Argentina
Technical Assistance Report-Report of the Technical Assistance Mission on External Sector Statistics (April 17-28, 2017)

At the request of the National Statistics and Census Institute (INDEC), a technical assistance mission on external sector statistics (ESS) visited Buenos Aires during April 17–28, 2017. This was a follow up to the November 2016 mission that evaluated the ESS methodology, information sources, and dissemination policy and made recommendations to improve quality, adapt the production of ESS to the methodology provided by the Balance of Payments and International Investment Position Manual, Sixth Edition (BPM6), and support the quarterly compilation and dissemination of the International Investment Position (IIP) in accordance with the Special Data Dissemination Standards (SDDS). The mission reviewed the implementation status of the tasks identified in the action plan prepared by the November 2016 mission; assisted compilers in preparing quarterly ESS in accordance with BPM6 guidelines for the next quarterly publication; and provided practical advice on the methodology to be used.

Abstract

At the request of the National Statistics and Census Institute (INDEC), a technical assistance mission on external sector statistics (ESS) visited Buenos Aires during April 17–28, 2017. This was a follow up to the November 2016 mission that evaluated the ESS methodology, information sources, and dissemination policy and made recommendations to improve quality, adapt the production of ESS to the methodology provided by the Balance of Payments and International Investment Position Manual, Sixth Edition (BPM6), and support the quarterly compilation and dissemination of the International Investment Position (IIP) in accordance with the Special Data Dissemination Standards (SDDS). The mission reviewed the implementation status of the tasks identified in the action plan prepared by the November 2016 mission; assisted compilers in preparing quarterly ESS in accordance with BPM6 guidelines for the next quarterly publication; and provided practical advice on the methodology to be used.

I. Introduction

1. At the request of the National Statistics and Census Institute (INDEC), a technical assistance mission on external sector statistics (ESS) visited Buenos Aires during April 17–28, 2017. The mission reviewed the implementation status of the action plan prepared by the November 2016 mission (Appendix I presents the action plan and status of the proposed actions); assisted compilers in preparing quarterly ESS data in accordance with the guidelines prescribed by the Balance of Payments and International Investment Position Manual, Sixth Edition (BPM6) in preparation for the forthcoming dissemination of ESS; and provided practical advice on the methodology to be used.

2. The mission worked with compilers from the INDEC National Directorate of International Accounts (DNCI) and met with staff members from the Central Bank of Argentina (BCRA) and the Argentine federal tax administration (AFIP) (see Appendix II).

3. Section II of this report reviews progress and pending issues and presents the mission’s recommendations for the preparation and release of ESS in Argentina in accordance with the BPM6 methodology, and for the quarterly compilation and dissemination of the International Investment Position (IIP) in accordance with the Special Data Dissemination Standards (SDDS).1 Section III addresses the ESS dissemination and revision policy and other aspects of the compilation of ESS in Argentina.

II. Progress, Pending Issues, and Recommendations Concerning the Compilation of ESS

A. Current Account and Capital Account

4. The DNCI has compiled the historical series for the balance of payments current and capital accounts in accordance with the BPM6 for the years 1990–2016. However, the BPM6-compliant balance of payments series data to be published in July 2017 will begin with data from 2006, because the information on financial liabilities and rents has yet to be validated by the National Public Credit Office for the years prior to 2006.

5. The changes made to the current and capital accounts in migrating to the BPM6 methodology, which are discussed below, do not entail differences in the combined result of those accounts (see Appendix III), although the data to be published in July 2017 may include revisions of several items independent of the adaptation to the BPM6, such as the revision of travel data discussed in paragraph 11.

Goods

6. In the BPM6-compliant data, this account includes goods purchases and sales using information from the unified foreign exchange market (MULC). Also, data for manufacturing services on physical inputs owned by others and maintenance and repair services were reclassified in the corresponding services headings.

7. The customs information allows migrants’ personal effects to be identified. Under the BPM6 methodology, these items should be excluded in the balance of payments goods account, as recommended by the November 2016 mission.

8. In late 2016, the DNCI began receiving data on transactions related to trade in goods broken down by taxpayer identification number (CUIT), which facilitates more detailed analysis of import and export operations to ensure data quality.

Services

9. In addition to the reclassification of goods purchases and sales, manufacturing services on physical inputs owned by others, and maintenance and repair services, as mentioned above, the migration to the BPM6 entailed reclassification of the expense of use of intellectual property; insurance services were calculated as premiums less claims paid; and financial intermediation services indirectly measured (FISIM) were included in services, offsetting the amount against investment income. Also, as recommended by the November 2016 mission, estimated income tax (impuesto a las ganancias) is no longer included in services payments, which reduces service debits and credits in the secondary income account with no net effect on the combined balance of the current and capital accounts.

10. The November 2016 mission recommended including passenger sea and land transport companies in the survey of transportation companies in order to improve the quality of transportation services estimates and, naturally, to improve coverage. The National Directorate of External Sector Statistics (DNSE) prepared a form for this purpose, and a pilot survey of those companies will be conducted. The final decision on including those companies in the survey will depend on the results of the pilot survey and the necessary budget appropriation.

Recommendation:

  • Include passenger sea and land transport companies in the survey of transport companies (December 2017).

11. With respect to travel, the November 2016 mission recommended improving the estimate of spending by Argentines traveling abroad and exploring the availability of information from counterparty countries for this purpose. The DCNI worked on this issue, although recently the procedure of imputing expenses from the International Tourism Survey (ETI) was revised to produce more robust results. Accordingly, this recommendation is considered unnecessary. The new procedure results in increased estimates of travel-related income and expenditure, with the larger increase seen in expenditure (roughly US$350 million in 2016). The procedure for imputing survey data was revised only for the 2016 results, and therefore the historical series for the prior years will need to be chained.

Recommendation:

  • Include information on travel income and expenditure, calculated under the new procedure for imputation of expenses from the International Tourism Survey (ETI), and chain the historical series (July 2017).

Primary income

12. As noted previously, the DNCI has calculated the amount of FISIM, which will be included in the data aligned with the BPM6. The result will reduce net investment income and increase net income from services by approximately US$16 million.

13. In addition to enhancing information on transportation services, the possible inclusion of passenger land and sea transport companies in the transport companies survey (see paragraph 10 above) will provide for updated information on wages and salaries paid by those companies to temporary employees, which is now estimated based on the data for other transportation companies.

Secondary income

14. In accordance with the November 2016 mission recommendations, estimated income tax withheld from payments for services is no longer included in the balance of payments data (as noted in paragraph 9 above, the effect of this change with respect to services is neutralized in both accounts).

B. Financial Account and IIP

15. Since November 2016, the DNCI has been working to implement the most important recommendations concerning the migration of ESS to the BPM6, and has begun publishing the IIP quarterly. The mission met with the compilers to review the quarterly data for the financial account, investment income, and the IIP prepared using the new methodology, and analyzed the consistency of transactions and balances.

16. The November 2016 mission recommended adjusting the information used to compile estimates of nonfinancial private sector financial assets. The difficulty in measuring those assets lies in the lack of appropriate sources of the information, which led to the use of assumptions based on information previously obtained, which dates back decades in some cases. A tax amnesty process in which unreported foreign assets were declared concluded in March 2017, and the main findings were published in April 2017. A total of US$116.8 billion was declared under the process (of which US$93.3 billion in assets held abroad).

17. The foreign assets reported by Argentine private sector residents to the AFIP under that process (i.e., the amounts previously declared on regular tax returns plus assets disclosed in the amnesty process) were generally in line with current IIP estimates except for holdings of foreign banknotes (US$7 billion according to AFIP data and US$115 billion reflected in the IIP), although in this case the amount disclosed in the amnesty process is assumed to be well below the actual amount. For foreign real property holdings, the AFIP figure is slightly less than the IIP figure (roughly US$11 billion and US$11.5 billion, respectively), and therefore, assuming that not all assets were disclosed, the amount reflected in the IIP should be adjusted upward. For foreign accounts and deposits, the amount reported to the AFIP (US$31 billion) is slightly below the IIP estimate (US$35.5 billion). Finally, the AFIP results for investments (US$66 billion) are higher than the investment portfolio estimated in the IIP (US$48 billion), although there is uncertainty as to whether the investments reported to the AFIP include investments classified as ID in the IIP. Any decision concerning revision of the estimates currently included in the ESS would require additional details of the foreign assets reported to the AFIP.

18. The mission and the compilers met with AFIP staff to explore the possibility of obtaining data in addition to the data released to the public (i.e., data by counterparty country, additional characteristics needed to properly classify investments, and information on a group of companies identified as ID) to support a decision on revising the current estimates of private sector foreign assets. While the data reported in the amnesty process is strictly confidential and no additional details can be provided, the AFIP expressed a willingness to provide those details for data reported on regular tax returns (which will foreseeably include all assets, i.e., assets previously reported and assets disclosed under the amnesty process). The data for December 20016 (the first data to include the disclosed assets) will be available in June 2017, at which point the requested details could be provided. With respect to corporations, the foreign assets declared in the amnesty process were not significant (only 4 percent of declarations were filed by corporations), and therefore the analysis of data previously reported on the 2015 tax returns can begin. Accordingly, the DNCI will request detailed data for corporations immediately, without waiting until June 2017.

Recommendation:

  • Obtain data from the AFIP on resident corporations and individuals’ foreign investment at the greatest available level of detail (June 2017).

19. The INDEC and the BCRA have begun exchanging detailed information on foreign asset and liability transactions and balances pursuant to an agreement concluded in November 2016. They have also begun work on reconciling their respective ID results and private liabilities and debt. The mission and the DNCI met with the BCRA to reach agreement on a detailed reconciliation procedure. A detailed timetable was agreed for the exchange and revision of detailed ID and private external debt data and for the reconciliation, release, and revision of aggregates compiled (the agreement was drawn up and ratified during the mission, see Appendix IV). Under the timetable, the first version of quarterly debt and ID data released by the BCRA may diverge, but following the first revision of the data, during the subsequent quarter, the information will be consistent. Successive revisions of the statistics will be coordinated and will therefore continue to be consistent. With the implementation of the agreement, Argentina’s contribution to the CDIS will be consistent with the IIP data for the first time with the submission of December 2016 data in the fall of 2017.

20. The BCRA is currently revising the questionnaires on ID (foreign assets and liabilities) and private external debt to adapt them to the BPM6 requirements, as recommended by the November 2016 mission. The mission and the compilers reviewed the current versions of the ID questionnaires and provided additional suggestions to the BCRA, which were incorporated (the quarterly form is reproduced in Appendix V). Also, the BCRA advised the mission that as recommended by the November 2016 mission, it planned to collect ID information quarterly (the current ID survey collects semiannual data). The quarterly data collection will begin in first quarter 2018 for December 2017 data. With this change, companies with ID holdings in excess of US$100 million would submit information quarterly, and those with ID holdings between US$50 million and US$100 would report annually. This change would result in more accurate quarterly balance of payments and IIP estimates.

Recommendation:

  • Establish a quarterly procedure to obtain DI data as part of the BCRA survey (December 2017).

21. The mission, the BCRA, and the DNCI reviewed the current procedure for collecting information for the financial account and IIP for nonfinancial corporations (NFCs) and households. The BCRA has been collecting private external debt and ID using the same procedure applied during the period when exchange restrictions were in place and DTCs served as the NFCs’ intermediaries in the collection of data. With the elimination of exchange controls, the procedure is less reliable. Moreover, channeling information on corporate assets and liabilities through DTCs could raise problems as to continuity of information. Accordingly, it can be expected that information collected in this manner will tend to gradually deteriorate, which will also affect MULC data.

22. The mission recommends that a procedure be defined to collect information on private sector financial asset and liability transactions and balances whereby the information would be obtained directly from companies and include all financial assets and liabilities (the current surveys only pertain to private debt and ID). To implement such a procedure, the BCRA must be authorized to collect this information directly from companies (it currently relies on its regulatory authority with respect to DTCs, which act as intermediaries). This could be accomplished through a delegation of INDEC authorities or by formally conducting a joint survey pursuant to statistics regulations.

Recommendation:

  • In cooperation with the BCRA, begin defining a new procedure for compiling information on foreign assets and liabilities of the private sector, to replace the current surveys of private debt, DI, and the MULC in the medium term (December 2017).

23. As recommended by the November 2016 mission, the DNCI has compiled the balance of payments financial account, the IIP, and rents in accordance with the BPM6 methodology. For the three items, it has prepared a consistent quarterly data series for the years 2006–2016. Although the IIP data were compiled quarterly for the entire period in question, the BPM6-compliant IIP data to be published in July 2017 will include annual information for the years 2006–2015 and quarterly information for 2016 and first quarter 2017 (see Appendix VII). The DNCI will address the dissemination of the complete quarterly series for 2006 and later years once the necessary testing has been completed.

24. The adaptation to BPM6 methodology did not entail substantial changes in the estimated transactions or the total value of Argentine residents’ foreign assets and liabilities, but rather an improved classification of assets and liabilities. The key changes introduced in the financial account and the IIP to adapt to the BPM6 concern the classification of foreign assets and liabilities by instrument and institutional sector. With respect to instruments, the principal changes related to foreign assets of ORSs, which had previously been included (as a combined total) in other investments and are now distributed between the investment portfolio (securities) and other investments (foreign currency and deposits).

25. With respect to the classification by institutional sector, the new data to be released in July and August 2017 will correctly delimit ORSs, and will also identify OFCs for which information on insurance companies’ foreign assets is available from the National Superintendency of Insurance (SSN) and the Mutual Fund Association (AFCI). The BCRA’s forthcoming modifications to the private external debt survey will improve this distinction for the corresponding liabilities.

Recommendations:

  • Identify the foreign assets of insurance companies (including technical reserves) and mutual funds separately from those of other ORSs.

  • Identify the financial liabilities (loans) of OFCs once the BCRA has completed and implemented the changes in progress to adapt the debt survey to the BPM6 (December 2017).

Direct investment

26. As indicated above, the BCRA is modifying the forms used and the frequency of the ID and private debt survey. The implementation of the changes, in 2018, will result in closer adherence to the BPM6 recommendations and improved data quality. In addition, the procedure for reconciliation of results obtained by the BCRA and the INDEC will improve the consistency of data released by the two entities, beginning with the December 2016 data. The mission assisted the compilers in identifying the priority transactions for which the [reconciliation] process should be addressed on an urgent basis. The reconciliation process will also make it possible to begin submission of Argentina’s contribution to the CDIS for the first time in Fall 2017.

27. The mission and the compilers analyzed the consistency of DI transactions and balances. The comparison revealed substantial differences in values for the years 2010–2014 reflecting the devaluation of the peso with respect to the dollar (the currency in which ESS are disseminated). Since the devaluation coincided with periods of high inflation in Argentina, a possible correction of the companies’ values to take account of the change in prices of their assets would largely offset that effect. The mission recommends consideration, in consultation with the BCRA and the ESS users, of the potential use of alternative estimates of the stock of ID that could be of use to analysts in identifying the effect of the change in the companies’ values.

Recommendations:

  • Conclude the process of reconciling DI data (including rents) with the BCRA so that Argentina’s contribution to the CDIS, the balance of payments, and the IIP are consistent beginning with December 2016 data (July 2017).

  • In cooperation with the BCRA and the users of the information, consider the possibility of using and disseminating alternative estimates of the stock of DI liabilities that take into account the change in price of the companies’ assets and the capitalization of DI companies listed on the stock exchange (December 2017).

28. In accordance with the November 2016 mission’s recommendation, the BCRA has provided for the identification of loans from nonresident DI companies to their Argentine direct investors in the changes to the private external debt survey it plans to introduce in late 2017. In addition, the new ID survey questionnaires mentioned above will implement the recommendations concerning the identification of loans from resident DI companies to their nonresident parent companies and affiliates, equity interests of less than 10 percent that may be held by affiliates, indirect holdings, and clear identification of exceptional income and treatment of taxes.

Recommendation:

  • In the BCRA debt survey, identify loans by nonresident DI companies to their resident direct investors and loans by the largest resident DI companies to their nonresident direct investors (December 2017).

Portfolio investment

29. The November 2016 mission recommended the inclusion of information available from the Caja de Valores, S.A. (the central securities depository, CSD) on nonresidents’ holdings of securities issued locally by the private sector. The DNCI confirmed that this information is already included in the BCRA debt survey, and therefore no additional action is required.

30. The DNCI obtained data on arrears broken down by instrument for the purpose of reclassification in the information adapted to the BPM6. The balance of payments and IIP data to be released in July and August 2017, respectively, will correctly classify the arrears as memorandum items in the balance of payments and in positions.

31. External assets in the investment portfolio presented according to the BPM6 will include estimates of securities held by ORSs (which are currently included in other investments, together with currency and deposits). As noted above, the estimate may be revised in light of information provided by the AFIP in June 2017.

32. In portfolio investments, transactions in the secondary market between residents and nonresidents involving the purchase or sale of securities issued by Argentine residents (in other words, transactions in negotiable debt instruments) have been allocated to the sector that executes the transaction. In the data to be disseminated in June 2017, the transactions will be classified as changes in the amount of issuing sector liabilities, as provided by the BPM6.

Recommendation:

  • Classify transactions between residents and nonresidents involving the purchase or sale of securities issued by Argentine residents in the foreign liabilities of the securities issuing sector (June 2017).

Financial derivatives

33. The mission assisted the compilers in reviewing provisions of the BCRA chart of accounts pertaining to DTCs to identify relevant accounts for the purpose of estimating financial derivatives transactions and balances with nonresidents. It did not appear possible to obtain the necessary data directly from the chart of accounts, because the relevant accounts do not distinguish between resident and nonresident counterparties. However, the BCRA may have other information from which the data can be obtained. Also, the November 2016 mission recommended investigating the possibility that information on nonresidents’ derivatives transactions may be available from the Rosario derivatives market.

Recommendation:

  • Review available data from the BCRA and the Rosario derivatives market to identify any information on transactions with nonresidents involving financial derivatives (June 2017). This information should be obtained for all sectors in Argentina that execute transactions in financial derivatives with nonresidents.

Other investment

34. The information on private sector foreign assets in the form of banknotes, deposits, and trade credits and advances may be revised in view of information provided by the AFIP in June 2017. For trade credits and advances, the November 2016 mission recommended that the survey of companies include a question on average collection times for goods and services exports for use in estimating Argentine companies’ assets and liabilities with the rest of the world. In November 2016, the DNSE agreed to consider that proposal.

35. As noted above, the BCA confirmed that starting at the end of the year, the revised debt survey will include the modifications to the data compiled and the compilation procedures identified by the November 2016 mission.

Recommendation:

  • Continue discussions with the DNSE on the possible inclusion of information required to estimate average import and export collection and payment times (December 2017).

Reserves

36. The DNCI has been receiving information from the BCRA on the quarterly reserves position, as well as detailed information (by currency and instrument) used to obtain data relating to the corresponding transactions, with the exception of valuation differences. However, the data compiled on transactions by instrument using this information is not fully consistent with the positions. The mission assisted the compilers in preparing a spreadsheet (Appendix V) with which the BCRA will provide transaction information fully consistent with the positions by including price information in its analysis.

Recommendation:

  • In the balance of payments, include data on transactions in reserves, broken down by instrument and consistent with the IIP, using data provided by the BCRA (June 2017).

III. Dissemination of ESS and Other Aspects of ESS Preparation in Argentina

37. The November 2016 mission recommended discontinuing the practice of revising the entire published historical series every time new balance of payments and IIP information is released, and defining an ESS revision policy that considers aspects of the compilation process and the needs of users. The DNCI will propose a new ESS dissemination policy that takes account of information exchanges with the BCRA, the need to maintain consistency with the private external debt and ID statistics released by the institution, the revision of other statistics compiled by the INDEC, and the advantages of maintaining a degree of stability in the data released. Under the proposal, each quarterly release of data from June 2017 onward will include a revision of the previous quarter’s data. Also, when the data for a full year is released, the previous year will be revised as well (in other words, in addition to the four quarters of the current year, the [revised] four quarters for the preceding year).

Recommendation:

  • Redefine the ESS revision policy to adapt it to information sources and compilation procedures to be implemented this year. (June 2017).

38. Following the November 2016 mission recommendations, the DNCI subsequently added one staff member, and another is in the contracting process. In addition, a database expert has been contracted on a temporary basis to promote the use of microdata in preparing the ESS. The mission believes it is essential to maintain adequate staffing to perform the tasks delegated to the DNCI by consolidating the current staffing (i.e., the 12 employees who had been hired prior to November and the three additional staff members). Also, the DNCI has continued strengthening its personnel capacities through training activities and seminars on ESS methodology, statistical operations, and IT tools (since the previous mission, DNCI staff have attended six international courses, four national courses, and seven online courses).

39. The DNCI has begun using SAS and other IT tools to efficiently exploit the detailed information it has begun to receive from the BCRA and the DNSE in compiling the ESS. The level of effort should be maintained to build staff capacities in the use of those tools.

Recommendation:

  • Provide further training for staff in managing complex databases (December 2017).

40. During the mission, the DNCI launched the process of preparing metadata adapted to the changes to be introduced to migrate the ESS to the BPM6, with the intention of releasing the metadata on the INDEC website together with the revised data.

Recommendation:

  • Prepare and disseminate ESS metadata incorporating the changes introduced in order to adapt the ESS to the BPM6.