).
Barro, Robert, 1998, “Notes on Growth Accounting,” NBER Working Paper 6654 (Cambridge, MA: National Bureau of Economic Research).
Barro, Robert, and Jong-Wha Lee, 1994, “Sources of Economic Growth,” Carnegie-Rochester Conference Series on Public Policy, Vol. 40, pp. 1–46.
Barro, Robert and Jong-Wha Lee, 2001, “International Data on Educational Attainment: Updates and Implications,” Oxford Economic Papers, Vol. 53, pp. 541–63.
Center for Research on the Epidemiology of Disasters (CRED), 2010, Emergency Disaster Database (EM-DAT).
Denison, Edward, 1962, “Sources of Growth in the United States and the Alternative Before Us,” Supplement Paper 13 (New York: Committee for Economic Development).
Denison, Edward F., Zvi Griliches and dale Jorgenson, 1972, “The Measurement of Productivity,” Survey of Current Business 52 (May, part 2), pp. 3–111 (special issue).
Fiscal Affairs Department (FAD), 2010, Fiscal Monitor (Washington: International Monetary Fund).
Hall, Robert and Charles Jones, 1999, “Why do Some Countries Produce so Much More Output per Worker than Others?,” Quarterly Journal of Economics, Vol. 114, pp. 83–116.
Heston, Alan; Robert Summers and Bettina Aten, 2009, “Penn World Table v. 6.3,” Center for International Comparisons of Production, Income and Prices, (Philadelphia: University of Pennsylvania).
International Monetary Fund, 2008–2010, World Economic Outlook (Washington: International Monetary Fund).
International Monetary Fund, 2010, International Financial Statistics (Washington: International Monetary Fund).
Jorgenson, Dale and Zvi Griliches, 1967, “The Explanation of Productivity Change,” Review of Economic Studies, Vol. 34, pp. 249–80.
Lee, Jong-Wha, 2005, “Human capital and productivity for Korea’s sustained economic growth,” Journal of Asian Economics, Vol. 16, pp. 663–87.
Loayza, Norman, Pablo Fajnzylber and Cesar Calderon, 2004, Economic Growth in Latin America and the Caribbean: Stylized Facts, Explanations, and Forecasts (Washington: World Bank).
Pattillo, Catherine, Helene Poirson and Luca Ricci, 2002, “External Debt and Growth,” IMF Working Paper 02/69 (Washington: International Monetary Fund).
Reinhart, Carmen and Kenneth Rogoff, 2010, “Growth in a Time of Debt,” American Economic Review, Vol. 100, pp. 573–78.
Solow, Robert,, 1957, “Technical Change and the Aggregate Production Function,” Review of Economics and Statistics, Vol. 39, pp. 312–20.
World Bank, 2010, World Development Indicators (Washington: World Bank).
World Travel Organization, (several years) 1980–2009, Yearbook of Tourism Statistics (Madrid: World Travel Organization).
Prepared by N. Thacker, S. Acevedo, J. Kang, R. Perrelli and M. Tashu.
Mostly in the early seventies.
For the purpose of this chapter, SI economies comprise 23 other small islands in the rest of the world. These are: Bermuda, Cape Verde, Comoros, Cuba, Cyprus, Dominican Republic, Fiji, Haiti, Kiribati, Maldives, Malta, Marshall Islands, Mauritius, Federated States of Micronesia, Palau, Papua New Guinea, Samoa, Sao Tome and Principe, Seychelles, Solomon Islands, Sri Lanka, Tonga, and Vanuatu. The other Caribbean countries refer to all the Caribbean countries minus the ECCU countries.
See Appendix 1 on the methodology. To the extent that TFP is a residual, it may reflect measurement errors on capital and labor.
Barbados is used as the reference country because, like most ECCU countries, it relies on tourism from different sources as the main driver of growth and it has one of the highest GDP per capita in the region.
The sample used covers 5 five-year periods: 1979-1983, 1984-1988, 1989-1993, 1994-1998, 1999-2003 and one four-year period 2004-2007.
The explanatory variables are included as the averages of the five-year periods.
Unfortunately Antigua and Barbuda, Bermuda, Comoros, Cuba, Fiji, Federated States of Micronesia, Kiribati, Marshall Islands, Palau, Sao Tome and Principe and Tonga did not have all the data required to be included in the estimations.
To correct for the possible correlation of investment and tourism with the individual effects of each country, we used the Hausman-Taylor estimator, allowing the use of a random effects model that includes time-invariant variables.
The results are not presented but they are available on request.
See also Chapter II on Public Debt in the ECCU.
The samples used in the debt regressions of Table 4 are smaller due to data unavailability for some countries. Table 5 in the Appendix details the countries included in the different estimations.
Following Kumar and Woo (2010) and Patillo et al. (2002) the approach explores the nonlinearities of the growth-debt relationship by introducing interactions terms between debt and dummies for three ranges of debt-to-GDP; 0 to 30 percent, 30 to 90 percent and 90 percent and above, and by including a quadratic specification for the debt variable.
Growth accounting has some important limitations. First, the TFP component is by definition a residual and therefore picks up measurement errors in the data. Also a failure to account for improvements in the quality and composition of the physical capital and the differences in human capital of the labor force will lead to an overestimation of TFP. Third, it does not provide any insight into why TFP changes from one period to another.
Since the focus of this study are small islands, particularly ECCU countries, which do not have sufficient data to construct educational attainment measured by average years of schooling à la Barro and Lee (2001), we use the most basic production function that only includes capital and labor as factor inputs, leaving out the contribution of human capital to growth.
The dots over the variables denote derivatives with respect to time as in most of the growth literature.
Data used is from 1970-2007, as the PWT has data only up to 2007.
The calculated TFP growth did not change much when we assumed α to be higher (0.45) or lower (0.30).
However, the data in EM-DAT is not exhaustive and has some limitations. For example, there are no estimates of damages in some years despite reported hurricanes in those years. Also, it is possible that the recorded damages are over or underestimating the true damages caused by hurricanes. In the case of St. Lucia in 1988 we omit EM-DAT’s entry of US$1 billion in damages because it appears erroneous, however it is difficult to know if there are other less conspicuous errors in the data.
