National Bank of Romania

Statistics Department

NBR Statistics Department, 25 Lipscani Street, Bucharest 3, 030031 Romania

The different domains relevant for external sector statistics (Balance of Payments -BOP, International Investment Position - IIP, Foreign Direct Investment - FDI, and International Trade in Services - ITS) sent to Eurostat are based on the BOP Vademecum reflecting requirements laid down in the Regulation (EC) No 184/2005 on Community statistics concerning BOP, ITS, and FDI, as amended by the Commission Regulation (EU) No 555/2012 of 22 June 2012 and Regulation (EU) No 2016/1013 of the European Parliament and of the Council of 8 June 2016.

These datasets are broadly in line with the sixth edition of the IMF’s Balance of Payments and International Investment Position Manual (BPM6), the OECD Benchmark Definition of Foreign Direct Investment (BD4) and the Manual on Statistics of International Trade in services 2010 (MSITS 2010).

Monthly and quarterly BOP summarize transactions between residents and nonresidents during a specific period. BOP data consist of the goods and services account, the primary income account, the secondary income account, the capital account, and the functional categories of the financial account (direct investment, portfolio investment, financial derivatives and employee stock options, other investment and reserve assets). Differences between the current and capital account on the one hand and the financial account on the other hand are visible under Net errors and omissions that result from imperfections in source data, inconsistent reporting by enterprises and compilation issues. 

Quarterly IIP shows for a country all financial claims on nonresidents and a country’s liabilities to nonresidents at a certain point in time. The breakdown follows the functional categories of the financial account (direct investment, portfolio investment, financial derivatives (other than reserve assets) and employee stock options, other investment, and reserve assets). The sign of the balance shows whether the domestic economic sectors have a net creditor or net debtor position vis-à-vis other countries. The other changes in financial assets and liabilities accounts (revaluations due to exchange rate, revaluations due to other price changes and other changes in the volume) reconcile the balance of payments and IIP for a specific period, by showing changes due to economic events other than transactions between residents and nonresidents.

Annual FDI statistics (consisting of financial account transactions, current account primary income figures and IIP position data) is a category of cross-border investment associated with a resident in one economy (direct investor) having control or a significant degree of influence on the management of an enterprise that is resident in another economy (direct investment enterprise). By convention, such a lasting interest exists when a direct investor owns 10% or more of the voting power or the equivalent (for an unincorporated enterprise). Operational definitions of control and influence are explained in BPM6 § 6.12. Furthermore, the definition of direct investment is the same as in the fourth edition of the OECD Benchmark Definition of Foreign Direct Investment.

Annual ITS statistics record services transactions between residents and non-residents and cover the following categories: manufacturing services on physical inputs owned by others; Maintenance and repair services, not included elsewhere; transport; travel; construction; insurance and pension services; financial services; charges for the use of intellectual property, not included elsewhere; telecommunication, computer and information services; other business services; personal, cultural and recreational services; and government goods and services, not included elsewhere. The services categories are listed in the Extended Balance of Payments Services Classification (EBOPS 2010).

Classification used for the BOP, IIP, FDI and ITS statistics are in broad conformity with guidelines outlined in the relevant manual (e.g., BPM6, BD4).

Nonfinancial transactions in the BOP are generally grouped according to their nature and characteristics. Produced assets are covered in the goods and services account. Primary income captures returns for the provision of labour and financial assets and for renting of natural resources. Secondary income captures further redistribution of income through current transfers, such as by governments, private households or charitable organization in cash or in kind. The capital account is split into gross acquisitions and disposals of nonproduced nonfinancial assets and other capital transfers.

Positions and flows of financial assets and liabilities are primarily grouped according to the BPM6 functional categories. Five functional categories of investment are distinguished: (a) direct investment, (b) portfolio investment, (c) financial derivatives (other than reserves) and employee stock options, (d) other investment, and (e) reserve assets. These functional categories reflect on economic motivations and patterns of behaviour. Positions, the associated income and financial account transactions, and other changes are based on three broad categories of financial assets and liabilities: (1) equity and investment fund shares, (2) debt instruments, and (3) other financial assets and liabilities.

Annual FDI flows are classified according to the directional principle, that is grouping the FDI transactions according to the status of the resident entity; 1) FDI abroad, if the resident entity is the direct investor or, in the case of transactions between fellows companies, if the “Ultimate Controlling Parent (UCP)” is also located in the compiling economy,  2) FDI in the reporting economy, if the resident entity is the direct investment enterprise or, in the case of transactions between fellows companies, if the “Ultimate Controlling Parent (UCP)” is not located in the compiling economy. In the directional presentation, reverse investment can be seen as equivalent to the withdrawal of investment. The instrument classification differentiates between equity (other than reinvestment of earnings), reinvestment of earnings and debt instruments.

Annual FDI income shows amounts payable and receivable between resident and non-resident entities in return for providing financial direct investment assets to the rest of the world, or incurring direct investment liabilities vis-à-vis the rest of the world. The instrument classification differentiates between dividends, reinvested earnings, and income on debt.

FDI positions are also classified according to the directional principle, split into net FDI positions abroad and net FDI positions in the reporting economy. The instrument classification differentiates between equity (including reinvested earnings) and debt instruments.

For all FDI statistics, the geographical allocation is made according to the economic residence of the immediate direct investor or immediate direct investment enterprise, and the recommended classification by activity is that of the direct investment enterprise (to avoid asymmetry issues).  The industrial activity level is based on ISIC4/NACE Rev.2.

International Trade in Services data are presented in line with the Extended Balance of Payments Services Classification (EBOPS 2010).

Known deviations:

Regulation (EU) No 549/2013 (ESA2010) (par. 5.119) and the BPM6 Manual (par. 5.42) provide that funds between deposit-taking corporations are always recorded as deposits. Concerning deposit/loans liabilities of deposit-taking corporations to other counterparts, BPM6 foresees the additional convention that these are to be recorded as deposits, irrespective of the maturity, while ESA 2010 restricts this to short-term funds. Concerning deposit/loans assets of deposit-taking corporations vis-à-vis other counterparts BPM6 foresees the additional convention that these are to be recorded as loans, while ESA 2010 (in § 5.118) makes a similar reference for short-term loans.

Specifically for euro area: The definition of reserve assets is in line with BPM6 guidelines. However, as a member of the euro area, the definition of monetary reserves in a MS’s BOP includes (inter alia) only those liquid claims denominated in foreign currency that the MS’s central bank holds on non-euro area residents. Claims on residents of other euro area countries, regardless of the currency, are part of other external assets.

Institutional units are grouped into institutional sectors according to similar economic objectives, functions, and behaviour.

The sector classifications based on the BOP Vademecum are:

Central bank (S.121); Monetary Financial Institutions (MFI) other than central bank (S.122) (incl. Deposit-taking corporations except the central bank; Money Market Funds (S.123)); General government (S.13); and Other sectors.

The Other Sectors consist of Financial Corporations other than MFIs (S.12); Non-Financial Corporations (S.11); Households (S.14); and Non-profit institutions serving households (S.15).

Exception to BPM6:

A deviation from the BPM6 sector classification is that Money Market Funds are part of Other sectors (in BPM6) whereas they are part of the MFI sector for European Statistics according to the Vademecum.

The overall conceptual framework of BOP, IIP, FDI and ITS are in broad conformity with the most recent manuals as well as the EC Guidelines and Eurostat’s Vademecum.

Statistical concepts and definitions relate to basic internationally accepted standards and guidelines for external sector statistics; for instance:

• All resident-nonresident transactions covered;
• The concept of residency adhered to;
• For the BOP, the concept of gross reporting is followed for the current and capital account; and the net basis for financial account transactions (separately for the individual asset and liability components);
• The change of economic ownership principle soundly applied;
• FDI is defined as equity ownership representing 10 percent or more of the voting power;
• The accrual basis is broadly applied;
• Market values or appropriate substitute measures are used;#
• the residence of Special Purpose Entities (SPEs) is attributed to the economy in which they are legally domiciled or incorporated;
• Overall, the classification, netting and ordering in the IIP is consistent with BPM6; current, capital, and financial accounts of the balance of payments statement are defined according to the BPM6.

Known Deviation (Source: Vademecum):

BOP/IIP data are to be compiled following the debtor/creditor approach, instead of the “transactor” approach. In other words, the geographical allocation of assets/credits is to be done on the basis of the residency of the issuer/debtor and not of the “transactor”. This is particularly relevant for portfolio and direct investment functional categories, which record tradable instruments. This approach is to be followed consistently in the geographical and sector allocation of investment income, financial transactions and stocks.

Institutional units are defined in conformity with BPM6 and relate to those that have a predominant centre of economic interest in the country. In principle, any individual, corporation or other institution that provides information on the transactions/positions between the residents and non-residents of a country during a given period is included. Resident institutional units engaged with nonresidents also cover in principle:

-       incorporated or unincorporated affiliates of nonresident companies; and SPEs with little or no physical presence;

-       resident territorial enclaves in the rest of the world (e.g., embassies, military basis);

-       free zones/bonded warehouses/factories operated by offshore enterprises under customs control;

-       Citizens who work or live temporarily in another country (seasonal and cross border commuters, students and patients).

Not applicable.

The reference area describes the geographical area covered by the data disseminated. According to the BOP Vademecum, the reference area is the economic territory, country, or region for which external sector statistics are provided. The country code list follows the ISO 3166-1 alpha-2 classification and is a "cross-domain" code list, used also in National Accounts. The codes used for various regional groupings are harmonized across international agencies that use the BOP-DSD.

Monthly BoP data are available since 2005 to date, while quarterly data are available back from 1999. Quarterly IIP is available starting with 2008.

Not applicable.

All data sent to Eurostat are in Millions of Euro for Euro Area countries and in Millions of National currency for non-Euro Area countries. 

The data is disseminated in both euro and national currency.

The monthly (MBOP), quarterly (QBOP) BOP, and quarterly FDI transactions summarize economic transactions between residents and nonresidents during the respective reference period.  The annual ITS dataset summarizes services transactions over the period of one year.

The quarterly IIP statement as well as the annual FDI stock statistics refer to a point in time at the end of the reference period; i.e., the last day of a quarter or year, respectively. The other changes in financial assets and liabilities of the IIP statement (revaluations due to exchange rate, revaluations due to other price changes and other changes in the volume) reconcile the BOP and IIP during a specific period.

(i) Law No. 312/2004 on the Statute of the National Bank of Romania: the National Bank of Romania is legally empowered to compile the balance of payments (Article 9 para. (2)); (ii) Law No. 226/2009 on the organisation and functioning of official statistics in Romania: (Article 6 paras. (3)–(5)); (iii) Regulation no. 4/2021 on the reporting of data and statistical information to the NBR.

Non-confidential data that are not published are shared with interested users upon their request. Microdata are shared only within ESCB and ESS, according to the Regulation (EC) No 223/2009 of the European Parliament and of the Council of 11 March 2009 on European statistics regulates protection and transmissions of confidential data within ESS and ESCB.

BOP Vademecum (2021): Chapter V “Statistical confidentiality” of Regulation (EC) No 223/2009 of the European Parliament and of the Council of 11 March 2009 on European statistics regulates protection and transmissions of confidential data within ESS and ESCB. Confidentiality status attribute is mandatory in BOP and FDI DSDs and thus each observation has to be flagged with a confidentiality status.

When the information is already released and disseminated by the national compiler (e.g. it is published on the national compiler's website) applying confidentiality status attributes such as C "Primary confidential statistical information", D “secondary confidentiality set and managed by the national compiler” or N "Not for publication, restricted for internal use only" is not justified.

Quality issues should be much less frequent for quarterly BOP and IIP data. The confidentiality status attribute "Not for publication, restricted for internal use only" - N flag – shall be used conservatively and rather in cases of more detailed breakdowns (e.g. geographical, instrument or sector breakdown). Using confidentiality status (C or D or N flags) to suppress publication of data with insufficient quality should be avoided. Observation status with U flag (low reliability) should be used instead. This would indicate existing observations and, at the same time, users will be aware of the low quality assigned.

This indicator refers to Eurostat Quality Report Chapter 1.1.2.The analysis in this section focuses on the availability of the data to the users. The tables present the number of cells flagged as confidential and non - publishable, compared with the total number of cells that have to be provided according to the BOP Regulation.

Eurostat release calendar can be found under http://ec.europa.eu/eurostat/news/release-calendar

The release calendar for both press releases and database updates is published on the NBR website, under: https://bnr.ro/News-calendar-3204.aspx

Eurostat Website:

BOP: monthly and quarterly

FDI flows and stocks: annually

IIP: quarterly and annually        

ITS: annually

The first dissemination of data is announced via the monthly press release Balance of payments and external debt, to be found under: https://bnr.ro/Press-Releases---Balance-of-Payments-and-External-Debt-4049.aspx

At Eurostat, the database for external sector statistics gives access to the following statistics:

• Monthly and quarterly BOP and quarterly IIP statistics;
• Annual data on ITS;
• Annual data on FDI positions, transactions and income;
• Detailed data on international transactions involving the European institutions.

Harmonized data is available for the European Union, the euro area, the EU Member States, the United Kingdom, EFTA countries, candidate and potential candidate countries.

 https://ec.europa.eu/eurostat/web/balance-of-payments/data/database

Not applicable.

At national level, the database for external sector statistics gives access to the following statistics:

• Monthly and quarterly BOP and quarterly IIP statistics;
• Annual data on ITS;
• Annual data on FDI positions, transactions and income;

The database is available on NBR webpage, under: https://bnr.ro/Interactive-database-1107.aspx

The microdata are shared only within the ESCB and ESS framework, for specific requests, with prior approval.

Not applicable

Not applicable.

Methodological notes for users are published on the NBR website: Balance of Payments - Methodology.

Not applicable.

Not applicable.

This indicator refers to the quality assurance framework (QAF) in place within an organisation to manage the quality of statistical products and processes.

The quality report for 2023 shows a good adequacy of BOP/IIP data in terms of compliance with integrity rules, of completeness and of most other assessment criteria. BOP/IIP data revisions are included in time series as soon as they are available and are closely monitored by NBR.

Follow up of the last year’s Eurostat Assessment

The quality assessment of the data included in 2023 Romanian Quality report indicates good levels for the most data quality criteria.  All data series have been transmitted to Eurostat on time and the consistency with integrity rules remained excellent. The completeness level indicator has been 100% for all required data sets.

Till the results of the surveys are available the compilation system is set to make estimates based on historical data or provisional data sources are taken into account. The 2019 Q2-2019Q4 period presented in the actual report was impacted in the context of quantitative assessment of revisions by the benchmark revision performed in 2019, affecting components like  services,  primary income (compensation of employees), secondary income (social contributions, taxes on income), financial account (currency and deposits, insurance, pension and standardized guarantee schemes). The period 2010-2020 was revised in March 2021 including in the Goods component, the NSI estimates for second-hand cars purchased by residents from abroad. 2020 was a special year, because of the outbreak of COVID-19 pandemic and the lockdown period across the world. NBR used some additional sources in order to adjust some BOP components till the results of the regular surveys became available (it is the case of Services) and to estimate components affected by the exceptional situation like compensation of employees (till the results of Labor force survey were received from NSI), air passenger transport or travel.

The NBR finished in 2021 the implementation of functional specifications for the automatic retrieval of the security holding statistics information from SHSDB, the database managed by ECB and Bundesbank, for the compilation of portfolio investment statistics and security statistics. The revised data for 2020 and 2021 will be send to Eurostat in March 2022.

BOP and NA statistics comparison is not too relevant in this reconciliation exercise as the NA data series, since the NA series starting with 2020 Q1 are missing.

Asymmetries on services generally remained at similar levels. Asymmetries on travel and compensation of employees diminished based on the newly implemented algorithms. Hopefully, some improvements will follow in case of travel (debit), air passengers transport (debit), computer and information services (debit) after the implementation in BOP system of the of the new available data source provided by the ECB Regulations on payment statistics (to be analyzed in the context of the next benchmark revision).

Starting with 2022, NBR is participating in the Asymetry Resolution Mechanism (ARM), under the ITSS Working Group, an initiative that is intended to reduce the level of asymetries in ITS data and improve the overall quality of these data.

The level of errors and omissions item is slightly above the median for the period 2019 Q3 – 2022 Q2, being affected by several factors, among which the inclusion of the estimations of the second-hand cars purchases of residents from abroad or the estimated status of the 2022 quarterly data. Both the cumulative sum of net errors and omissions and cumulative sum of relative E&O, related to CA and IIP tend to increase, but the same trend is observed for the EU27 median. NBR monitors this level on a continuous basis and tries to improve the data quality of BOP components by finding new data sources, calculating reliable estimates or cross checking the primary data available.

Overall Assessment

The quality report for 2023 shows a good adequacy of BOP/IIP data in terms of compliance with integrity rules, of completeness and of most other assessment criteria. BOP/IIP data revisions are included in time series as soon as they are available and are closely monitored by NBR.

The asymmetries related to ITS and FDI are close to those presented in the previous Quality Reports. Asymmetries on travel and compensation of employees diminished based on the new algorithms implemented in the last years. Some improvements in case of travel (debit), air passengers transport (debit), computer and information services (debit) are expected after the new reporting of the payment institutions based on ECB Regulations on payment statistics will enter into force, offering details about the value of online payments made by residents using credit cards on foreign websites, broken down by merchant codes.

NBR constantly checks all primary BOP/IIP data sources in order to ensure a high degree of quality of the disseminated statistics.

The estimations method used for compensation of employees developed in the context of benchmark revision from 2019, changed the positive lasting level of errors and omissions. The level of errors and omissions item (ARE(EO)% indicator) is slightly above the median, since the period under review is affected by the inclusion of the estimations for the second-hand cars purchases of residents from abroad and for Revolut payments, as well as by the provisional status of the 2022 data. Both the cumulative sum of net errors and omissions and cumulative sum of relative E&O, related to CA and IIP tend to increase, as it is the case for the reference (EU27 median).

In 2023, NBR will continue its work to improve the external statistics data by participating in discussions with NSI and other governmental agencies (Ministry of Finance, Ministry of European investments and projects) regarding new data flows, related to EU funds received by Romania and as well as regarding the household sector, aiming at the harmonization of the BOP and NA data in the context of the next benchmark revision in 2024.

To guarantee the quality of results, European statistics shall be developed, produced and disseminated on the basis of uniform standards and of harmonised methods. In this respect, Regulation (EC) No 223/2009 of the European Parliament and of the Council quality criteria (a) ‘relevance’ refers to the degree to which statistics meet current and potential needs of the users. Meeting the requirements of the EC Regulation on community statistics concerning balance of payments, international trade in services and foreign direct investment confirms relevance for the user “Eurostat” as recipient of the data sets on behalf of its user community.

Furthermore, relevance is also indirectly accomplished by countries participating in domain specific committees and working groups as well as the worldwide update of the manuals whose aim it is to keep the standards and statistics as relevant as possible going forward.

User needs are accomplished by keeping statistics harmonized with EU standards, which on their turn follow the worldwide update of the manuals.

NBR closely monitors users’ needs and offers support and specific answers via a dedicated email adress available on its website: statistici-externe@bnro.ro

This section refers to the current Eurostat Quality Report 1.1.1. Data availability- completeness.

Data completeness – the provided cells expressed as % of required cells refers to the completeness of BOP, IIP, FDI and ITS data provided to Eurostat based on the requirements of the BOP Regulation.

Accuracy of data is the closeness of computations or estimates to the exact or true values that the statistics were intended to measure.

Accuracy is being measured using three concepts: Reliability; Vintage Analysis; and Plausibility. See 13.1.

This section refers to the current Eurostat Quality Report Chapter 2.  

The 2019 – 2022 monthly and quarterly data presented in the tables of the 2023 Romanian quality report, chapter Quantitative assessment of revisions, include and are affected by the revisions for the year 2020, when the NBR used some additional data sources in order to adjust some BOP components till the results of the regular surveys became available (this is the case of Services, where NBR used the data on enterprises turnover published by NSI to estimate services for months affected by the lockdown and also the case of compensation of employees, where the NSI and border police press releases related to the number of the individuals who came and left the country compared to similar periods of the previous years were used till the results of Labor Force Survey were received from the NSI). Also, in March 2021, the period 2010-2020 was revised to include, under the Goods component, the NSI estimates for second-hand cars purchased by residents from abroad. As the NIS provides data on second-hand cars on an annual basis, for the subsequent periods the first release includes NBR’s estimates based on historical data.

BOP Monthly data, Period 2019 M04 – 2022 M03, data as reported at end of September 2022:

G Goods (W1 – credit and debit): In case of goods the amounts are slightly revised each month both on credit and debit. The GEO allocation is also updated especially for the previous month. The main source of data is the foreign trade statistics compiled by the NSI based on Intrastat system for intra – EU trade and supplemented by information on extra – EU provided by the National Customs Authority (which process custom trade declarations for export and import of goods outside EU). Each month NBR receives the data series revised starting with the first month of the year both for credit and debit. They are included in the BOP system, the previous compiled months being revised accordingly.

Starting with 2020 September revision of the year 2019, in Goods compilation, debit side, a new data source was introduced. REVOLUT LTD has reported quarterly, since June 2020, the online transactions on foreign websites made by resident individuals using a Revolut card (first quarter reported is Q4 2018). Also, in March 2021 a revision was performed by NBR for 2010-2020 period, including NSI estimates for second-hand cars purchased by residents from abroad.

SMAPE is very low (between 0% and 1%).

S Services (D6 – credit and debit) ( W1 – credit and debit): The main source of discrepancies arises from the method of compiling monthly services data. Preliminary data from the ITS survey are available at T+44 days after the end of the reference quarter and final data are available at T+85 days. Monthly estimates are compiled using historical data which are subject to revisions after the survey data became available. Additionally, there are some reasons behind the revisions for specific periods, such as the 2019 benchmark revision (affecting the period 2019M04-2019M09, i.e. the revision of the grossing-up coefficients of the data collected through the Survey on international trade in services – ITS, as well as some new sources used to improve the travel component, i.e data provided by the National Institute of Statistics (NSI) regarding the touristic expenses made by non-residents in their origin country for the trips on the Romanian territory and data obtain from compensation of employees model related to the expenses made by foreign short-term workers on Romanian territory and also the expenses made by resident short-term workers abroad). For 2020, the year affected by the COVID-19 pandemic, the data for the first months estimates of the lockdown period were adjusted using some information collected by NSI, regarding the turnover of the companies in the most important economic activities, till the ITS survey results became available. The usual model based on historical data was not appropriate to derive estimates for the pandemic exceptional period.

IN1 Primary income (W1 – Credits and Debits): The differences between revisions for the period 2019M04-2019M09 are explained mainly by benchmark revision performed in 2019 by NBR, related to compensation of employees (COE). A new calculation model was used both for credit and debit. In 2020, the first pandemic year, NBR used some additional sources (the NSI and border police press releases related to the number of the individuals who came and left the country compared to previous years) in order to estimate compensation of employees (credit) till the results of the quarterly Labor force survey were received from NSI (usually at 120 days after the end of the quarter). Another source of discrepancies between revisions, on the credit side of primary income, is caused by the fact that NBR currently conducts only annual surveys on outward direct investments, due to the less significant figures. During the year, estimates based on historical data are used to compile income for outward direct investment.

On the debit side of the primary income the component that registered differences at revisions are direct investment (DI) and other investment (OI) income.

Related to DI, debit side, NBR is conducting a quarterly survey on inward direct investment (only first 3 quarters are covered) in addition to the annual survey. First monthly data are estimated using historical data and they are replaced by the results of the surveys when become available.

 For OI income, the OFIN (Other financial information related to nonresidents) survey provides quarterly results for short-term loans interest and for deposits interest; the previous year periods collected data  are subject to revisions each time a new survey is ongoing. Data are reported on a cumulated basis from the beginning of the year, and the quarterly BOP transactions are compiled by subtracting data between two consecutive reporting periods. The months are estimated till the quarterly survey data became available, based on historical data. The discrepancies between revisions are rather small in term of absolute values.

Also, NBR is using the Debt Management and Financial Analysis System (DMFAS) for monitoring the long term external debt and the interest paid by residents for MLT loans received. Revisions on MLT interests can occur from time to time when new reporting agent is identified and he has to report back data or if not all the loans were reported to NBR by an economic agent on time.

IN2 Secondary income (D6, W1 – Credits and Debits): The differences between revisions for the period 2019M04-2019M09 are explained by benchmark revision performed in 2019 by NBR, related to compensation of employees. Social contributions paid by employers and employees and taxes on income regarding compensation of employees (COE) were calculated and included in the debit and credit side of secondary income.

Workers’ remittances are compiled based on the quarterly reporting of most important MTOs that act on Romanian territory.  The monthly data are derived from historical series until the MTOs reports are available.

KA Capital account (W1 – Credits and Debits): The differences in KA transactions are mainly due to the fact that for some components (Acquisitions and Disposals of Nonproduced, Nonfinancial Assets, capital transfers) data requested in the ITS survey are used. First the months are estimated based on historical data and then the survey results replace the BOP items.

BOP Quarterly data, Period 2019Q2-2022Q2, data as reported in September 2022:

G Goods and S Services: SMAPE is very low (between 0% and 3%).

Primary income (W1 – Credits):

Compensation of employees: The differences between revisions for the period 2019Q2-2019Q4 are explained mainly by benchmark revision performed in 2019 by NBR including compensation of employees. A new model of deriving COE was developed for credit and debit. Estimates are performed from historical data till the usual data sources become available (120 days after the end of the quarter, quarterly frequency). In 2020, the first pandemic year, NBR used some additional sources (the NSI and border police press releases related to the number of the individuals who came and left the country compared to previous years) in order to estimate compensation of employees (credit) till the results of the quarterly Labor force survey were received from NSI (usually at 120 days after the end of the quarter).

IN2 Secondary income (D6 and W1 Debits): The important differences between revisions for the period 2019Q2-2019Q4 are explained by benchmark revision performed in 2019 by NBR, related to compensation of employees. Social contributions paid by employers and employees and taxes on income regarding compensation of employees (COE) were calculated within the model used to derive COE and included in the debit and credit side of secondary income. Estimates are performed from historical data till the data sources become available (120 days after the end of the quarter, quarterly frequency).

Financial account transactions (2019Q2-2022Q2):

Direct investment: Starting with the first quarter of 2013, NBR is conducting a quarterly survey on inward direct investment in addition to the annual survey. Even though the quarterly survey is conducted with a higher frequency, the annual survey has a better coverage.  In the quarterly surveys, data are reported on a cumulated basis and the quarterly BOP transactions are compiled by subtracting data between two consecutive periods. Reporting entities are allowed to revise their previous data transmission when needed. Data for the fourth quarter are initially estimated using historical trends and other sources and are later revised, when the annual survey data become available. For outward direct investment, only the annual survey is conducted. The results for the analyzed quarters from 2019 to 2021 were incorporated in a later stage, when the results of the annual survey became available.

In case of Other investment the differences between revisions come from the OFIN survey which is a cumulative survey (each quarter the data are reported starting from the beginning of the year) allowing the reporters to correct data from previous periods. Also, two important cross-checks with financial statements when they become available are performed in order to correct the data reported to NBR which could impact the first transmission of data.

Another source of revisions between quarters is due to BIS data related to deposits held abroad by residents, which is available 120 days after the end of the quarter.

IIP Quarterly data (2018Q2-2021Q1):

Overall, IIP financial World account (W1) has followed the same pattern as it did last year (2021) in the quality report review, showing compilation specificities method during pandemic years (2020,2021). The change in sign occurred in the asset side during the third quarter of 2019 when the data received an impact due to the benchmark revision occasion, while in the liability side the last quarter of 2020 has shown a reverse sign and a drop in the final estimate of the first quarter of 2021.

The upper revision indicator (UPP REV) for IIP assets (100%) falls much over the prescriptive target, showing the impact of the annual outward FDI survey as well as other sources updates during the quarters/years. The directional reliability (DIR REL) indicator for both asset and liabilities (91%) stood well above the prescriptive target revealing high Delta sums always for the fourth quarter of each year with minor corrections between quarters, following the FDI quarterly/annual surveys pattern. Consequently, the delta for net IIP financial account shows a net relative revision (NRR) indicator of 0.85% which is well accepted.

Inward and outward FDI data is compiled using the annual survey. Quarterly FDI surveys brings data for inward direction, only.

Data collected through quarterly surveys (on a cumulated basis) aims to deliver information on initial estimates (as of 1st, 2nd and 3rd quarter), while the final estimates are using annual survey figures when the fourth’s quarter data is derived accordingly (year annual – first 3 quarters). Before the final annual data become available, the fourth quarter is previously estimated (based on historical data and other data sources/estimations).

The annual survey has a better coverage while the quarterly one is more frequent. Reporting entities revise their initial data transmission when the case.

In the context of the Guideline (EU) 2018/1151 of the European Central Bank of 2 August 2018 amending Guideline ECB/2011/23 on the statistical reporting requirements of the European Central Bank in the field of external statistics (ECB/2018/19), starting with 2019, the annual FDI survey conducted in May 2020, was updated in order to include information about other accounts receivable/payable, besides the existing ones related to equity, loans, trade credits and corresponding income. The data for 2019 were revised accordingly.

Since the initial release of data encompasses figures from quarterly surveys and the final estimates have been derived from annual survey (which offers a better coverage), the quantitative assessment indicators show better for directional reliability for both W1 and Extra EU 27 Direct Investments (assets and liabilities). Remains in the prescriptive target for upward revisions ratio only on the liabilities side.

Portfolio investment asset and liability side updates are significant for both Extra EU-27 and World, in that way that assets were underestimated (SMAPE 1.74%), while the liabilities have been slightly overestimated (SMAPE 0.16%). Between the first quarter of 2021 and the corresponding of 2022, a new data source for securities has been put in place (SHSS), which together with CPIS survey completed and refined the data on the holdings side. The information gathered from SHSS updated the preliminary data from Romanian custodian authorities.

Other investment asset/liability quantitative indicators show upwards revision ratios in the prescriptive target for D6 and falling over in case of World figures (92%) - assets, while directional reliability ratios are well above 80 percent indicative mark, likewise previously review. 

Regarding the liability side, the directional reliability indicator is above its prescriptive target showing a slight lower percentage related to W1 area. The symmetric mean percentage error (SMAPE) allows for figures situated between 0.2-0.3 percent (indicating a good predictive, forecast model) in case of other liabilities, while the asset side of other investment relies on SMAPE values between 0.6-0.8 percent. This is because the asset side of other investment depends on quarterly/annual surveys and on some administrative/reporting data not ready available at the time of the compilation process.  

In case of Other Investments W1, the differences between initial and final estimates come from quarterly OFIN survey, which collects data for all major items of this functional category (deposits, loans, trade credits and advances and other accounts receivable/payable) from nonfinancial corporations and nonbank financial corporations. Being a cumulative survey, the data are reported starting from the beginning of the year for each quarter, so the reporting entities have the opportunity to correct information for previous quarters of that year. Also, the survey results are cross-checked with financial statements semiannually, when they become available, in order to correct the data reported to NBR which could impact the first transmission of data.

Not applicable. 

Not applicable. 

Not applicable. 

Not applicable. 

Not applicable. 

Not applicable. 

Not applicable. 

Not applicable. 

Not applicable. 

Not applicable. 

Not applicable. 

Not applicable. 

According to the provisions of the Commission Regulation (EU) No 184/2005 and ECB Guideline ECB/2011/23 datasets are reported by countries to Eurostat with the following timeliness:

The BOP regulation defines the timeliness and sets the deadlines for the data transmission to Eurostat as follows:

-      Monthly BOP: 44 days after the end of the reference period;

-      Quarterly BOP, quarterly IIP and quarterly revaluations: 85 days after the end of the reference period;

-      ITS: 9 months after the end of the reference period;

-      FDI: 9 months after the end of the reference period (21 months for the activity breakdown).

Not applicable. 

Not applicable. 

The punctuality for the current edition of the Quality Report is stated in the table below (14.2.1)

This indicator refers to Eurostat Quality Report 3.1 Punctuality.

Punctuality is calculated as the actual date of data delivery minus the scheduled date of transmission to Eurostat. It shows how many calendar days this was behind (positive value) or ahead of (negative value) the legal deadline.

Coherence refers to the adequacy of the data to be reliably combined in different ways and for various uses.

In the Eurostat Quality Report, the analysis of coherence focuses on two aspects: internal consistency, that examines to which extent data are coherent within the dataset, and external consistency, that examines to which extent data are coherent with others statistics (e.g., NA) obtained by different sources or within different statistical frameworks.

Comparability refers to the measurement of the impact of differences in applied statistical concepts and methodologies, measurement tools and procedures applied, when statistics are compared between geographical areas, sectoral domains (e.g., with QSA, ITGS data) or over time. 

This indicator refers to Chapter 5.3.1 and the corresponding tables of the Eurostat Quality Report: Asymmetries with regard to main ITS and FDI items.

Further information and country-specific feedback is provided below. 

ITS 2021 as reported in September 2022

The main asymmetries analyzed by partner countries revealed discrepancies for items such as Transport (Credit and Debit), Travel (Debit), Charges for the use of intellectual property (Debit), Telecommunications, computer and information services (Credit), Other business services (Credit and Debit) and Personal, cultural and recreational services (Credit).

In case of services, except for Travel component, Air Passenger transport (Debit), Telecommunications, computer and information services (Debit), the only compilation source, is the quarterly survey on international trade in services. The sample of ITS survey consists of about 8500 reporting entities and is established annually in close cooperation with the Romanian National Institute of Statistics (NSI) based on administrative reports on intra Community deliveries/ procurements/ supplies (VIES), for intra EU trade in services. The sample coverage is about 85% overall. Statistical procedures for grossing up are used in order to estimate for 100% coverage. In the recent past years NBR noticed that the international services are largely concentrated in SMEs, small companies tend to have a great impact on total services reported. As a result a possible cause for asymmetries between countries may be the sample itself. The ITS sample used by our main partner countries, Germany,  Italy, France, Spain, Hungary may not be as relevant if they do not capture the services in relation with Romanian SMEs.

In 2020, the pandemic year, in case of travel item, the main data sources were the monthly banking reports on the credit/debit cards transactions made by non-residents in Romania (credit) and by residents’ abroad (debit), the data collected from foreign exchange bureaus, the resident travel agencies data regarding payments made abroad by residents, all these data adjusted with NSI press release data related to the arrivals of foreign visitors and the departures of Romanian visitors abroad. Other additional data source used in the compilation of travel items is data estimated from compensation of employees’ model, related to the expenses made by foreign short-term workers on Romanian territory and also to the expenses made by resident short-term workers abroad.  NBR noticed that the asymmetries are lower on the debit side than in the previous year, as it was developed an algorithm to derive country breakdown taken into account mainly two important sources: monthly banking reports on the credit/debit cards transactions and also resident travel agencies data regarding payments made abroad by residents. An average between the two data sources at the level of common countries is used. Previously, in travel compilation (debit side), the country structure resulted from cards transactions had a larger impact, as the amounts reported for this data source are larger than the ones coming from other data sources.

Significant asymmetries are recorded mainly with Germany, Italy, Ireland and Malta. The ITS survey has a good coverage and NBR considers that the results related to these type of transactions are quite accurate in terms of GEO allocation.

The significant asymmetries in Transport (Credits and Debits) are registered mainly with Germany, Spain and France (Credits) and Hungary (Debits). The primary data source for these services is also the ITS survey which has a good coverage, especially for the credit side, where our data indicate larger values. Starting from January 2017, estimations have been implemented in the current BOP compilation system considering the online transactions with credit/debit cards made by residents on the foreign websites, reported by credit institutions. A specific ratio, obtained from the ITC survey conducted annually  by NSI, is applied in order to estimate the amounts related to air passenger transport paid online using cards. The amounts from online transactions are distributed to the countries reported by resident banks. In 2023, NBR expects to obtain, in the context of the ECB Regulations on payment statistics update, the information regarding the value of online payments made using credit cards on foreign websites, broken down by merchant codes. NBR hopes that this will improve the current estimates regarding the distribution of online transactions between Goods and Services and also between types of services, which will determine a reducing in the value of the asymmetries both for air passenger transport and travel.

As for Telecommunications, computer and information services (Credit), Other business services (Credit and Debit), the primary data source is the ITS survey. NBR considers that it has a good coverage.

The differences on compensation of employees’ item have been reduced in the context of the new estimation model developed in 2019.

FDI 2021 as reported in September 2022

Data on inward direct investment in 2021 were obtained mainly by means of a statistical survey, conducted annually and quarterly by the National Bank of Romania (NBR) and the National Institute of Statistics. The other data source is direct reporting undertaken according to the law by financial enterprises (i.e. banks, non-bank financial institutions and insurance companies).

The sampling frame of the foreign direct investment (FDI) survey consisted of 36,881 FDI enterprises, out of which 7,951 enterprises were surveyed (either via comprehensive or random survey).  Specifically, a number of 6,485 enterprises were subject to a comprehensive survey,  including enterprises with at least 20 employees (5,073 entities), all credit institutions  (27 entities), non-bank financial institutions in the Special Register (41 entities) and insurance companies with foreign capital (36 entities), as well as the enterprises referred to as atypical, namely all direct investment enterprises with less than 20 employees whose turnover or share capital was of at least lei 30 million or which took medium and long-term loans of at least lei 15 million from their foreign direct investors or from fellow companies, as well as those for which there is information that they record significant FDI flows/positions (1,308 entities). A sample of 704 companies out of the 6,479 FDI enterprises with 5 to 19 employees was subject to a random survey.

Regarding the survey on the direct investment of residents abroad, 357 entities were subject to a comprehensive survey, apart from 218 FDI enterprises that also reported direct investment abroad.

The overall response rate of both the FDI survey and the Direct Investment Abroad survey was 95.90 percent.

Numerous validation checks are undertaken aiming at ensuring data quality. The data on FDI in Romania are cross-cheeked with administrative records (financial statements from Ministry of Finance, supervisory records for banks and non-banks financial institutions) and other in-house databases such (e.g. External Debt Register).

As a general remark, based on our experience, inward direct investment data cover better the direct investment phenomena (due to a better capturing of the inward investment - the debtor generally  knows very well the borrowings).

Regarding DI_D4P_D_F Income on direct investment in the reporting economy (DIRE) (debit), NBR has supplementary verified the figures and they are confirmed.

Regarding DI_D43S_D_F5 Reinvested earnings (debit) NBR has verified the figures once again and they are correct.

Regarding DI_FA_D_F5B Reinvestment of earnings (liabilities), as previously mentioned, a lot of validation steps are carried out in determining the components of this indicator, such as checks with administrative data (financial statements form Ministry of Finance etc.).

Regarding DO-FA-D-F (liabilities) Direct investment positions abroad, NBR conducts a comprehensive survey. The surveyed entities consists of all resident units (including individuals) that invested abroad. The population is determined based on the notifications transmitted to NBR according to the law  and is complemented with data from financial statements submitted to the Ministry of Finance, Euro Group Register and media releases. In the same way as for direct investment, NBR carries out many validation steps also for the direct investment abroad statistics.

The most important discrepancy is likely to have an explanation in the DI_FA_D_FP for the same counterparty country, most probably due to fellow investment.

Regarding DI_FA_D_F (assets) Direct investment positions in the reporting economy (assets), we have performed supplementary verifications and the figures are correct.

Regarding DI-FA-D-F (liabilities) Direct investment positions in the reporting economy (liabilities) we have undertaken additional checks, and we confirm the data (DIRE figures are always better captured).   

Comparability over time is ensured by updates of historical data whenever a change in data sources or compilation methods occurs.

Methodological comparability is ensured by the international methodological framework, as detailed below.

Methodological guidelines for compiling BOP, IIP, FDI and ITS statistics are defined by the BPM6, the Manual on Statistics of International Trade in Services of the United Nations, and the OECD Benchmark Definition of Foreign Direct Investments (BD4).

The methodology for Goods and Services is outlined in Chapter 10 of the IMF BPM6. For community statistics, countries are expected to record different values for Goods “national” and “community” concepts, if applicable. 

The methodology for Goods and Services is outlined in Chapter 10 of the IMF BPM6.

The methodology for Primary Income is outlined in Chapter 11 of the IMF BPM6.

The methodology for Secondary Income is outlined in Chapter 12 of the IMF BPM6.

The methodology for the Capital Account is outlined in Chapter 13 of the IMF BPM6.

The methodology for Direct Investment is outlined in Chapters 6, 8 and Appendix 6 of the IMF BPM6.

The methodology for Portfolio Investment is outlined in Chapters 6, and 8 of the IMF BPM6.

The methodology for Other Investment is outlined in Chapters 6, and 8 of the IMF BPM6.

The methodology for Financial Derivatives is outlined in Chapters 6, and 8 of the IMF BPM6.

Not applicable

Not applicable

These indicators refer to Chapter 6 and the corresponding tables of the Eurostat Quality Report: Coherence.

The comparability between BOP, IIP, FDI, ITS and National accounts is ensured by the application of common concepts and definitions of BPM6 and the 2008 SNA/ESA 2010.

Further information and country-specific feedback is provided below. 

This indicator refers to Eurostat Quality Report Chapter 6.2.1: BOP/ITGS reconciliation. The corresponding table – to be completed by countries is part of the Quality Report Annex 2 and is included in the Metadata Handler in form of a separate Excel Sheet.  

ITGS cover goods “which add to or subtract from the stock of material resources of a country by entering (imports) or leaving (exports) its economic territory” (United Nations IMTS: Concepts and Definitions 1998, paragraph 14). This basis differs from the change of ownership between residents and nonresidents required for BOP, so adjustments are needed.

Please refer to BPM6 § 10.17 for cases that are included in the BOP definition of general merchandise because there is a change of ownership of goods between a resident and a nonresident.

Please refer to BPM6 § 10.22 for items to be excluded from general merchandise because there is no international transaction because there is no change of ownership of goods between a resident and a nonresident, or because the goods have no value. 

The information displayed in the tables for the period 2019 Q3-Q4 in case of National accounts are extracted from the June 2020 NA data transmission to Eurostat. Differences, taking into account the latest transmissions of NA, occur on secondary income and capital transfers as NSI registers the European funds received from European Commission and the contribution paid to European Commission on an accrual basis and also on property income as it includes some small estimates related to income attributable to insurance policy holder and pension entitlements.

For the period 2020Q1-2022Q2 no data for NA is available for comparison in the tables presented.

These indicators refer to Chapter 6 and the corresponding tables of the Eurostat Quality Report: Coherence. 

Quarterly and annual data are consistent.

There monthly and quarterly data are fully consistent.

IIP asset/liability positions are broadly consistent by functional category and by instrument. Small differencies may occur when the data disseminated in RON currency reveals, at basis level, rounding up operations.

For the period analysed in the current report, 2019Q3 – 2022 Q2, ARE(EO)IIP exceeds the 3rd quartile threshold of the EU27, while ARE(EO)CA for the same time span is in the third quartile, but closer to the median. 2019 was a year that registered high positive level of errors and omissions, while 2020, 2021 and Q1-Q2 2022 registered negative errors. Cumulated sum of relative E&O indicates declining values in the last 6 quarters.

Not applicable. 

Source: A Harmonised European Revision Policy for Macroeconomic Statistics; CMFB October 2017. Macroeconomic statistics, such as national accounts, the balance of payments and the international investment position, are produced from a large variety of data sources. These data sources are reconciled using an approach based on an agreed set of international guidelines. The sources used to estimate macroeconomic aggregates are provided with varying degrees of timeliness, taking up to three years or more in the case of structural sources. As users need national and international data as fast as possible, particularly on certain key aggregates like gross domestic product (GDP), data are produced using the sources and related indicators that are more readily available. As more complete data are obtained from these sources in due course and the structural sources are made available, the statistics are updated to incorporate the new information.

Such revisions of macroeconomic statistics are necessary to improve quality, but can be inconvenient for users. To minimise this inconvenience, revisions should ideally be coordinated within one country, across different statistics, and then across countries. International comparability – and the compilation of EU and euro area aggregate statistics – is hampered when different revision policy schemes are applied in different countries. As the schedule of revision of national accounts and balance of payments statistics varies from country to country, this creates inconsistencies among different statistical domains.

The European Statistical System (ESS) and the European System of Central Banks (ESCB) try to strike the right balance between incorporating the necessary statistical revisions and maintaining an acceptable degree of consistency across domains and countries. To this end, the two systems have worked together to draw up guidelines for a harmonised revision policy for macroeconomic statistics.

A distinction should be made between 'routine' revisions and 'major' or 'benchmark' revisions. Routine revisions refer to the changes made to the economic data published initially and to its subsequent releases for a particular reference quarter or year. The earlier estimates typically undergo the most significant revision. In routine revisions, the number of past periods being revised (the 'depth' of the revision) is typically relatively limited. Benchmark revision is carried out at much longer time intervals. Its purpose is to incorporate the main new data sources and major changes in international statistical methodology (such as ESA 2010 or BPM6). In benchmark revision, many years are open for revision in order to create the longest possible consistent time series.

The National Statistical Offices and the National Central Banks are not legally bound by this common policy, but voluntarily agree to it and commit to gradually implement it with the aim of delivering more consistent statistics to users. The level of adherence to the guidelines of countries' revision policies will be monitored regularly.

Routine revisions for the current year are made regularly for monthly data for the items based on quarterly surveys (monthly estimates are replaced by data resulted from the surveys, such as ITS, FDI, Other investment). Also monthly ITGS data from the beginning of the current year reported as revised by the NIS are adjusted accordingly and included in the BOP data.  

Not applicable. 

This quality concept refers to whether the composition of data sources (surveys, ITRS (International Transactions Reporting System), administrative data, ITGS, monetary and financial statistics, etc.), in principle, sufficiently covers the compilation of BOP, IIP, FDI and ITS. 

Please refer to Table Data-Sources Detail.

Not applicable.

The following sub-categories refer to the Extended Balance of Payments Services Classification (EBOPS 2010) data collection. The EBOPS 2010 classification provides a breakdown of the Balance of Payments Trade in Services items (debit and credit) as defined in BPM6, by types of services. The classification thereby meets a number of user requirements, including the provision of more detailed information on Trade in services as required in connection with the General Agreement on Trade in Services (GATS).

EBOPS 2010 is a primarily product-based classification of types of services, which in many cases may be described in terms of international classification of products as contained in CPC Ver. 2. However, the classification also includes transaction-based criteria (for example, Travel, Government services, and Construction Services).

Yes

Yes

Yes, where applicable

In the quarterly survey framework, the data are collected on a monthly basis and denominated in the national currency. The conversion into EUR is made at the monthly average exchange rate.

EBOPS 2010 items are compiled broken down by partner country.

Currently no information beyond EBOPS 2010 is published, but work is in progress as regards the implementation of Regulation (EU) 2019/2152 of the European Parliament and of the Council on European Business Statistics – basic act, and  Regulation (EU) 2020/1197 – implementation act, which refer to several other classifications of services, like STEC and MOS.

Data validation is made by verifying the integrity rules of the Vademecum and via Conval (previously known as EDIT tool).

Not applicable.

Not applicable.

Applied to ITGS data in order to comply with the change in ownership principle.

Not available