A. Collection and Compilation of Statistics on Offshore Activities

11. The offshore activity in Seychelles is significant and continues growing. According to the FSA the number of offshore entities exceeds 200,000 in 2021; however, only about 25 percent are active (in good standing).¹ The work for covering the offshore activities in ESS started in 2012 and comprises two phases. The first phase covers collecting data on specific type of offshore entities—the CSL—created for the purpose of avoiding double taxation.² The surveyed population are CSPs that are CSLs financial records keepers and report on behalf of their CSLs clients. The CSPs survey was launched in 2014 (for data for 2012 and 2013) and is carried out through the FSA—the regulatory body of the offshore activity in the country—who then submits the reports to the CBS without executing any validation/checks. The CSPs survey results are consolidated with the balance of payments, IIP, and external debt for the rest of the country and the three consolidated datasets are disseminated. The CBS disseminates the balance of payments and IIP in two formats—including and excluding data on offshore entities.

12. Collecting data on offshore activities in Seychelles remains challenging and requires joint efforts of all stakeholders. Since 2017 TA mission, the authorities made some progress in improving the data on offshore activities in ESS. It included adjusting the CSPs survey form by (i) requesting the number of new CSLs covered in the report; (ii) adding additional breakdowns needed for a better integration of data in the balance of payments and IIP; and (iii) adjusting the title for some components to better aligned them to reporters’ terminology.

13. The mission assessed the results of the CSPs’ survey conducted in 2021 covering data for 2019 and 2020. The data have been reported by CSPs aggregated for all covered CSLs with the indication of number of total CSLs in their portfolio, of those covered in the report, including the number of new CSLs. General check of the consistency and adequacy of the reported data for 2019 and 2020 have been conducted, and, when needed, in comparison with data for the previous years. The mission noticed improvements have been noticed in the quality of reported data. Inconsistencies between positions at the end of the previous period and position data at the beginning of next period have been eliminated, and misclassifications have been observed only in a few instances. However, there are still some irregularities yet to be addressed.

14. A challenge in conducting the CSPs survey faced by ESS compilers is the liaison with reporters. The CBS compilers do not have direct access to reporters for data verification and for following up on identified inconsistencies as the collection of CSPs data is still done with FSA intermediation. As the FSA staff has no expertise in the ESS methodology they are not able to exercise an adequate follow up with the CSPs on inconsistencies. The previous ESS mission stressed the need of allowing the CBS to contact directly the CSPs for follow up; however, this recommendation has not been implemented. To facilitate the direct follow up with the CSPs by the CBS staff, the mission agreed with the FSA representatives that the FSA would inform CSPs that the CBS could contact them to request clarifications on the 2019/2020 report.

15. For cross-checking and validating data reporting by the CSPs, access to the financial statements of CSLs covered by the survey is the most recommended tool. According to existing regulatory acts, CSPs are obliged to submit to the FSA the financial statements of CSLs by end-June of the following year. However, in many instances the financial statements are submitted to FSA with considerable delay. During the 2017 mission, the FSA committed to provide the CBS with the CSLs’financial statements, but this commitment has not been fulfilled because the financial statements were not available timely. Also, starting from 2017, the FSA was expected to implement more aggressive strategy to collect financial statements from CSPs timely, applying penalties for non or late submission. However, no actions have been undertaken in this regard due the absence of a strong regulation that would support the application of penalties.

16. The following immediate actions were agreed with the FSA and the CBS to wrap-up the processing of data from the CSPs survey for 2019–2020 and prepare for the participation in the IMF’s SPEs Survey:

• (i) the FSA sending the feedback questions to the CSPs with the deadline to answer the questions by November 19 (October 15, 2021);

• (ii) the CBS reaching out the CSPs directly for non-response and inconsistencies in reported data (by end-November 2021);

• (iii) the CBS updating the balance of payments and IIP for 2019–2020 by replacing the estimated offshore data with the reported ones (by December 15, 2021).

17. New amendments to the offshore activity regulatory act have been introduced in 2021 that could result in two important implications for the Seychelles’ offshore activities statistics. According to the amendments, the CLSs that are created for taking advantage of tax concessions offered by the Double Taxation Avoidance Treaties will be taxed applying the domestic rates. This may result in the majority, if not all CSLs, leaving Seychelles given that they will no longer be able to take advantage of tax concessions. As of other types of offshore entities—IBCs—starting from 2021, they are obliged to keep in Seychelles their financial records for the last seven years. This could facilitate their survey for the balance of payments purpose. Taking advantage of this change related to IBCs, the CBS should consider initiating the second phase of the project on covering offshore activities in ESS. For this purpose, the CBS together with FSA should develop a strategy for collecting the IBCs data. This would require a careful stratification of IBCs to be included in the survey population;³ further TA may be required for this purpose.

18. The return C(viii) has been replaced by return C1 Deposits by Residency, Institutional Units, Economic Sectors & Currency of Denomination to identify asset and liability positions of offshore entities, introducing recommended amendments by the 2017 mission. A few further changes need to be made before the first collection of the return in 2022, including providing a clearer definition of SPEs in the instruction to the form.

Recommendations:

• Follow up with CSPs to address the identified issues in reported data (e.g., misclassifications, inconsistencies between assets and liabilities, missing data, etc.);

• Adjust the 2019–2020 CSPs Survey results by addressing mistakes and inconsistencies based on updates from CSPs, applying checks and validation procedures discussed during the mission;

• Adjust the CSPs survey form based on the mission’s recommendations (to be used for the 2022 round of the CSPs survey);

• Initiate the preparatory work for expanding the coverage of CSPs survey to IBCs;

• mend returns C1 Deposits by Residency, Institutional Units, Economic Sectors & Currency of Denomination based on mission’s recommendations.

B. Participation in the IMF Survey on SPEs

19. In early 2021, the IMF launched the survey “Resident Special Purpose Entities: Cross Border Flows and Positions” and Seychelles expressed interest to participate in it. The mission discussed with compilers the template for reporting the data as well as the main elements of the operational guidance for the implementation of the survey. The focus was on the statistical definition of SPEs and the application of the decision tree (as per the “Special Purpose Entities: Guidelines for a Data Template”) for deciding if offshore entities incorporated in Seychelles meet the SPE definition. Analyzing the characteristics of offshore entities incorporated in Seychelles following the decision tree, it was concluded that all of them fit the statistical definition of SPEs.⁴

20. The assessment of the report form used by the CBS for collecting data from CSLs confirmed that they contain all main components requested for the IMF survey. In addition, as the CBS compiles and disseminates the consolidated balance of payments and IIP that include data for onshore and offshore entities, the available data allows to fill in all required components of the IMF’s SPE Survey: (i) the totals to be drawn from balance of payments and IIP, and (ii) “of which” data for SPEs by indicator.

21. The mission assisted the CBS in preparing for the submission of the IMF’s SPEs Survey. Thus, the data collected through the CSPs Survey for 2019 and 2020 have been aggregated recognizing that they are row data that will be updated when the CSPs address the identified inconsistencies (Appendix II). Further, the aggregated indicators have been mapped with the indicators requested in the IMF’s SPEs Survey (Appendix III). For finalizing the data before the deadline for the submission of the survey to the IMF, compilers should (i) reach out to the CSPs for clarifications/changes regarding the identified issues in reported data; (ii) adjust the aggregated SPEs data based on changes to the CSPs reports; (iii) revise balance of payments and IIP for 2020 (and 2019) by replacing the estimated indicators for SPEs with the reported ones; and (iv) fill in the IMF’s SPEs Survey with adjusted data.

Recommendations:

• Revise the balance of payments, IIP, and external debt data for 2019–2020 by incorporating the adjusted SPEs data;

• Fill in the IMF SPEs Survey based on the revised balance of payments and IIP data for 2020;

• Submit to STA the IMF’s SPEs Survey for 2020.

C. Compiling Data on Travel Services

22. The data on travel services—the main component of Seychelles current account— included in the balance of payments are deemed to be considerably underestimated. The CBS has been estimating the travel based on data on FX transactions, after the switch from the goods and services taxes paid by services providers to value added taxes and the discontinuation of the provision of data on tourism sales by the Seychelles Revenue Commission (SRC). The method to estimate travel services based on FX transactions data is causing a significant underestimation of tourism services exports as tours booked outside the country in foreign currencies are not covered in the estimation method.

23. The CBS has developed a new estimation model based on available statistics. It sums earnings generated from tourism in the four areas: accommodation, extra expenses, domestic ferry, and domestic flights (all international flights are operated by nonresident airlines). Accommodation revenue is estimated as number of beds occupied multiplied by bed rate and by number of visitors. The data are gathered from the National Bureau of Statistics (NBS), internet, and Immigration and Civile Status Department, respectively. Assumptions are made on what proportion of tourists would be residing in the different establishment types. Extra expenses are calculated by multiplying the total number of visitors by expenditure per visitor. The latter is estimated considering past years trends and is taken at level of 11 percent of accommodations. Tourist spending on ferry and domestic flights are estimated based on NBS statistics on number of visitors assuming that one-fourth of them visit Pralin of which half uses ferry and another half takes domestic flights. The comparison of estimates based on new model and current balance of payments data for travel services is presented in Table 3.

^{Table 2.}

Seychelles: Travel Services Based on New Estimation Model (mil. USD)

^{Table 2.}

Seychelles: Travel Services Based on New Estimation Model (mil. USD)

Year	Estimates based on new model					Current balance of payments data
	Q1	Q2	Q3	Q4	Annual total	BOP	Difference	Ratio in data from new model
2013	162.6	165.4	180.0	175.0	682.9	430.2	252.7	63.0%
2014	150.0	150.7	150.8	144.6	596.1	397.6	198.5	66.7%
2015	131.8	134.8	146.2	143.4	556.2	392.4	163.8	70.5%
2016	138.2	140.3	145.6	139.4	563.6	413.7	150.0	73.4%
2017	146.2	146.8	169.2	172.5	634.7	482.7	152.0	76.1%
2018	173.8	153.1	167.3	179.9	674.1	563.9	110.2	83.6%

View Table

^{Table 2.}

Seychelles: Travel Services Based on New Estimation Model (mil. USD)

Year	Estimates based on new model					Current balance of payments data
	Q1	Q2	Q3	Q4	Annual total	BOP	Difference	Ratio in data from new model
2013	162.6	165.4	180.0	175.0	682.9	430.2	252.7	63.0%
2014	150.0	150.7	150.8	144.6	596.1	397.6	198.5	66.7%
2015	131.8	134.8	146.2	143.4	556.2	392.4	163.8	70.5%
2016	138.2	140.3	145.6	139.4	563.6	413.7	150.0	73.4%
2017	146.2	146.8	169.2	172.5	634.7	482.7	152.0	76.1%
2018	173.8	153.1	167.3	179.9	674.1	563.9	110.2	83.6%

24. The mission assessed the new estimation model and found it suitable. It can estimate expenses of tourists regardless if payments are entering the domestic banking system or not. Some adjustments are recommended such as analyzing the visitors by type and applying different level of expenses for different types of visitors. Also, compilers should consider capturing also tourists’expenses for domestic flights to other than Pralin islands.

25. The CBS should start incorporating the new estimates for travel services in balance of payments compilation beginning with balance of payments forQ1 2022, with backward adjustments from 2016. The 2016 year is recommended because estimates for years before 2016 are less accurate due to issues in reliabilities in the source data. The preliminary calculations shows that new travel services estimates would significantly reduce the current account deficit (see Appendix IV). While using the new estimates for travel services in balance of payments, compilers should consider also estimating offsetting entries to avoid imbalances that ultimately would lead to an increase in errors and omissions. Offsetting entries could be in other sectors’ currency and deposits (assets), trade credit and advances (liabilities), or investment income. The corresponding components of the IIP should also be adjusted. Considering the significant changes in balance of payments indicators while using new travel estimates, compilers should inform users about the foreseen changes in balance of payments before its release.

Recommendations:

• Compile the data on travel services applying the new estimation model;

• Revise backward data on travel services based on new estimation model.

D. Staff Resources

26. The staff resources allocated to the compilation of the ESS need to be enhanced. Currently, three staff members are mainly compiling balance of payments, IIP, and external debt. A few other staff members help data collection through surveys. If they are assigned only to the ESS, such staff resources would be sufficient for performing all tasks related to data collection and compilation. However, the staff members in charge of the compilation of balance of payments, IIP, and external debt perform also other functions. It considerably reduces the time they can devote to ESS data collection, validation, and compilation. This impacts the timeliness and coverage of the ESS, including developing new data sources and following up with reporters for improving the quality of data. The CBS should consider reducing their involvement in other tasks so that they would be able to improve the quality of collected data by regularly reaching out the reporters to address identified issues.

27. Two of the three staff members assigned to the balance of payments, IIP, and external debt compilation have not undertaken any training on ESS methodology. This makes it more difficult to compile sound ESS following the international standards, specifically while dealing with such complex area as offshore entities. The mission recommended compilers to undertake the new online course on Balance of Payments and IIP Statistics launched in 2021. It is delivered every four months; the next round will start in January 2022.

Recommendations:

• Consider reducing the assignment of balance of payments and IIP compilers to tasks not related to ESS;

• Assign the CBS ESS compilers to take the IMF’s online course on Balance of Payments and IIP Statistics.