– This Technical Assistance Report paper on St Lucia was prepared by a staff team of the International Monetary Fund. It is based on the information available at the time it was completed in March 2022.
USA / ST LUCIA – The following is a summary of mission outcomes and priority recommendations:
- A remote technical assistance (TA) mission on external sector statistics (ESS) was conducted for the Central Statistical Office (CSO) of Saint Lucia during March 22–26, 2021. The mission was part of the Caribbean Regional Technical Assistance Centre (CARTAC) work program on External Sector Statistics (ESS).
2. The mission assisted in addressing data collection issues, with emphasis on those related to the COVID-19 pandemic. An official from the Eastern Caribbean Central Bank (ECCB) participated during one working day. The main outcomes of the mission were: (i) assessing available source data and main changes in the collection system affected by the pandemic; (ii) providing further guidance to estimate nonresponse and missing data; and (iii) assessing available monthly and quarterly data and their timeliness with the support of the ECCB.
3. The COVID-19 outbreak partially affected balance of payments administrative data sources (e.g., trade, number of tourism visitors, remittances, government transfers, financial statements from enterprises) due delays stemming from work from home circumstances. Tourism-related stayover expenditure surveys applied by the Tourism Board of Saint Lucia were not possible because of the inability to conduct in-person interviews. The response rate to balance of payments surveys was lower than 50 percent and around 30 percent for the hotels sector. Even though the CSO processes the data received through surveys, the CSO does not make estimations for missing data or low response rate. For that reason, the ECCB must make those estimations as part of their revision process. To avoid delaying the compilation process the TA mission recommended to designate officials for the validation process of balance of payments surveys and for the follow up of the data.
4. Once a year, the ECCB works directly with CSO officials cross-checking the data and contacting key respondents as needed. The mission provided guidance to the CSO compilers on the validation procedures provided by the sixth edition of the IMF Balance of Payments and International Investment Position Manual Compilation Guide (BPM6CG). It also proposed estimation methods to consider the relative importance of the enterprises that did not respond (compared with those that responded), sector activity, and sample expansion (where possible). It also recommended a permanent validation process, since the sooner an inconsistency is detected the easier it would be to receive a clarification or correction from the respondent.
5. The business registry needs to be updated to provide a better framework to improve the grossing up ratios employed by the ECCB. Recommendations were provided to the CSO on the useful sources (current registers for National Accounts focusing on key respondents, taxation records, information from Invest Saint Lucia, the official investment promotion agency, and media reports) for this purpose.
6. The mission found that the CSO’s annual report continue disseminating exports data that are inconsistent with those included in the balance of payments by the ECCB, provided by the CSO with adjustments. During a previous TA mission (February 2018) it was recommended to review re-exports of aviation fuel that did not have their corresponding precedent imports and the export values of alcoholic beverages, both improving artificially the trade balance. In addition, adjustments were recommended for balance of payments purposes. The mission recommended to regularly assess data sources (for coverage or response errors) and to validate these data from Saint Lucia Customs and Excise Department (CED) against direct information provided by enterprises and to disseminate export statistics consistent with those provided to the ECCB for balance of payments purposes.
7. On the production of flash indicators in advance of the release of preliminary balance of payments data, some variables (e.g., monthly trade balance, monthly number of tourism visitor arrivals) are available. It must be mentioned that during the March 2021 TA mission to the ECCB—which compiles balance of payments and international investment position (IIP) for the eight Eastern Caribbean Currency Union (ECCU) member countries—it was remarked that even though some monthly or quarterly variables may be available, they would not make a commitment on their prompt dissemination as they will first define the work program of the new Balance of Payments Unit.
8. To support progress in the above areas, the mission recommended an action plan with the following priority recommendations (Table 1).
A. Action plan
9. The below action plan includes steps to be undertaken by the CSO to accomplish relevant milestones. Actions are formulated as high (H) and priority recommendations (PR). Appendix I provides the assessment on the progress made in the implementation of the previous mission’s action plan.
B. Source data and main changes in the collection system due to COVID-19 pandemic
10. The COVID-19 outbreak partially affected balance of payments administrative data sources (e.g., trade, number of tourism visitors, remittances, government transfers, financial statements from enterprises) due delays stemming from work from home circumstances. Tourism-related stayover expenditure surveys applied by the Tourism Board of Saint Lucia were not possible because of the inability to conduct in-person interviews.
11. Data source for goods was assessed. Exports include locally produced goods exports, exports from free zones, and re-exports from free zones and warehouses. Total exports and imports are provided by the CED the customs office for trade in goods data (including the information of the free trade areas). The frequency of the data received is monthly and the timeliness is 6-8 weeks after the end of the reference month. There have been some lags since March 2020 onwards, but the data was received.
12. The mission reviewed the published information of trade data from the annual report of the CSO and observed there are some differences with exports published by the ECCB. The CSO provides trade statistics with adjustments to the ECCB for balance of payments purposes, nevertheless, it is disseminating exports data in its annual report without those adjustments. During a previous TA mission (2018) two issues were observed:
a. Re-exports of aviation fuel did not have their corresponding precedent imports records in the CED database. There is an inconsistency that is improving the trade balance.
b. Export values of alcoholic beverages were too high according to the Trade Export Promotion Agency.
13. Recommendations were provided to follow the guidelines for the accuracy and reliability of the data. Even though the collection of exports and imports of goods employs merchandise trade statistics from the CED, it is important to assess source data regularly (coverage, response errors). Validation against the direct information provided by enterprises for significant products would be a good practice.
14. It is a good practice to produce a reconciliation table of the differences between merchandise trade statistics and goods on the balance of payments. Table 10.2 of BPM6 show a sample of the adjustments that balance of payments compilers may need to do to record total goods on a balance of payments basis.
15. To comply with the serviceability of the data, it is important that the statistics are consistent with that obtained through other data sources or statistical frameworks. This would be the case for the value of exports disseminated in the CSO Annual Report and the information published by the ECCB. Also, regarding the serviceability of the data, the mission observed that the Table 11.4 of the annual report does not correspond to the balance of payments of Saint Lucia but is the consolidated balance of payments of the ECCU member countries. It would be important to substitute the table or clarify the title of the table.
16. Due to COVID-19 pandemic, the main challenge is the inability to conduct in-person interviews. For 2020, the tourism-related stayover visitor expenditure survey, was not conducted by the Tourism Board of Saint Lucia. The Board provides the CSO with the stayover and cruise ship passenger arrival numbers, and average length of stay (from migration statistics) as well. The frequency of the data is monthly, and the timeliness of the data is about three months after the end of the reference month.
The mission recommended to extrapolate the average daily expenditures (ADEs) using the consumer price index (CPI) related to each visitor expenditure categories and weighting them using the weights calculated from the existing visitor expenditure surveys for the nearest year. This daily expenditure can be used together with arrival numbers and average length of stay to estimate the stayover visitor expenditure. The mission also suggested to use the daily expenditure estimated for stayover visitor (exclusive of the accommodation expenditure) to extrapolate the ADEs for cruise ship visitors and yacht visitors when their expenditure is not available. The CSO remarked that they do not make estimations and that the ECCB conduct this process, a fact that adds burden to the compilation process of the ECCU member countries’ balance of payments. The ECCB confirmed that the methodology suggested by the TA mission is applied regularly.
17. On travel services credit also, expenditure of students enrolled in the offshore universities (OU) is not significant relative to other ECCU member countries. There is only one university which has a participation of international students. The information on tuitions and other income of the main university is taken from the Inland Revenue Department (IRD). Such information was received.
18. On workers’ remittances, the CSO receives information from the financial regulator authority. The frequency of the data collected is quarterly, and the timeliness is two quarters after the end of the reference month. The information is received. There are no estimations for informal remittances. During the pandemic these informal channels (through travelers) were closed and may be substituted by banks or money transfer operators, for that reason some increase in formal remittances may be observed.
19. On compensation of employees the source data are the ESS surveys. Due to the pandemic the challenge is the lower response rates of surveys. Estimations should consider that some employees became unemployed (for example the workers in hotels and restaurant activities, construction activities and other). The ECCB commented that workers in these areas are usually residents.
Recommended Actions:
- Assess source data regularly, particularly exports of goods provided by the CED, to comply with the guidelines for data accuracy.
- Disseminate exports data consistent with that provided to the ECCB to improve the serviceability of the data.
- Correct Table 11.4 of the Annual Report which does not correspond to the balance of payments of Saint Lucia.
C. Nonresponse, Timeliness of Dissemination, and Longer Time Series
20. The response rates for the balance of payments surveys are low. ESS surveys are carried out by the Survey Unit (SU) of the CSO which is also in charge of other economic surveys such as the National Accounts and the Consumer Price Index surveys. The response rate to balance of payments surveys is lower than 50 percent and around 30 percent for the hotels sector.
21. The CSO processes the data received through surveys, nevertheless their validation process is insufficient, as it does not make estimations for missing data or low response rate. The ECCB works directly with CSO officials cross-checking the data and contacting key respondents as needed. Some of the missing data are completed mainly by the ECCB through the financial statements provided by the IRD. Other administrative sources employed are Invest Saint Lucia, the Land Registry Department, the Airport Authority, the Financial Regulator Authority, and the Ministry of Finance. During the pandemic the information from administrative sources has been received.
22. The mission provided guidance to the CSO compilers on the validation procedures provided by the BPM6CG:
a. Saint Lucia has conducted surveys previously, so they may use a benchmark survey as a basis for the grossing up factors, considering the relative importance of the enterprises that did not respond, compared with those that responded. It is recommended to have an updated business register otherwise the weights could be misleading.
b. The mission also proposed to consider the sector activity, and sample expansion (where possible).
c. A permanent validation process, since the sooner an inconsistency is detected the easier it would be to receive a clarification or correction from the respondent. Telephone contact may help to obtain information from the enterprises that are considered the largest nonrespondents to the survey.
d. It is important to follow a method of estimating the data of nonrespondents. Published information such as financial statements on the Internet may be useful to impute data for large nonrespondent companies. The ECCB confirmed that this is the practice they applied currently.
23. The business registry needs to be updated to provide a better framework to improve the grossing up ratios employed by the ECCB. Recommendations were provided to the CSO on the useful source data (current registers for National Accounts focusing on key respondents, taxation records, information from Invest Saint Lucia and media reports) for this purpose.
24. Actions are required on data validation and cooperation with the ECCB for compilation purposes. The validation process needs to be permanent. It is convenient that the questionnaires include quality control checks that make this process easier. If it does not have this control questions, there are other sources to cross check for example information that the enterprises provide for National Accounts. As the information is processed, results can be observed, if they are not what expected then the data of the main enterprises could be revised accordingly. The absence of a CSO statistician that is in charge specifically at least of the validation process affects the compilation process.
25. Data for 2020 would include alternative data sources to fill data gaps as well as statistical techniques to estimate visitors’ expenditure for the travel account. During a recent mission to the ECCB on March 15–19, 2021, the preliminary 2019 balance of payments and IIP for the eight ECCU member countries were revised. The CSO would provide 2020 balance of payments survey data to the ECCB around mid-year 2021.
26. Saint Lucia’s balance of payments and IIP is disseminated with delays. For surveillance purposes, it is important to reduce the delays or to have some frequent indicators of the ESS. As mentioned in the previous section, according to the frequency of available data the CSO receives monthly trade of goods and remittances (collected on a quarterly basis) and information from the financial sector (banking sector collected directly by the ECCB). Travel could be estimated with the number of travelers provided by the Tourism Board of Saint Lucia and estimated expenditure of stayover, cruise, and yacht visitors.
27. On the backcasting of the balance of payments to complete time series for 2000–13, the CSO mentioned they are not involved in this exercise and the ECCB would oversee it. The recommendation is that the CSO contribute at least with available information of trade and travel of the period 2000–13 for the ECCB.
Recommended Actions:
- Designate officials for the validation process of balance of payments surveys and for the follow up of the data.
- Update and extend the business registry in line with changes in the economy.
- Produce frequent indicators of balance of payments in advance of the preliminary annual estimation. This must be coordinated with the ECCB as they disseminate ESS for all the ECCU member countries.
- A draft revised 2000–13 trade data, number of visitors, visitors’ expenditure, and other information for balance of payments purposes directly received by the CSO is prepared to be shared with the ECCB.
