Abstract
This Selected Issues paper on the Eastern Caribbean Currency Union (ECCU) underlies key features of business cycles. To obtain new measures of classical and growth cycles, simple rules were applied to date turning points in the classical business cycle, and a recently developed frequency domain filter was used to estimate the growth cycle. At the regional level, the ECCU countries are facing two shocks, i.e., the depreciation of the U.S. dollar and the depreciation of the Dominican Republic’s peso. The countries of the ECCU have experienced modest erosion in their price and nonprice competitiveness.
III. Competitiveness in the ECCU: Measures of the Real Exchange Rate1
A. Introduction
1. The external environment faced by the Eastern Caribbean Currency Union (ECCU) has been more adverse in the last decade than it was prior to the 1990s. A series of exogenous shocks have hit the ECCU region in recent years, including: the terrorism-induced decline in tourism; a series of natural disasters affecting the region; the dismantling of preferential trade agreements and their impact on traditional agricultural exports (bananas and sugar); and the Financial Action Task Force (FATF) initiatives on offshore financial services arrangements.
2. The region is also experiencing a loss of market share in its principal export—tourism—though receipts have not declined significantly (see International Monetary Fund, 2003). Tourism indicators show that the ECCU’s share of the Caribbean tourism market has declined as other non-CARICOM tourist destinations—notably Cuba, Dominican Republic and Mexico—have rapidly increased their market share (Table III.1). The composition of tourists visiting the ECCU has also changed: stayover tourist arrivals in the ECCU have declined relative to the Caribbean totals, while cruise passengers arrivals have increased. Since tourism receipts have not declined as much in relative terms, this could signal that the ECCU is moving towards a more high-income (smaller) market segment.
Selected Caribbean Market Share Indicators 1/ 2/
(Figures are expressed as ratios to the respective Caribbean totals)
‘CARICOM’ refers to English-Speaking, traditional CARICOM members.
The 1990 column is only indicative, since the data for key “Other Caribbean countries” (such as Cancun/Mexico, Cozumel/Mexico, the Dominican Republic, and Puerto Rico) is not available.
Selected Caribbean Market Share Indicators 1/ 2/
(Figures are expressed as ratios to the respective Caribbean totals)
| Destination | 1990 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | Average 1995–97 |
Average | Change 1995–97 2/ |
Change 1998–2002 |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Stayover tourist arrivals | ||||||||||||||
| CARICOM | 31.3 | 27.8 | 27.2 | 26.4 | 25.8 | 25.4 | 24.6 | 24.6 | 25.5 | 27.1 | 25.2 | −1.4 | −0.4 | |
| ECCU | 5.4 | 5.0 | 4.9 | 4.9 | 4.8 | 4.7 | 4.4 | 4.4 | 4.6 | 4.9 | 4.6 | −0.2 | −0.2 | |
| Non-ECCU | 25.9 | 22.7 | 22.3 | 21.5 | 21.0 | 20.7 | 20.2 | 20.3 | 20.8 | 22.2 | 20.6 | −1.2 | −0.2 | |
| Other | 68.7 | 72.2 | 72.8 | 73.6 | 74.2 | 74.6 | 75.4 | 75.4 | 74.5 | 72.9 | 74.8 | 1.4 | 0.4 | |
| Cruise passenger arrivals | ||||||||||||||
| CARICOM | 50.7 | 37.3 | 36.2 | 34.9 | 35.2 | 37.7 | 39.1 | 38.0 | 35.9 | 36.1 | 37.2 | −2.4 | 0.7 | |
| ECCU | 9.8 | 10.1 | 9.7 | 10.0 | 11.3 | 10.8 | 10.7 | 10.7 | 7.5 | 9.9 | 10.2 | −0.1 | −3.8 | |
| Non-ECCU | 40.9 | 27.2 | 26.5 | 24.9 | 23.9 | 26.9 | 28.4 | 27.4 | 28.4 | 26.2 | 27.0 | −2.3 | 4.6 | |
| Other | 49.3 | 62.7 | 63.8 | 65.1 | 64.8 | 62.3 | 60.9 | 62.0 | 64.1 | 63.9 | 62.8 | 2.4 | −0.7 | |
| Estimates of visitor expenditure | ||||||||||||||
| CARICOM | 35.2 | 28.9 | 28.0 | 27.2 | 26.1 | 26.3 | 25.8 | 25.4 | 26.3 | 28.1 | 26.0 | −1.7 | 0.2 | |
| ECCU | 6.8 | 5.6 | 5.3 | 5.4 | 5.2 | 4.9 | 4.5 | 4.8 | 4.8 | 5.5 | 4.8 | −0.3 | −0.4 | |
| Non-ECCU | 28.4 | 23.3 | 22.7 | 21.8 | 20.9 | 21.4 | 21.3 | 20.6 | 21.5 | 22.6 | 21.2 | |||
| Other | 64.8 | 71.1 | 72.0 | 72.8 | 73.9 | 73.7 | 74.2 | 74.6 | 73.7 | 71.9 | 74.0 | 1.7 | −0.2 | |
| Rooms in tourist accommodation | ||||||||||||||
| CARICOM | 58.1 | 30.6 | 29.9 | 28.8 | 28.1 | 27.9 | 27.4 | 27.2 | 27.1 | 28.8 | 28.2 | −1.8 | −1.0 | |
| ECCU | 8.7 | 7.0 | 6.4 | 6.1 | 6.0 | 5.8 | 6.0 | 5.8 | 5.4 | 6.2 | 6.0 | −0.9 | −0.6 | |
| Non-ECCU | 49.3 | 23.6 | 23.6 | 22.7 | 22.1 | 22.2 | 21.4 | 21.4 | 21.7 | 22.6 | 22.2 | −0.9 | −0.4 | |
| Other | 41.9 | 69.4 | 70.1 | 71.2 | 71.9 | 72.1 | 72.6 | 72.8 | 72.9 | 71.2 | 71.8 | 1.8 | 1.0 | |
‘CARICOM’ refers to English-Speaking, traditional CARICOM members.
The 1990 column is only indicative, since the data for key “Other Caribbean countries” (such as Cancun/Mexico, Cozumel/Mexico, the Dominican Republic, and Puerto Rico) is not available.
Selected Caribbean Market Share Indicators 1/ 2/
(Figures are expressed as ratios to the respective Caribbean totals)
| Destination | 1990 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | Average 1995–97 |
Average | Change 1995–97 2/ |
Change 1998–2002 |
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Stayover tourist arrivals | ||||||||||||||
| CARICOM | 31.3 | 27.8 | 27.2 | 26.4 | 25.8 | 25.4 | 24.6 | 24.6 | 25.5 | 27.1 | 25.2 | −1.4 | −0.4 | |
| ECCU | 5.4 | 5.0 | 4.9 | 4.9 | 4.8 | 4.7 | 4.4 | 4.4 | 4.6 | 4.9 | 4.6 | −0.2 | −0.2 | |
| Non-ECCU | 25.9 | 22.7 | 22.3 | 21.5 | 21.0 | 20.7 | 20.2 | 20.3 | 20.8 | 22.2 | 20.6 | −1.2 | −0.2 | |
| Other | 68.7 | 72.2 | 72.8 | 73.6 | 74.2 | 74.6 | 75.4 | 75.4 | 74.5 | 72.9 | 74.8 | 1.4 | 0.4 | |
| Cruise passenger arrivals | ||||||||||||||
| CARICOM | 50.7 | 37.3 | 36.2 | 34.9 | 35.2 | 37.7 | 39.1 | 38.0 | 35.9 | 36.1 | 37.2 | −2.4 | 0.7 | |
| ECCU | 9.8 | 10.1 | 9.7 | 10.0 | 11.3 | 10.8 | 10.7 | 10.7 | 7.5 | 9.9 | 10.2 | −0.1 | −3.8 | |
| Non-ECCU | 40.9 | 27.2 | 26.5 | 24.9 | 23.9 | 26.9 | 28.4 | 27.4 | 28.4 | 26.2 | 27.0 | −2.3 | 4.6 | |
| Other | 49.3 | 62.7 | 63.8 | 65.1 | 64.8 | 62.3 | 60.9 | 62.0 | 64.1 | 63.9 | 62.8 | 2.4 | −0.7 | |
| Estimates of visitor expenditure | ||||||||||||||
| CARICOM | 35.2 | 28.9 | 28.0 | 27.2 | 26.1 | 26.3 | 25.8 | 25.4 | 26.3 | 28.1 | 26.0 | −1.7 | 0.2 | |
| ECCU | 6.8 | 5.6 | 5.3 | 5.4 | 5.2 | 4.9 | 4.5 | 4.8 | 4.8 | 5.5 | 4.8 | −0.3 | −0.4 | |
| Non-ECCU | 28.4 | 23.3 | 22.7 | 21.8 | 20.9 | 21.4 | 21.3 | 20.6 | 21.5 | 22.6 | 21.2 | |||
| Other | 64.8 | 71.1 | 72.0 | 72.8 | 73.9 | 73.7 | 74.2 | 74.6 | 73.7 | 71.9 | 74.0 | 1.7 | −0.2 | |
| Rooms in tourist accommodation | ||||||||||||||
| CARICOM | 58.1 | 30.6 | 29.9 | 28.8 | 28.1 | 27.9 | 27.4 | 27.2 | 27.1 | 28.8 | 28.2 | −1.8 | −1.0 | |
| ECCU | 8.7 | 7.0 | 6.4 | 6.1 | 6.0 | 5.8 | 6.0 | 5.8 | 5.4 | 6.2 | 6.0 | −0.9 | −0.6 | |
| Non-ECCU | 49.3 | 23.6 | 23.6 | 22.7 | 22.1 | 22.2 | 21.4 | 21.4 | 21.7 | 22.6 | 22.2 | −0.9 | −0.4 | |
| Other | 41.9 | 69.4 | 70.1 | 71.2 | 71.9 | 72.1 | 72.6 | 72.8 | 72.9 | 71.2 | 71.8 | 1.8 | 1.0 | |
‘CARICOM’ refers to English-Speaking, traditional CARICOM members.
The 1990 column is only indicative, since the data for key “Other Caribbean countries” (such as Cancun/Mexico, Cozumel/Mexico, the Dominican Republic, and Puerto Rico) is not available.
3. This chapter assesses the extent to which ECCU real exchange rates have appreciated over time, and whether the region has lost competitiveness. Two approaches are taken to assess the movement of the real exchange rate over time. The first approach looks at traditional real effective exchange rate measures, based on the ratio of nontradable to tradable prices and on the ratio of domestic to foreign prices.2 The second is based on specially constructed measures that are more appropriate for tourism-dominated economies. These measures are based on real exchange rates vis-à-vis: (i) major competitors of the ECCU in the tourism sector; and (ii) major customers of the ECCU tourism sector.
B. Analysis of Traditional Real Exchange Rate Measures
4. Real exchange rates are traditionally measured in one of two ways:
First, as the ratio of the domestic price of nontradable goods to the domestic price of tradable goods. In equation (1), RERj,d1 is the real exchange rate of country j (d1 indicates definition 1), PNT and PT are, respectively, the domestic price indexes of nontradable and tradable goods
Second, as the ratio of domestic prices to foreign prices (usually the consumer price index). In equation (2) below, RERj,d2 denotes the second definition of the real exchange rate of country j. CPIj and Ej denote, respectively, the consumer price index and (an index of) the nominal exchange rate (measured in foreign currency per unit of domestic currency). P* denotes an index of foreign prices (measured in foreign currency). The index of foreign prices may be of a single country or of a group of countries (see for instance equation (3) in the next section). Since this definition allows for a comparison of domestic and international prices over time, changes in this ratio indicate whether goods and services in the home country are becoming cheaper or more expensive than in other countries
Real Exchange Rate Developments
Definition 1
5. In the absence of official measures, developments in four alternative measures of definition 1 are presented below. The measures use various proxies for the prices of nontradable and tradable goods.
6. The first measure of the real exchange rate uses information on the disaggregated consumer price index (CPI) to classify each subgroup of the CPI as a tradable or a nontradable group (there are between seven and eleven subgroups in each country’s CPI).3 Price indexes for both the tradable and nontradable groups are constructed and the nontradable index is divided by the tradable index to obtain a measure of the real exchange rate.
7. The first measure indicates that real exchange rates at end-June 2003, are, in general, more appreciated than in September of 1995—the start of the growth slow down in the region—although the degree varies from country to country (Figure III.1). In Grenada, St. Kitts and Nevis, and St. Vincent and the Grenadines the real exchange rate appreciated by more than 15 percent. In contrast, in Antigua and Barbuda (where data are available only until December 2001), Dominica, and St. Lucia, real exchange rates were at levels comparable to those of the base period. The large appreciations observed in St. Kitts and Nevis and St. Vincent and the Grenadines are, to a large extent, associated with increases in housing prices.
ECCU: Real Exchange Rates—Based on Disaggregated Consumer Price Index, 1990–2003 1/
(September 1995=100)
Citation: IMF Staff Country Reports 2004, 335; 10.5089/9781451811667.002.A003
Source: Fund staff calculations using data from the Eastern Caribbean Central Bank.1/ Calculated as the country-specific price index of nontradable consumer goods divided by the country-specific consumer price index of tradable goods. An increase (decrease) indicates an appreciation (depreciation). Excludes Anguilla and Montserrat in the calculation of the ECCU average. ECCU average excludes Antigua and Barbuda and Dominica before March of 1994 and Antigua and Barbuda after December of 2001. Data up to June 2003 for all countries except Antigua and Barbuda which has data up to December 2001.8. The second measure of the real exchange rate classifies all but food items as nontradable goods. This measure avoids two disadvantages of the first measure, namely: uncertainty about the composition of each of the subgroups (for example, some tradable goods may be part of a subgroup that has been defined as nontradable), and lack of crosscountry homogeneity at the subgroup level.
9. The second measure portrays a similar picture to that indicated by the first measure, but the magnitudes of the appreciations are typically smaller (Figure III.2). The only significant difference is the real exchange rate of St. Lucia, which appears slightly more appreciated under this measure when compared with the first measure.
ECCU: Real Exchange Rates—Non-Food and Food Price Indexes, 1990–2003 1/
(Sep. 1995=100)
Citation: IMF Staff Country Reports 2004, 335; 10.5089/9781451811667.002.A003
Source: Fund staff calculations using data from the Eastern Caribbean Central Bank.1/ Calculated as the price index of non-food consumer goods divided by the price index of food. An increase (decrease) indicates an appreciation (depreciation). Excludes Anguilla and Montserrat in the calculation of the ECCU average. ECCU average excludes Antigua and Barbuda and Dominica before March of 1994 and Antigua and Barbuda after December of 2001. Data up to June 2003 for all countries except Antigua and Barbuda which has data up to December 2001.10. The third measure uses the CPI as the measure of the price of nontradable goods and the (weighted) unit prices of exports and imports as the measure of the price of tradable goods. This approach not only simplifies cross-country comparisons, but also refines the measure of tradable goods, because even highly tradable goods can have a large nontradable component given the importance of distribution costs (which is a nontradable service).4 While consumer price indexes are readily available for the ECCU countries, unit price measures for exports and imports are not, and were estimated using data on global commodity prices and data on the composition of each country’s trade.
11. The third measure shows a modest appreciation in the initial years and a substantial one in later years. During 1990 to 1996 there is a 12 percent appreciation of the real exchange rate, while during 1997 to 2002 there is a 31 percent appreciation (Figure III.3). Overall, the ECCU index indicates a real exchange appreciation of 46 percent from 1990 to 2002, including a 2 percent depreciation in 2002. The estimates for individual countries were broadly similar, though a more substantial appreciation than the ECCU average is visible for St. Lucia and St. Kitts and Nevis.
ECCU: Trade-Based Measure of the Real Exchange Rate, 1990–2002 1/
(1990=100)
Citation: IMF Staff Country Reports 2004, 335; 10.5089/9781451811667.002.A003
Source: Eastern Caribbean Central Bank and Fund staff estimates.1/ Calculated as the consumer price index divided by a weighted index of the unit value indexes for imports and exports. An increase (decrease) indicates an appreciation (depreciation). Excludes Anguilla and Montserrat in the calculation of the ECCU average. The indexes are set at 1990=100 for all series, except for Grenada and Antigua and Barbuda, whose series are set at 1994=100.12. The fourth measure of the real exchange rate is based on sectoral GDP deflators. The tradable goods deflator is calculated by adding GDP in current prices of the agriculture sector to GDP in current prices of the manufacturing sector, and dividing this amount by the sum of the corresponding GDP in constant prices. The nontradable goods deflator is calculated similarly by using GDP that excludes the agriculture and manufacturing sectors.5 As a means to test the efficacy of purchasing power parity as a theory of exchange rate determination, this definition of prices is favored on theoretical grounds (Goldstein and Officer, 1979). In addition, GDP-based indexes are measures based on production rather than consumption, which is of key importance in considering country competitiveness issues.6 Moreover, the GDP-based measures of the real exchange rate are also current-weighted price indexes, accounting for the changing pattern of sectoral production over time.
13. The fourth measure shows that the largest appreciation occurred in St. Kitts and Nevis and the least in Antigua and Barbuda and Dominica (Figure III.4). The appreciation in St. Kitts and Nevis has been primarily associated with a sustained rise in the relative price of nontradables, consistent with the findings related to the first measure. The GDP-weighted ECCU index indicates about a 6 percent real appreciation over the 1990–2002 period, with the appreciation peaking in 1998 at about 9 percent above its 1990 level.7
ECCU: Real Exchange Rates—Based on Sectoral GDP Deflator for Nontradables and Tradables, 1990–2002 1/
(1990=100)
Citation: IMF Staff Country Reports 2004, 335; 10.5089/9781451811667.002.A003
Source: Eastern Caribbean Central Bank and Fund staff estimates1/ Real exchange rate is measured as the price of nontradables relative to the price of tradables. An increase (decrease) indicates and appreciation (depreciation). Tradable sectors include agriculture and manufacturing; nontradable sectors include mining and quarrying and all service sectors.Definition 2
14. The traditional IMF measure of the real effective exchange rate, a version of equation (2), uses a weighted average of foreign prices (with the weights reflecting the home country’s bilateral trade with each country). As mentioned above, the advantage of this definition is that it allows for a comparison of domestic and international prices over time.
15. Definition 2 confirms a similar picture for all countries: an appreciation during the 1990s and a depreciation beginning in 2002 (Figure III.5).8 The appreciation was most pronounced in St. Lucia and least pronounced in Antigua and Barbuda. A comparison of this measure with movements in the U.S. real exchange rate reveals the substantial influence of the latter on the former. A recent example is the real depreciation of the Eastern Caribbean (EC) dollar since early 2002, which has been associated with a sharp depreciation of the U.S. dollar against major currencies, particularly in the latter part of 2002.
ECCU: Real Exchange Rate—Real Effective Exchange Rate, 1990–2003 1/
(1990=100)
Citation: IMF Staff Country Reports 2004, 335; 10.5089/9781451811667.002.A003
Sources: IMF Information Notice System; and Fund staff estimates.1/ Trade-weighted index of nominal exchange rates deflated by seasonally adjusted relative consumer prices. An increase (decrease) indicates an appreciation (depreciation). Excludes Anguilla and Montserrat in the calculation of the ECCU average. Data up to September 2003 for all countries.Regularities Across Real Exchange Rate Measures
16. The largest appreciation of the real exchange rates occurred in St. Kitts and Nevis and St. Vincent and the Grenadines. All five measures of the real exchange rate provide evidence of large appreciations since 1990.
17. Antigua and Barbuda and Dominica appear to have experienced the least appreciation of their real exchange rates. Of the real exchange rate measures, these two countries indicate a large appreciation by one measure only.
18. Evidence on real exchange rates in Grenada and St. Lucia are mixed, depending on the measure used. Grenada’s real exchange rate appreciated based on three measures but not the remaining two, while St. Lucia’s real exchange rate appreciated based on only two measures.
19. The traditional IMF measure of the real exchange rate indicates that current real exchange rate levels are at par with historical levels. However, a more detailed assessment based on the fundamental determinants is needed to determine the degree of undervaluation or overvaluation of the EC dollar.
20. The exchange rate of the EC dollar appears to be close to its long-run equilibrium level, based on estimates that take into account the fundamental determinants of real exchange rates (Box III.1) The recent depreciation of the U.S. dollar against major currencies has contributed to a real depreciation of the EC dollar (through the peg the EC dollar has to the U.S. dollar). The empirical analysis suggests that the real depreciation of the EC dollar has reversed the real appreciation of the currency that occurred during the period 1998–2001.
C. Real Exchange Rates Based on Customers and Competitors of the Tourism Sector
21. For the analysis in this section, real exchange rate measures based on currencies of tourism customer and tourism competitor countries were constructed. These measures are variants of the traditional real effective exchange rate index calculated by the IMF. The index is presented in equation (3), where RERj denotes the real exchange rate of country j; CPIj and Ej denote, respectively, the consumer price index and (an index of) the nominal exchange rate (measured in U.S. dollars per unit of domestic currency) of country j; and wi denotes the weight assigned to each of the partner countries i. It can be seen that when domestic prices (measured in U.S. dollars) increase more than prices in partner countries, the real exchange rate will increase and the domestic currency will appreciate. Similarly, a depreciation occurs when domestic prices (measured in U.S. dollars) fall more in the domestic country than in partner countries.
The Long-Run (Equilibrium) Real Exchange Rate of the EC Dollar
Comparing the ECCU’s real exchange rate with only its historical values to determine the equilibrium exchange rate does not take account of recent macroeconomic shocks that have impacted the region. Therefore, it is important to explore whether or not the current real exchange rate of the EC dollar is consistent with the behavior of its fundamental determinants. Johansen’s method can be used to investigate the existence of a long-run, cointegrating, relationship between the real exchange rate and a set of fundamentals—terms of trade, net foreign liabilities (proxied by public sector external debt), and size of the government (measured as the ratio of government expenditure to GDP). Economic theory suggests that these fundamentals affect the real exchange rate by raising the price of nontradable (NT) goods relative to tradable (T) goods, with increases (declines) in the NT to T ratio producing an appreciation (depreciation) of the real exchange rate. More specifically, an increase (decline) in the terms of trade or a decline (increase) in net foreign liabilities increases (reduces) an economy’s wealth, and puts upward (downward) pressure on the price of NT relative to T due to the more limited supply of NT goods. Higher (lower) government spending usually increases (reduces) the price of NT goods, as government spending tends to be biased toward NT goods.
Given that the EC dollar is a regional currency, regional aggregates are used in estimating the equilibrium real exchange rate, which are weighted by GDP. Data are annual and cover the period 1979–2003. The analysis suggests that a cointegrating (long-run) relationship exists between the real exchange rate and its fundamental determinants, and the signs of the coefficients are consistent with economic theory. The estimated coefficients and the trend component of the fundamentals (derived using the Hodrick-Prescott filter) are then used to generate an estimate of the equilibrium real exchange rate.
The cointegrating equation obtained for the ECCU’s real effective exchange rate (reer) is presented below (LN denotes natural log). The estimated coefficients in the equation below are semi-elasticities for government size and net foreign liabilities (both are measured in percent of GDP) and an elasticity for the terms of trade:
LN(reert) = 2.40 + 1.39*LN(tott) + 0.07*(government sizet) - 0.015*(net foreign liabilities t).
The estimation suggests (see Figure) that the EC dollar went through a period of overvaluation that started in 1998. The recent depreciation of the U.S. dollar against major currencies, and its impact on the real value of the EC dollar (through the EC dollar’s peg to the U.S. dollar) seems to have corrected the previous overvaluation since, as of 2003, the actual real exchange rate is very close to its equilibrium level. A caveat to the analysis is that the long-run determinants of the equilibrium real exchange rate vary from country to country, and for some ECCU countries no long-run relationship was found. However, the results presented serve as an indication of the situation in the region as a whole.
ECCU. Actual and Equilibrium Real Exchange Rates
(Indexes, 1990=100)
Citation: IMF Staff Country Reports 2004, 335; 10.5089/9781451811667.002.A003
Tourism Customer-Based Real Exchange Rates
22. The customer-based real exchange rate indicates in general that the rate has been fairly stable since 1990 (Figure III.6). Partner countries considered are the three main customers of each country’s tourism sector (Table III.2). While the exchange rate for the region as a whole remained fairly stable, there are several country-specific results that are worth noting. First, Antigua and Barbuda and Dominica have become marginally cheaper destinations for their major customers since 1990, although the change is only around 5 percent for the period 1990–2003. Second, Grenada and St. Vincent and the Grenadines have also experienced a depreciation of their customer-based real exchange rates, especially since 1993. This depreciation has been mainly driven by a sustained appreciation of the real exchange rate of Trinidad and Tobago, a key tourism customer of both countries. Third, and in contrast to the other countries, St. Kitts and Nevis and St. Lucia have become more expensive than they were in 1990. The extent of their appreciations is moderate, however, ranging from around 8 percent for St. Kitts and Nevis to around 2 percent for St. Lucia.
ECCU: Major Customers of the Tourism Sector
Denotes major customers of each ECCU country in 2001.
ECCU: Major Customers of the Tourism Sector
| Host | Customers 1/ | Shares in Index (%) |
|---|---|---|
| Antigua and Barbuda | Canada, United Kingdom, United States | 9.1, 48.2, 42.8 |
| Dominica | France, United Kingdom, United States | 11.5, 25.9, 62.6 |
| Grenada | Trinidad and Tobago, United Kingdom, United States | 18.3, 38.4, 43.4 |
| St. Kitts and Nevis | Canada, United Kingdom, United States | 13.2, 22.1, 64.6 |
| St. Lucia | Canada, United Kingdom, United States | 7.2, 39.3, 53.5 |
| St. Vincent and the Grenadines | Trinidad and Tobago, United Kingdom, United States | 19.9, 28.7, 51.4 |
Denotes major customers of each ECCU country in 2001.
ECCU: Major Customers of the Tourism Sector
| Host | Customers 1/ | Shares in Index (%) |
|---|---|---|
| Antigua and Barbuda | Canada, United Kingdom, United States | 9.1, 48.2, 42.8 |
| Dominica | France, United Kingdom, United States | 11.5, 25.9, 62.6 |
| Grenada | Trinidad and Tobago, United Kingdom, United States | 18.3, 38.4, 43.4 |
| St. Kitts and Nevis | Canada, United Kingdom, United States | 13.2, 22.1, 64.6 |
| St. Lucia | Canada, United Kingdom, United States | 7.2, 39.3, 53.5 |
| St. Vincent and the Grenadines | Trinidad and Tobago, United Kingdom, United States | 19.9, 28.7, 51.4 |
Denotes major customers of each ECCU country in 2001.
ECCU: Real Exchange Rates With Respect to Main Customers of the Tourism Sector. 1990–2003 (1990=100) 1/ 2/
Citation: IMF Staff Country Reports 2004, 335; 10.5089/9781451811667.002.A003
Sources: IMF Information Notice System; and Fund staff estimates.1/ Customers-weighted index of nominal exchange rates deflated by seasonally adjusted relative consumer prices. An increase (decrease) indicates an appreciation (depreciation). Excludes Anguilla and Montserrat in the calculation of the ECCU average. Data up to October 2003 for all countries.2/ Customers: Antigua and Barbuda-(Canada, UK, US), Dominica-(France, UK, US), Grenada-(Trinidad and Tobago, UK, US), St. Kits and Nevis-(Canada, U.K., U.S.), St. Lucia-(Canada, U.K., U.S.), St. Vincent and the Grenadines-(Trinidad and Tobago, U.K., U.S.).23. A common phenomenon for all ECCU countries is the continued real depreciation of the customer-based real exchange rates in 2002–03. The main factor underpinning this behavior is the depreciation of the U.S. dollar against other major currencies (for these indexes the relevant currencies are the Canadian dollar, the Euro, and the U.K. pound).
Tourism Competitor-Based Real Exchange Rates
24. On average, the competitor-based real exchange rates depreciated during the 1990s, a trend that was reversed in 2002-03 because of the large real depreciation of the Dominican Republic’s peso (Figure III.7). The competitor-based real exchange rate indexes were constructed using the following countries and weights: The Bahamas (23.4 %), Barbados (8.0 %), Dominican Republic (43.5 %), Jamaica (19.4 %), and Trinidad and Tobago (5.7 %). The weights were chosen based on their relative size (tourism arrivals) in the tourism market in the Caribbean in 2001.
ECCU: Real Exchange Rates With Respect to Tourism Competitors, 1990–2003 1/ 2/
(1990=100)
Citation: IMF Staff Country Reports 2004, 335; 10.5089/9781451811667.002.A003
Sources: IMF Information Notice System; and Fund staff estimates.1/ Competitors-weighted index of nominal exchange rates deflated by seasonally adjusted relative consumer prices. An increase (decrease) indicates an appreciation (depreciation). Excludes Anguilla and Montserrat in the calculation of the ECCU average. Data up to October 2003 for all countries.2/ The competitors with their respective weights in the index are: The Bahamas (23.4 %), Barbados (8.0 %), Dominican Republic (43.5 %), Jamaica (19.4 %), and Trinidad and Tobago (5.7 %). The weights are chosen based on the share of tourism arrivals to the Caribbean in 2001.25. The ECCU countries’ bilateral real exchange rates with The Bahamas, a country with a large share of Caribbean tourism, have depreciated on average since 1990 (Figure III.8). The Bahamas is a good example of a close competitor to the ECCU, as it is a country with a large share of tourism in the Caribbean, and whose market segment is similar to that of the ECCU (the high-income or upper-end segment). In addition, like the ECCU, its nominal exchange rate has been pegged to the U.S. dollar throughout the period, so that real exchange rate movements can be calculated on the basis of changes in relative prices. As of 2003, only St. Kitts and Nevis has experienced an appreciation of its Bahamas-based real exchange rate, while the real exchange rates of other ECCU countries have depreciated. However, the appreciation in St. Kitts and Nevis was moderate, at around 5 percent with respect to 1990. Antigua and Barbuda, Dominica, Grenada, and St. Vincent and the Grenadines have all experienced depreciations of their Bahamas-based real exchange rate with respect to the rates prevailing in 1990. The depreciations in the other countries range from around 3 percent in Grenada to around 7 percent in the remaining three countries.
ECCU: Real Exchange Rates With Respect to The Bahamas, 1990–2003 1/
(1990=100)
Citation: IMF Staff Country Reports 2004, 335; 10.5089/9781451811667.002.A003
Sources: IMF Information Notice System; and Fund staff estimates.1/ Index of nominal exchange rates deflated by seasonally adjusted relative consumer prices. An increase (decrease) indicates an appreciation (depreciation). Excludes Anguilla and Montserrat in the calculation of the ECCU average. Data up to October 2003 for all countries.D. Conclusions
26. The real exchange rates of ECCU countries are (as of September 2003) at about the same level that they were in 1990. At a country-specific level, only St. Kitts and Nevis appears to have lost competitiveness, although the loss is moderate. Antigua and Barbuda and Dominica consistently seem to have gained competitiveness, although the gains also look moderate.
27. At the regional level the ECCU countries are facing two shocks: the depreciation of the U.S. dollar and the depreciation of the Dominican Republic’s peso. The first shock is increasing the attractiveness of ECCU countries to their customers, especially Canada and Europe, while the second is making it more difficult for ECCU countries to compete (on the basis of price) in the tourism market.
28. An analysis of the fundamental determinants of the ECCU equilibrium real exchange rate indicates that the real exchange rate of the EC dollar is neither over- nor undervalued. The empirical analysis suggests that the recent real depreciation of the EC dollar has corrected the real appreciation that occurred during 1998–2001.
References
Goldstein, M., and L. Officer, 1979, “New Measures of Prices and Productivity for Traded and Nontraded Goods,” Review of Income and Wealth, Vol. 25, No. 4.
International Monetary Fund, 2003, “Tourism in the Eastern Caribbean: Meeting the Competitive Threat,” in Eastern Caribbean Currency Union: Selected Issues, Prepared by Anthony Boote and others, International Monetary Fund, Washington, D.C.
Prepared by Paul Cashin, Patrick Njoroge, and Pedro Rodriguez.
The ratio of nontradable to tradable prices is used more frequently when the purpose is to analyze the allocation of resources between the tradable and nontradable sectors; the ratio of domestic to foreign prices is mostly used when the purpose is to assess the competitiveness of one country vis-à-vis another country.
Disaggregated CPI data were obtained from the ECCB for each of the six Fund members of the ECCU, at quarterly frequency for the period 1990:01–2003:02 (data for Antigua and Barbuda were only available for the period 1994:01–2001:04 while data for Dominica start at 1994:01). The typical disaggregation contains the following subgroups, some of which were classified as tradables (food and beverages, clothing and footwear, and furniture and household appliances) and some as nontradables (housing, medical care and health, transportation, education, entertainment and recreation). In 2001, most countries modified the subgroups of the CPI (in most cases to increase the level of disaggregation of some subgroups). The analysis is based on the pre-2001 subgroups and, where necessary, the new subgroups were collapsed into the old ones using their corresponding weights, in order to guarantee homogeneous components throughout the period.
The margins of the distribution sector, and with them the final price of the goods, will vary in the same direction as the price of nontradable goods, given that the distribution sector is a nontradable service. As a consequence, the final price of tradable and nontradable goods may behave very similarly when the distribution costs of the tradable goods are large.
The current and constant price measures of sectoral GDP were obtained from the ECCB for each of the eight members of the ECCU, at annual frequency, for the period 1990–2002.
While the consumer price index focuses on domestic consumption (it includes import prices but excludes export prices), the GDP deflator focuses on domestic production (it includes export prices but excludes import prices).
For this measure, the ECCU average includes Anguilla and Montserrat. The exclusion of these two countries, however, would not change the results much, given their small size relative to the other ECCU countries.
Data on the real effective exchange rate of the six Fund members of the ECCU were taken from the Fund’s Information Notice System (INS) database, at monthly frequency, for the period January 1990 to September 2003. The INS’ monthly data are calculated by interpolating the quarterly CPI data supplied by the ECCB (and monthly nominal exchange rate data). Data for Antigua and Barbuda for 2002 and 2003 are estimates (provided by the database), as the ECCB has not reported CPI data for that country beyond December of 2001.
