A. Developments in 2000

2. In 2000, economic performance weakened. Real GDP growth slowed to 1.7 percent, from over 8 percent a year in 1998–99, reflecting in part a deterioration in the terms of trade, the end of the construction phase of a large foreign direct investment project by INTEL, and the effect of high real interest rates on domestic demand.¹ Inflation remained high at 10 percent during the year, in line with the objectives of credit and exchange rate policy. The decline in the terms of trade contributed to a widening of the external current account deficit and a loss of net international reserves of about US$150 million. At end-2000, gross international reserves amounted to 1.7 months of imports of goods and nonfactor services and 22 percent of broad money, a relatively low level by regional standards.

3. The overall public sector deficit rose from 3.5 percent of GDP in 1999 to 4.2 percent of GDP in 2000, as the primary surplus virtually disappeared.² Net external financing (including a US$250 million Eurobond issue) amounted to about 1½ percent of GDP, with the balance covered through further domestic borrowing; total public debt rose to 48 percent of GDP by end-2000. In the central government, transfers to the private sector declined in relation to GDP, reflecting the policy to phase out the export subsidies on nontraditional exports by 2001. However, while the higher world oil prices raised indirect tax revenues (the import duties and value-added taxes on petroleum products are ad valorem), the earmarking system channeled the additional revenues to public social programs. Central bank operating losses rose to 1.8 percent of GDP, reflecting interest on the large stock of open market instruments that had been issued in recent years. In September the government began to absorb some of the central bank’s losses by repaying part of its debt to the central bank, on which it was not paying interest. The overall surplus of the rest of the public sector fell slightly to 0.6 percent of GDP. This estimate of the enterprise surplus includes an estimated 0.6 percent of GDP of expenditure by a trust fund set up in early 2000 by the state electricity and telecommunications enterprise (ICE).

4. The central bank sought to hold inflation stable at 10 percent a year. The central bank’s net domestic assets contracted slightly in 2000.³ In March 2000, reserve requirements on domestic currency deposits were lowered by 2 percentage points to 12 percent to try to narrow the intermediation spread (the difference between deposit and lending interest rates) and encourage lower interest rates on domestic currency loans. (The reserve requirement on U.S. dollar deposits stayed at 5 percent.) The central bank held the interest rate on its open market instruments steady at 17 percent (for 6-month paper) through September, and then reduced the rate to 15 percent following the repayment by the central government. Lending interest rates in domestic currency declined from 29 percent at end-1999 to 27.8 percent at end-2000, although the intermediation spread held steady at about 13 percent. Exchange rate policy aimed to depreciate the colón to offset the differential between the domestic inflation target and estimated foreign inflation. During 2000, the central bank slowed the rate of crawl to 6.6 percent, from 9.9 percent during 1999, as it expected higher foreign inflation. However, the colón appreciated by 5 percent in real effective terms during the year, owing largely to the unexpected strength of the U.S. dollar (Figure).

5. Bank credit to the private sector rose by 31 percent during the year, led by a 52 percent rise in foreign currency lending. Moreover, a growing share of the foreign currency loans went to mortgage refinancing, consumer credit, and other activities without income in foreign currency. Borrowers were taking advantage of the significantly lower cost of U.S. dollar loans, which carried an interest rate of 12.3 percent on average during 2000 plus expected currency depreciation of about 6½ percent. The rapid credit expansion was financed partly by the use of net foreign assets, as broad money rose by 20 percent during the year. By end-2000, foreign currency loans amounted to 46 percent of total loans, compared with 36 percent at end-1999, while foreign currency deposits rose from 42 percent to 45 percent of broad money. Reported nonperforming loans rose from 2.9 percent of total loans at end-1999 to 3.6 percent of total loans at end-2000, reflecting mostly problem loans at the National Bank (the largest state bank). The risk-weighted capital asset ratio amounted to 19.5 percent at end-2000.

View Full Size

Figure. Costa Rica. Exchange Rate Developments (1990=100)

Citation: IMF Staff Country Reports 2002, 089; 10.5089/9781451809633.002.A001

Source: IMF, Information Notice System.1/ Trade-weighted index of nominal exchange rates deflated by seasonally adjusted relative consumer prices. An increase (decrease) indicates appreciation (depreciation)

• Download Figure
• Download figure as PowerPoint slide

6. The external current account deficit reached almost 5 percent of GDP in 2000, from about 4½ percent of GDP a year earlier, reflecting lower world coffee and banana prices as well as the surge in world oil prices. General merchandise exports in U.S. dollar terms declined for the second consecutive year, as world coffee and banana exports remained weak and the decline in subsidies and the strength of the U.S. dollar affected nontraditional exports. INTEL decided to retool the plant built in 1998–99, which led to almost equal declines in exports of goods for processing and profit remittances. Net capital inflows dropped by over 50 percent, partly because foreign direct investment declined, as the construction of INTEL’S plant ended in 1999. Other private capital inflows also fell sharply, reflecting largely the decline in domestic interest rates, while net borrowing by the public sector remained steady at slightly less than 1 percent of GDP.

7. In the structural area, the assembly approved legislation to open up the telecommunication and electricity sectors to private investment, but subsequently strong popular opposition and an unfavorable supreme court ruling stopped this reform. In early 2000, the superintendent of financial institutions began to require banks to place off-balance-sheet operations on their balance sheets, report regularly on their offshore operations, and apply stricter norms for loan classification. Trade liberalization continued, with a reduction in tariffs on imports from outside Central America in January, the abolition of a small tax on trade with Central America, and the approval of a free trade agreement with Chile.

B. Developments so far in 2001

8. The authorities’ economic program for 2001 (which was published in December 2000)⁴ envisaged a recovery in real GDP growth to 3.2 percent, on the assumption that lower domestic interest rates would stimulate a recovery in investment. Inflation was to be held at 10 percent during the year. The external current account deficit was projected to remain constant in relation to GDP, but with increased net capital inflows, net international reserves would remain constant. The fiscal plan was to reduce the overall public sector deficit in relation to GDP, financed by another international bond issue and domestic borrowing. The government planned to repay the remainder of its debt to the central bank, which would allow the central bank to redeem a significant amount of open market instruments without expanding credit. This policy was expected to permit a reduction in interest rates on open market instruments of about 50 basis points. Consistent with the inflation target, the central bank planned to depreciate the colón by 6.6 percent during the year.

9. So far in 2001, preliminary information suggests that real economic activity declined by about 2 percent (year on year) in the first four months, as INTEL had not yet finished retooling its plant. (Excluding INTEL operations, economic activity rose by an estimated 2 percent year on year in the same period). Price increases in housing and food raised 12-month inflation to 13.3 percent in May. In the first quarter, the central government deficit amounted to 3.2 percent of quarterly GDP, in line with expectations. Although the government repaid the remainder of its debt to the central bank, net domestic assets of the central bank contracted through end-May. The central bank lowered interest rates on its open market instruments by 50 basis points to 14.5 percent in late April, much less than the decline in interest rates on U.S. treasury bills since end-2000, and continued to sell open market securities. Owing to stagnant exports, the external current account deficit was about 3 percent of quarterly GDP in the first quarter, considerably higher than in the same period in 2000. Through end-May, net international reserves rose by US$32 million, benefiting from the proceeds of a US$250 million bond issue by the government in March.

10. With regard to structural reforms, in March the government submitted financial sector legislation to increase the independence of the central bank, strengthen financial supervision, and extend limited deposit insurance to all banks. The proposals for the central bank include steps to lengthen the term of the president from four years to eight years and separate the beginning of the term from the change in government. The legislation seeks to improve financial supervision through several changes, including by requiring annual inspections of financial institutions and allowing the supervisor to share information with his counterparts in other countries. The legislation would establish deposit insurance for all financial institutions. Currently, public banks—which together account for about two-thirds of banking system assets—have a full deposit guarantee, while private banks have no deposit insurance. The insurance would cover deposits up to 2.5 times the previous year’s per capita national income per person, and financial institutions would finance the insurance through premia that varied with each bank’s risk. The legislation would also allow foreign banks to set up branches in Costa Rica (the existing law only allows the establishment of fully owned subsidiaries).

APPENDIX Data Sources

The data for real GDP, investment, and labor force are the IFS and WEO databases. The growth accounting exercise follows the methodology described in Barro and Sala-i-Martin’s Economic Growth, Chapter 10 (1995).⁶ The capital stock in a given year equals the capital stock of the previous year minus depreciation plus investment in the same year, that is, K(t)=K(t-1)^*(l-d)+I(t), where d is the depreciation rate. The depreciation rate is 5 percent, consistent with Elias (1992).⁷ The initial capital stock is determined through the “rough-guess” method suggested by Barro and Sala-i-Martin. The average annual real rate of return of capital is 7 percent, the long-term international average rate of return of capital estimated by Siegel (1998).⁸ The average years of schooling of the workforce are drawn from the Barro and Lee database. Human capital is estimated in terms of average years of schooling following the standard definition used by Lucas (1988).⁹

A. Introduction

17. Costa Rica has been following an exchange rate policy aimed at maintaining the external competitiveness of domestic goods through daily adjustments of the exchange rate on the basis of the differential between inflation in Costa Rica and its main trading partners. More specifically, the Central Bank of Costa Rica (CBCR) manages the exchange rate with the goal of keeping the real effective exchange rate (REER) broadly stable around its level in 1997, when macroeconomic conditions were deemed by and large satisfactory and the level of competitiveness adequate. However, the poor performance of nontraditional exports in recent years raises questions about the quality of the method used to measure the REER. The CBCR index is actually based upon Costa Rica’s bilateral trade flows, and misses the effect of increased competition from third-countries, especially Asian countries, in Costa Rica’s main export markets.

18. This paper develops a new REER index that takes into account third-country competition and allows competitiveness weights to change over time. In particular, the method used to calculate competitiveness weights confirms that competition from a number of Asian countries has intensified over time. The results show that, although the REER index calculated by the CBCR broadly follows the path of the proposed index as well as that calculated by the IMF, it may underestimate the magnitude of appreciation of the REER starting at the end of 1999 as indicated by the other two indices.

19. The paper is organized as follows: Section B provides a brief overview of some methodological difficulties usually encountered in calculating a country’s real effective exchange rate. Section C reviews the approach used by the CBCR¹¹ and by the IMF¹² to calculate Costa Rica’s REER. Section D presents the alternative methodology. Section E discusses the results and draws conclusions.

B. Some Methodological Issues

20. The indices of the nominal and real effective exchange rate are important tools for exchange rate policy. The former, calculated as a weighted average of bilateral exchange rates of a country’s currency with respect to its main trading partners, provides a summary indicator of a currency’s external value. The latter, by comparing price developments in a country with those in its trading partners, offers an assessment of the evolution of a country’s competitiveness in international markets. Although these indices are common, their actual calculation is not straightforward and poses both conceptual and empirical difficulties. As emphasized by Hinkle and Montiel (1999), there are several conceptual definitions of the real exchange rate (RER) depending on the underlying economic model.¹³ In addition, the empirical measurement of the RER in many developing countries is made more complex by a number of practical problems that are not often encountered in the case of industrial countries (e.g., lack of data, multiple exchange rates, unrecorded or misreported trade flows).

21. In order to calculate a country’s REER, it is necessary to have four elements: (1) an operational formula; (2) nominal exchange rates;¹⁴ (3) a country-weight scheme; and (4) appropriate price or cost indices. Among these four elements, the latter two present a number of empirical problems.

22. In constructing an index of a country’s multilateral or effective RER (REER), the choice of the weighting scheme determines how developments in exchange rates and prices in different partner-countries affect the competitive position of the home country. Without considering more general problems associated with unrecorded or misreported trade, the choice of competitiveness weights may be complicated by a number of factors. For instance, the home country’s geographical trade patterns may differ significantly between exports and imports. In this case, there may be reasons to calculate separate REER indices for imports and exports rather than averaging them in a single indicator. Further complications may arise when home-country’s exporters face significant competition from producers in third countries with which home country’s bilateral trade is negligible. Ideally, competitiveness weights should take into account not only the effect of direct trading partners but also the impact of third countries with which home-country exporters compete in their outlet markets. Finally, home country’s trading patterns may change over time. In this case, fixed weight averages become less representative and it would be appropriate to update periodically the set of main trading partners as well as the weights attached to them.

23. As far as the choice of price or cost indices is concerned, it is preferable to use either producer prices or unit labor costs or export unit values. Since these indices are not always available in the case of developing countries, consumer price indices are usually used as a proxy. However, this option has at least two important drawbacks (Heinkel and Nsengiyumva, 1999). First, CPIs for different countries are not based on the same (or even comparable) basket of goods. This implies that bilateral exchange rates cannot be uniquely determined as the relative cost of two baskets of goods since their composition may differ significantly. Second, CPI developments may be affected by the existence of price controls and subsidies as well as by changes in indirect taxes. This makes it more difficult to disentangle the real forces driving changes in a country’s CPI over time.

C. A Comparison Between Indices

24. As stated above, the two available indices of Costa Rica’s REER are the one calculated by the CBCR and the one estimated by the IMF. The differences between the CBCR and the Fund methodology are: (a) the weighting scheme; (b) the set of trading partners; and (c) the operational formula.

Selection of comparator countries

31. The REER index calculated by the CBCR includes 19 countries (Table 1) that were chosen on the basis of three general criteria: (1) data availability; (2) relative importance in Costa Rica’s external trade; and (3) the share of Costa Rica’s total trade represented by the selected set of trade partners. In the case of the REER index calculated by the LMF, Costa Rica’s set of comparator countries is limited to those whose total competitiveness weights are greater than 1 percent (Table 2).

Table 1.

Country Set in the CBCR REER Index (ITCER-97)

Table 1.

Country Set in the CBCR REER Index (ITCER-97)

Country		1992	1995–1996	1997
North America		53.9	50.6	47.7
	United States	52.2	48.9	45.8
	Canada	1.7	1.7	1.9
Latin America		18.7	23.6	26.3
	Mexico	2.7	4.9	6.3
	Guatemala	5.1	4.7	5.0
	El Salvador	3.5	3.7	3.9
	Panama	3.3	3.4	3.0
	Nicaragua		2.2	2.5
	Venezuela	1.5	1.4	2.0
	Colombia	1.1	1.6	1.9
	Honduras	1.5	1.7	1.7
Europe		22.2	22.5	21.2
	Germany	8.2	5.6	5.4
	Italy	3.9	4.0	3.4
	Netherlands	1.8	2.7	2.6
	Belgium	3.1	3.3	2.5
	France	1.6	2.1	2.1
	Spain	1.5	1.6	2.0
	United Kingdom	1.6	2.3	1.9
	Sweden	0.5	0.9	1.3
Asia		5.2	3.4	4.9
	Japan	5.2	3.4	4.9
Total		100.0	100.0	100.0

View Table

Table 1.

Country Set in the CBCR REER Index (ITCER-97)

Country		1992	1995–1996	1997
North America		53.9	50.6	47.7
	United States	52.2	48.9	45.8
	Canada	1.7	1.7	1.9
Latin America		18.7	23.6	26.3
	Mexico	2.7	4.9	6.3
	Guatemala	5.1	4.7	5.0
	El Salvador	3.5	3.7	3.9
	Panama	3.3	3.4	3.0
	Nicaragua		2.2	2.5
	Venezuela	1.5	1.4	2.0
	Colombia	1.1	1.6	1.9
	Honduras	1.5	1.7	1.7
Europe		22.2	22.5	21.2
	Germany	8.2	5.6	5.4
	Italy	3.9	4.0	3.4
	Netherlands	1.8	2.7	2.6
	Belgium	3.1	3.3	2.5
	France	1.6	2.1	2.1
	Spain	1.5	1.6	2.0
	United Kingdom	1.6	2.3	1.9
	Sweden	0.5	0.9	1.3
Asia		5.2	3.4	4.9
	Japan	5.2	3.4	4.9
Total		100.0	100.0	100.0

Table 2.

Country Set in the IMF REER Index

Table 2.

Country Set in the IMF REER Index

Country	1988–1990
North America	43.47
United States	40.03
Canada	3.42
Latin America	18.77
Panama	6.85
Brazil	3.61
Guatemala	2.52
Mexico	2.46
El Salvador	1.94
Venezuela	1.39
Europe	26.72
Germany	7.18
France	4.26
Netherlands	3.79
United Kingdom	3.63
Italy	3.55
Spain	2.39
Belgium	1.92
Asia	11.07
Japan	7.17
Korea	2.03
Taiwan Province of China	1.87
Total	100.00

View Table

Table 2.

Country Set in the IMF REER Index

Country	1988–1990
North America	43.47
United States	40.03
Canada	3.42
Latin America	18.77
Panama	6.85
Brazil	3.61
Guatemala	2.52
Mexico	2.46
El Salvador	1.94
Venezuela	1.39
Europe	26.72
Germany	7.18
France	4.26
Netherlands	3.79
United Kingdom	3.63
Italy	3.55
Spain	2.39
Belgium	1.92
Asia	11.07
Japan	7.17
Korea	2.03
Taiwan Province of China	1.87
Total	100.00

32. There are some important differences in the comparator countries of the two indices. Brazil, Korea, and Taiwan Province of China are included in the IMF basket only, while the CBCR’s set of competitor countries gives more weight to Latin American countries, by including Colombia, Honduras, and Nicaragua.

Operational formula

33. The CBCR computes its REER index applying a “chain rule”. In this case, the change in the REER index between time t-1 and time t is computed as a weighted geometric average of the changes in the home country’s consumer prices relative to that in its main trading partners. Therefore, a change in the competitiveness index between two nonconsecutive years is equal to the sum of the changes in the competitiveness index in each consecutive year calculated with the weights pertaining to the corresponding year.¹⁹ This, in turn, allows for possible changes in the set of competitor countries as well as in their respective weights in order to better reflect developments in Costa Rica’s trading patterns.

$\begin{array}{cc}\left(5\right)& \mathrm{R}\mathrm{E}\mathrm{E}{\mathrm{R}}^{\mathrm{C}\mathrm{B}\mathrm{C}\mathrm{R}}\left(\mathrm{t}\right)=\frac{\left\{{\mathrm{\Pi }}_{\mathrm{j}}\right[{\mathrm{P}}_{\mathrm{j}}\left(\mathrm{t}\right)/{\mathrm{P}}_{\mathrm{j}}(\mathrm{t}-1){]}^{\mathrm{W}\mathrm{j}}*{\mathrm{\Pi }}_{\mathrm{j}}[{\mathrm{E}}_{\mathrm{j}}\left(\mathrm{t}\right)/{\mathrm{E}}_{\mathrm{j}}(\mathrm{t}-1){]}^{\mathrm{W}\mathrm{j}}]\}}{[{\mathrm{P}}^{\mathrm{C}\mathrm{R}}\left(\mathrm{t}\right)/{\mathrm{P}}^{\mathrm{C}\mathrm{R}}(\mathrm{t}-1)][{\mathrm{E}}^{\mathrm{C}\mathrm{R}}\left(\mathrm{t}\right)/{\mathrm{E}}^{\mathrm{C}\mathrm{R}}(\mathrm{t}-1)]}*\mathrm{R}\mathrm{E}\mathrm{E}{\mathrm{R}}^{\mathrm{C}\mathrm{B}\mathrm{C}\mathrm{R}}(\mathrm{t}-1)\end{array}$

Given the formula, an increase (decrease) in the REER^cbcr indicates a depreciation (appreciation). Therefore, in Figure 1, where the REER indices are compared, the inverse of the index is plotted.

View Full Size

Costa Rica; Real Effective Exchange Rates Indexes, 1995–2001 ^1/

(Monthly Data, 1997 = 100)

Citation: IMF Staff Country Reports 2002, 089; 10.5089/9781451809633.002.A001

Sources: Central Bank of Costa Rica; IMF, Information Notice System; IMF, Direction of Trade: and IMF staff estimates.^1/ An increase in the index denotes an appreciation

• Download Figure
• Download figure as PowerPoint slide

34. On the contrary, in the case of the IMF indicator, the REER index is computed as a weighted geometric average of the level of consumer prices in the home country relative to that in its main trading partners:

$\begin{array}{cc}\left(4\right)& \mathrm{R}\mathrm{E}\mathrm{E}{\mathrm{R}}^{\mathrm{I}\mathrm{M}\mathrm{F}}\left(\mathrm{t}\right)={\mathrm{\Pi }}_{\mathrm{j}}\ne \mathrm{i}[{\mathrm{P}}^{\mathrm{C}\mathrm{R}}\left(\mathrm{t}\right){\mathrm{E}}^{\mathrm{C}\mathrm{R}}\left(\mathrm{t}\right)/{\mathrm{P}}_{\mathrm{j}}\left(\mathrm{t}\right){\mathrm{E}}_{\mathrm{j}}\left(\mathrm{t}\right){]}^{\mathrm{W}\mathrm{j}}\end{array}$

where: P_j(t) = consumer price index of trading partner j at time t;

E_j(t) = exchange rate of trading partner j vis-à-vis the U.S. dollar at time t;

p^cr(t) = consumer price index of Costa Rica at time t; and

E^cr(t) = exchange rate of Costa Rica vis-à-vis the U.S. dollar at time t.

Therefore, an increase in the index indicates an appreciation of the REER. Although the IMF index has the advantage of applying a more detailed and accurate methodology to calculate the competitiveness weights, it has the shortcoming of computing those weights for a determined reference period, namely 1988–90. Although such a choice was at the time a compromise between use of up-to-date and comprehensive information, it is unavoidably arbitrary. Hence the use of fixed weights represents an important weakness of the IMF methodology especially in the case where a country’s geographical composition of external trade changes significantly over the medium term.

D. The Proposed Methodology

35. The present exercise tries to combine the more complex methodology used by the IMF in computing competitiveness weights with the operational formula used by the CBCR. However, given the lack of updated information on the commodity breakdown of trade flows and on tourism services, total competitiveness weights are computed as weighted averages of export and import weights using data on trade of goods only from the IMF—Direction of Trade data bank.²⁰

36. The general formula of the overall competitiveness weight of country-j is the following:

where:

ω_j(t) = total weight attached to country-j by Costa Rica;$\mathrm{\alpha }\left(\mathrm{t}\right)=\frac{{\mathrm{X}}^{\mathrm{C}\mathrm{R}}\left(\mathrm{t}\right)}{{\mathrm{X}}^{\mathrm{C}\mathrm{R}}\left(\mathrm{t}\right)+{\mathrm{M}}^{\mathrm{C}\mathrm{R}}\left(\mathrm{t}\right)}=$ export share of total trade of Costa Rica;

θ_j(t) = export weight attached to country-j by Costa Rica;${\mathrm{\mu }}_{\mathrm{j}}\mathrm{\left(}\mathrm{t}\mathrm{\right)}\mathrm{=}\frac{{{\mathrm{M}}_{}^{\mathrm{C}\mathrm{R}}}_{\mathrm{j}}\mathrm{\left(}\mathrm{t}\mathrm{\right)}}{{\mathrm{\Sigma }}_{\mathrm{j}}{{\mathrm{M}}_{}^{\mathrm{C}\mathrm{R}}}_{\mathrm{j}}\mathrm{\left(}\mathrm{t}\mathrm{\right)}}\mathrm{=}$ import weight attached to country-j by Costa Rica = share of Costa Rica’s import from country-j.

It is worth noting that all the variables are computed as three-year moving averages in order to smooth possible shocks in trade patterns.

37. The export weight attached to country-j by Costa Rica is calculated as a linear combination of Costa Rica’s bilateral trade with country-j and the competition from country-j’s exporters on third markets. For the sake of simplicity, only four markets are considered: the United States, Canada, the Euro area, and the United Kingdom. In the 1990s, about three quarters of Costa Rica’s exports were indeed directed to those markets. Similarly to the IMF methodology, competition in third markets is calculated as the product between the share of total exports to market-k stemming from country-j and Costa Rica’s export share to market-k:

${\mathrm{\beta }}_{\mathrm{j}}\left(\mathrm{t}\right)=\frac{{\mathrm{X}}_{\mathrm{j}}^{\mathrm{C}\mathrm{R}}\left(\mathrm{t}\right)}{{\mathrm{\Sigma }}_{\mathrm{j}}{\mathrm{X}}_{\mathrm{j}}^{\mathrm{C}\mathrm{R}}\left(\mathrm{t}\right)}=$ Costa Rica’s export share to country-j

where: k = the United States, Canada, the Euro area, and the United Kingdom. The parameters ν_j(t) are scaled so that they add up to 1.

38. Changes in the index of Costa Rica’s nominal effective exchange rate are therefore calculated by applying the following chain rule:

E^cr_j(t) = country-j’s exchange rate index defined as the amount of colones per unit of country-j’s currency.

where the changes in country-j’s exchange rate index between t and t-1 are weighted by the average of the weights attached to country-j in the two years.

39. Changes in the index of Costa Rica’s real effective exchange rate are therefore calculated as follows:

Trade partner selection

40. In order to take into account differences in Costa Rica’s trading patterns between exports and imports, a double threshold was used in the selection of Costa Rica’s trade partners. Countries were chosen if (1) their overall competitiveness weights (ω_j(t)) were equal or greater than 1 percent, or, at least, (2) their export weights (θ_j(t)) were greater than 1 percent.²¹ For instance, Brazil, Colombia, and the Philippines are not very significant export outlets for Costa Rica (less than 1 percent of total exports is directed to these markets, respectively) but they are (or have become over time) important sources of Costa Rica’s imports. On the contrary, the overall competitiveness weight of Sweden and Switzerland have fallen below the 1 percent threshold over time while their respective export weights have continued to satisfy the second criterion. To avoid changing the composition of the set of comparator countries too often that would introduce some noise in the calculation of the REER, countries that satisfy one of the previous two criteria irregularly (e.g., out-in-out) and for brief periods of time were excluded (Table 3).²² Furthermore, Venezuela was dropped from the group of competitor countries since the bulk of Costa Rica’s imports from this country is represented by oil imports. Once the set of countries is selected their respective total competitiveness weights are scaled so that they add up to 1.

Table 3.

Selection of Costa Rica’s Main Trading Partners

Table 3.

Selection of Costa Rica’s Main Trading Partners

Countries that Satisfy Criteria (1)	Period	Countries that Satisfy Criteria (2)	Period
United States	1990–2000	United States	1990–2000
Canada	1990–2000	Canada	1990–2000
Japan	1990–2000	Japan	1990–2000
Euro area	1990–2000	Euro area	1990–2000
Sweden	1990	Sweden	1990–2000
Switzerland	1990–1997	Switzerland	1990–2000
People’s Republic		People’s Republic
of China	1993–2000	of China	1990–2000
Korea	1990–1996	Korea	1990–2000
Malaysia	1999–2000	Malaysia	1998–2000
		Singapore	1995–2000
Brazil	1990–2000
Colombia	1992–2000
El Salvador	1990–2000	El Salvador	1990–2000
Guatemala	1990–2000
Mexico	1990–2000	Mexico	1990–2000
Nicaragua	1992–1997	Nicaragua	1991–2000
Panama	1990–2000	Panama	1990–1999

View Table

Table 3.

Selection of Costa Rica’s Main Trading Partners

Countries that Satisfy Criteria (1)	Period	Countries that Satisfy Criteria (2)	Period
United States	1990–2000	United States	1990–2000
Canada	1990–2000	Canada	1990–2000
Japan	1990–2000	Japan	1990–2000
Euro area	1990–2000	Euro area	1990–2000
Sweden	1990	Sweden	1990–2000
Switzerland	1990–1997	Switzerland	1990–2000
People’s Republic		People’s Republic
of China	1993–2000	of China	1990–2000
Korea	1990–1996	Korea	1990–2000
Malaysia	1999–2000	Malaysia	1998–2000
		Singapore	1995–2000
Brazil	1990–2000
Colombia	1992–2000
El Salvador	1990–2000	El Salvador	1990–2000
Guatemala	1990–2000
Mexico	1990–2000	Mexico	1990–2000
Nicaragua	1992–1997	Nicaragua	1991–2000
Panama	1990–2000	Panama	1990–1999

41. By comparing Table 1, 2, and 4, it is worth noting that the proposed REER index takes into account a larger number of competitor countries than the previous two indices. The number of countries selected goes from 15 in 1990 to 20 in 2000. In particular, while the calculated REER index gives less weight to Latin American countries than the CBCR and the IMF indices, it gives more weight to Asian countries, reflecting growing bilateral trade between Costa Rica and those countries.

Table 4.

Costa Rica: Alternative REER Competitiveness Weights

Table 4.

Costa Rica: Alternative REER Competitiveness Weights

Country	1990	1991	1992	1993	1994	1995	1996	1997	1998	1999	2000
North America	46.74	47.35	48.62	50.03	49.66	48.04	46.77	48.52	50.71	49.37	47.18
United Stales	41.66	42.31	44.09	45.24	45.06	43.34	42.42	43.70	45.69	44.25	42.22
Canada	5.08	5.04	4.53	4.79	4.60	4.69	4.36	4.82	5.03	5.13	4.96
Latin America	16.13	16.82	17.57	17.26	17.75	18.42	20.05	19.46	17.39	15.42	14.81
Mexico	4.52	4.34	4.17	4.09	4.18	4.43	5.43	6.12	6.18	5.82	5.64
Guatemala	4.03	3.89	3.62	3.17	3.18	3.21	3.48	3.24	2.82	2.45	2.32
El Salvador	2.56	2.52	2.36	2.18	2.23	2.36	2.60	2.52	2.17	1.85	1.72
Panama	2.27	2.33	1.94	1.67	1.71	2.19	2.58	2.36	1.90	1.55	1.48
Colombia	…	…	1.29	1.67	1.90	2.09	2.27	2.17	1.77	1.46	1.34
Brazil	2.76	2.62	2.64	2.68	2.83	2.69	2.42	1.82	1.42	1.26	1.33
Nicaragua	…	1.13	1.55	1.80	1.72	1.44	1.25	1.23	1.13	1.02	0.97
Europe	23.88	23.01	21.35	20.57	20.53	21.55	21.91	20.86	19.34	19.70	20.80
Euro area	17.31	16.71	15.56	15.21	15.23	15.93	15.91	15.08	13.85	13.79	14.36
United Kingdom	3.76	3.58	3.24	3.01	3.05	3.33	3.63	3.51	3.41	3.93	4.48
Switzerland	1.55	1.59	1.58	1.51	1.43	1.40	1.33	1.19	1.03	1.00	1.02
Sweden	1.26	1.12	0.97	0.84	0.83	0.89	1.04	1.08	1.06	0.97	0.94
Asia	13.25	12.81	12.45	12.15	12.06	12.00	11.26	11.17	12.55	15.52	17.21
Japan	9.41	9.06	8.57	8.30	8.02	7.42	6.53	6.51	6.95	6.98	6.60
People’s Republic of China	0.68	0.82	0.92	1.18	1.42	1.77	1.89	2.16	2.34	2.61	2.77
Hong Kong SAR	0.98	0.90	0.97	1.05	1.01	0.99	0.99	0.95	0.92	0.88	0.87
Malaysia	…	…	…	…	…	…	…	…	0.82	1.76	2.43
Philippines	…	…	…	…	…	…	…	…	…	1.35	2.17
Korea	2.18	2.03	1.99	1.62	1.61	1.21	1.21	0.86	0.78	1.19	1.67
Singapore	…	…	…	…	…	0.61	0.64	0.70	0.76	0.74	0.70

View Table

Table 4.

Costa Rica: Alternative REER Competitiveness Weights

Country	1990	1991	1992	1993	1994	1995	1996	1997	1998	1999	2000
North America	46.74	47.35	48.62	50.03	49.66	48.04	46.77	48.52	50.71	49.37	47.18
United Stales	41.66	42.31	44.09	45.24	45.06	43.34	42.42	43.70	45.69	44.25	42.22
Canada	5.08	5.04	4.53	4.79	4.60	4.69	4.36	4.82	5.03	5.13	4.96
Latin America	16.13	16.82	17.57	17.26	17.75	18.42	20.05	19.46	17.39	15.42	14.81
Mexico	4.52	4.34	4.17	4.09	4.18	4.43	5.43	6.12	6.18	5.82	5.64
Guatemala	4.03	3.89	3.62	3.17	3.18	3.21	3.48	3.24	2.82	2.45	2.32
El Salvador	2.56	2.52	2.36	2.18	2.23	2.36	2.60	2.52	2.17	1.85	1.72
Panama	2.27	2.33	1.94	1.67	1.71	2.19	2.58	2.36	1.90	1.55	1.48
Colombia	…	…	1.29	1.67	1.90	2.09	2.27	2.17	1.77	1.46	1.34
Brazil	2.76	2.62	2.64	2.68	2.83	2.69	2.42	1.82	1.42	1.26	1.33
Nicaragua	…	1.13	1.55	1.80	1.72	1.44	1.25	1.23	1.13	1.02	0.97
Europe	23.88	23.01	21.35	20.57	20.53	21.55	21.91	20.86	19.34	19.70	20.80
Euro area	17.31	16.71	15.56	15.21	15.23	15.93	15.91	15.08	13.85	13.79	14.36
United Kingdom	3.76	3.58	3.24	3.01	3.05	3.33	3.63	3.51	3.41	3.93	4.48
Switzerland	1.55	1.59	1.58	1.51	1.43	1.40	1.33	1.19	1.03	1.00	1.02
Sweden	1.26	1.12	0.97	0.84	0.83	0.89	1.04	1.08	1.06	0.97	0.94
Asia	13.25	12.81	12.45	12.15	12.06	12.00	11.26	11.17	12.55	15.52	17.21
Japan	9.41	9.06	8.57	8.30	8.02	7.42	6.53	6.51	6.95	6.98	6.60
People’s Republic of China	0.68	0.82	0.92	1.18	1.42	1.77	1.89	2.16	2.34	2.61	2.77
Hong Kong SAR	0.98	0.90	0.97	1.05	1.01	0.99	0.99	0.95	0.92	0.88	0.87
Malaysia	…	…	…	…	…	…	…	…	0.82	1.76	2.43
Philippines	…	…	…	…	…	…	…	…	…	1.35	2.17
Korea	2.18	2.03	1.99	1.62	1.61	1.21	1.21	0.86	0.78	1.19	1.67
Singapore	…	…	…	…	…	0.61	0.64	0.70	0.76	0.74	0.70

42. In the calculations the People’s Republic of China and Hong Kong SAR have been considered as only one country for the whole period²³ although the respective exchange rates and prices have been used for the calculation. Similarly, the countries participating in the Euro area have been considered as only one country and the exchange rate and price data for the area as a whole (provided by the INS database) have been used for the calculation.

E. Results and Conclusions

43. The alternative REER for Costa Rica was calculated for the period 1995–2001 using monthly data. As shown in Figure 1, where the proposed REER index (labeled as “Alternative”) is compared with those calculated by the IMF and by the CBCR, the proposed methodology succeeds in tracking sufficiently well the IMF REER. Small differences between the two indices emerge at the beginning and at the end of the period considered, reflecting some differences in the two indices of nominal effective exchange rate (1995–96 and 2000-01) and relative prices (1999–2001). The main reasons for this can be ascribed to the lower weight attributed in the alternative index to the countries of the Euro area, and to Europe in general,²⁴ as well as at the higher (and increasing over time) importance of Asian countries among Costa Rica’s trade partners.

44. Starting from the end of 1999, while the CBCR index shows a broadly unchanged competitiveness position of Costa Rica, both the IMF and the “Alternative” indices point to a more marked appreciation of the REER.

Table 1.

Costa Rica: National Income Accounts

Sources: Central Bank of Costa Rica; and Fund staff estimates.

Table 1.

Costa Rica: National Income Accounts

		1996	1997	1998	1999	Prel. 2000
(In billions of colones)
Consumption		2,151	2,562	3,010	3,456	3,968
Private sector		1,821	2,172	2,541	2,891	3,313
Public sector		330	390	469	565	655
Gross domestic investment		391	538	709	735	830
Fixed capital formation		413	530	731	789	838
	Private sector	342	434	605	639	667
	Public sector	71	97	127	150	171
Change in inventories		-22	8	-22	-54	-8
Gross domestic expenditure		2,541	3,100	3,719	4,191	4,798
Resource gap		-86	-117	-95	276	66
Exports of goods and nonfactor services		967	1,213	1,714	2,329	2,340
Imports of goods and nonfactor services		-1,053	-1,331	-1,809	-2,052	-2,273
GDP at market prices		2,456	2,982	3,624	4,467	4,864
(In billions of 1991 colones)
Consumption		985	1,035	1,089	1,110	1,135
Private sector		857	901	953	971	993
Public sector		128	133	136	139	142
Gross domestic investment		200	250	312	270	259
Fixed capital formation		208	240	302	294	282
	Private sector	172	196	250	238	224
	Public sector	36	44	52	56	58
Change in inventories		-8	10	10	-24	-23
Gross domestic expenditure		1,185	1,285	1,401	1,380	1,394
Resource gap		-63	-98	-116	12	22
Exports of goods and nonfactor services		455	494	624	761	750
Imports of goods and nonfactor services		-518	-592	-740	-750	-728
GDP at market prices		1,122	1,186	1,285	1,392	1,416

Sources: Central Bank of Costa Rica; and Fund staff estimates.

View Table

Table 1.

Costa Rica: National Income Accounts

		1996	1997	1998	1999	Prel. 2000
(In billions of colones)
Consumption		2,151	2,562	3,010	3,456	3,968
Private sector		1,821	2,172	2,541	2,891	3,313
Public sector		330	390	469	565	655
Gross domestic investment		391	538	709	735	830
Fixed capital formation		413	530	731	789	838
	Private sector	342	434	605	639	667
	Public sector	71	97	127	150	171
Change in inventories		-22	8	-22	-54	-8
Gross domestic expenditure		2,541	3,100	3,719	4,191	4,798
Resource gap		-86	-117	-95	276	66
Exports of goods and nonfactor services		967	1,213	1,714	2,329	2,340
Imports of goods and nonfactor services		-1,053	-1,331	-1,809	-2,052	-2,273
GDP at market prices		2,456	2,982	3,624	4,467	4,864
(In billions of 1991 colones)
Consumption		985	1,035	1,089	1,110	1,135
Private sector		857	901	953	971	993
Public sector		128	133	136	139	142
Gross domestic investment		200	250	312	270	259
Fixed capital formation		208	240	302	294	282
	Private sector	172	196	250	238	224
	Public sector	36	44	52	56	58
Change in inventories		-8	10	10	-24	-23
Gross domestic expenditure		1,185	1,285	1,401	1,380	1,394
Resource gap		-63	-98	-116	12	22
Exports of goods and nonfactor services		455	494	624	761	750
Imports of goods and nonfactor services		-518	-592	-740	-750	-728
GDP at market prices		1,122	1,186	1,285	1,392	1,416

Sources: Central Bank of Costa Rica; and Fund staff estimates.