EUROPEAN COMMISSION

Brussels, 30.3.2016

SWD(2016) 68 final

COMMISSION STAFF WORKING DOCUMENT

on quality assessment of balance of payments, international trade in services and foreign direct investment statistics for the year 2014 transmitted by Member States

Accompanying the document

Report from the Commission

assessing the quality of data reported by Member States in 2014 on balance of payments, international trade in services and foreign direct investment

{COM(2016) 163 final}

Table of contents

Executive summary

1.Introduction

2.Overview of Balance of Payments, International Trade in Services and Foreign Direct Investment Statistics

2.1. Legal basis

2.2. Balance of payments and the related statistics

2.3. Quality assessment and reports

3.Relevance

3.1.Data availability

3.1.1.Completeness

3.1.2.Confidentiality

4.Accuracy

4.1.Stability

5.Timeliness and punctuality

5.1.Punctuality

6.Comparability

6.1Asymmetries

6.2FDI Network

7.Coherence

7.1Internal consistency

7.1.1.Consistency with integrity rules

7.1.2.Consistency between quarterly and annual data

7.1.3.Consistency between monthly and quarterly data

7.1.4.Consistency between BOP and IIP data

7.1.5.Errors and omissions

7.2.External consistency

7.2.1.Overall consistency between ITGS and BOP

7.2.2.Directional consistency

8.Conclusions

Annex 1: List of national institutions compiling BOP

Annex 2: Overall quality assessment results

Annex 3: List of abbreviations and codes

Abbreviations

Geographical aggregates and country codes

Annex 4: Glossary

 

Executive summary

This Commission Staff Working Document presents an in-depth analysis of the quality assessment on Balance of Payments (BOP), International Trade in Services Statistics (ITSS) and Foreign Direct Investment (FDI) statistics for the year 2014 transmitted by Member States of the European Union (EU), as well as Iceland and Norway. The quality assessment has been carried out according to Article 4 of Regulation (EC) No 184/2005( 1 ). The report shows the first assessment results since the update of the data requirements by Commission Regulation (EU) No 555/2012 of 22 June 2012 amending Regulation (EC) No 184/2005 and the introduction of the methodology outlined in the 6th edition of Balance of Payments and International Investment Position Manual of the International Monetary Fund (BPM6).

The results of the first quality assessment after the introduction of the BPM6 meet the expectations. All Member States except for Croatia coped with the changed data requirements and methodology. The situation regarding BOP data is the most satisfactory from a quality point of view while improvements are most often needed in the area of FDI statistics. In terms of quality criteria, the overall results are as follows:

Relevance	Completeness was over 95% in most of the cases. Data availability to the final users was satisfactory, in most cases with 95% of main items publishable. However, some Member States exceedingly flag their data as 'non-publishable' or as 'confidential'.
Accuracy	The preliminary analysis of revisions shows stable estimates for the monthly and quarterly current account items and lower stability for the primary income and financial accounts items. The indicator on directional reliability shows a good predictability of the first assessments for monthly BOP.
Timeliness and punctuality	The timeliness of BOP and IIP kept gradually improving during the assessed period. The majority of Member States respected the transmission deadlines for ITSS and FDI.
Comparability	The asymmetries for the current account components have improved during the reference periods but continue to remain an issue.
Coherence	The overall consistency with integrity rules improved slightly but not yet sufficiently. There have been almost no discrepancies between the quarterly and annual ITSS data, while there have been some differences for FDI flows and income. Significant efforts have been made by the Member States to reduce the size of errors and omission; however, their size is still high. The BOP/ITGS directional consistency remains high except for few Members States.

The overall quality of data transmitted according to the requirements of Regulation (EC) No 184/2005 and Commission Regulation (EU) No 555/2012 is good. However, all Member States as well as Iceland and Norway need to address the existing deficiencies while bearing in mind that this first assessment could not explore all the dimensions of the data quality. Moreover, the sound implementation of the new methodology is yet to be analysed and assessed. On the basis of the current quality assessment, Eurostat recommends to the Member States to address with priority the data completeness, the application of the confidentiality rules as well as the reduction of asymmetries and the size of errors and omissions. Eurostat in cooperation with the Member States has already started such an analysis within the remits of the Balance of Payments Working Group. Eurostat and the ECB are also conducting missions to Member States to address, where relevant, country-specific issues. Particular attention and support will be provided to Croatia.

1.Introduction

BOP, ITSS and FDI statistics undergo annual quality assessment as required by Article 4, Regulation (EC) No 184/2005. This Commission Staff Working Document presents an in-depth analysis of the quality assessment on these statistics for the months October 2011-October 2014 for monthly BOP, the quarters 2011Q2-2014Q3 for quarterly BOP, IIP and revaluations and the year 2013 for annual ITSS and FDI. The assessment covers data transmitted by EU Member States as well as Iceland and Norway according to Article 2(1), Regulation (EC) No 184/2005.

This is the first round of quality reporting since the update of the data requirements by Commission Regulation (EU) No 555/2012 of 22 June 2012 amending Regulation (EC) No 184/2005 and the introduction of the methodology outlined in the 6th edition of Balance of Payments and International Investment Position Manual of the International Monetary Fund. Therefore, the report has been adapted to the changes. It also accommodates needs related to the Macroeconomic Imbalances Procedure Scoreboard exercise.

The quality report has been produced on the basis of national quality reports pre-filled by Eurostat and completed by Member States. Certain chapters are complemented by information published as metadata or in relevant articles of Statistics Explained( 2 ). Eurostat is grateful to the many experts who contributed to the quality exercise and the members of the Balance of Payments Committee who assisted in the review process.

This report is the first intended for public dissemination. Eurostat commits to follow this practice of transparency in the coming quality assessment rounds.

2.Overview of Balance of Payments, International Trade in Services and Foreign Direct Investment Statistics

2.1. Legal basis

Regulation (EC) No 184/2005 establishes the common framework for the systematic production of Community statistics on BOP, ITSS and FDI. It has been amended several times since its adoption to accommodate to the economic changes and the technical progress. Most recently, Commission Regulation (EU) No 555/2012 of 22 June 2012 amending Regulation (EC) No 184/2005 updated the data requirements and definitions, aligning them to those of BPM6. A new amendment of Regulation (EC) No 184/2005 is being discussed by the European Parliament and the Council. In 2012, the Commission made a legislative proposal as regards conferring of delegated and implementing powers upon the Commission for the adoption of certain measures to align the existing Regulation with the institutional context after the Lisbon Treaty.

2.2. Balance of payments and the related statistics

BOP as produced by Eurostat records all economic transactions between resident and non-resident entities of the EU or individual Member States during a given period. The Balance of payments provides harmonized information on international transactions which are part of the current account (goods, services, primary and secondary income), as well as on transactions which fall in the capital account and the financial account. The balance of payments is based on a double entry system, therefore the sum of the balances of the current account, the capital account and the financial account is in principle equal to zero, any discrepancy being due to statistical errors. International investment position presents the value of financial assets owned outside the economy and indebtedness of the economy to the rest of the world.

BOP is an important macro-economic indicator used to assess the position of an economy (of credit or debit for current and capital account, net acquisition of financial assets or net incurrence of liabilities for BOP financial account and international investment position) towards the external world.

The compilation of BOP relies on data provided by primary statistics, i.e. ITGS, ITSS and FDI. ITGS measure the value and quantity of goods traded between the Member States and goods traded by the Member States of the EU with third countries. ITSS record the international transactions of services between the EU and its main trading partners. FDI record the direct investment financial flows and income of the EU resident entities as well as the foreign direct investment positions. Securities statistics is used as input for the financial account of balance of payments. Items compiled specifically for balance of payments include compensation of employees, investment income, current transfers and the capital account.

All these statistics are needed for the implementation of the economic and trade policies of the EU. BOP data are used in the compilation of national accounts as well as for producing various indicators.

In the EU, the compilation of BOP at European level is a shared responsibility between the Commission (Eurostat) and the European Central Bank (ECB). While Eurostat is focusing on the BOP current account, the ECB looks closer into the BOP financial account. Annex 1 contains a list of national institutions compiling BOP.

2.3. Quality assessment and reports

Article 4 of Regulation (EC) No 184/2005 provides the requirements for the quality reporting and assessment of BOP, ITSS and FDI. Commission Regulation (EU) No 1227/2010 of 20 December 2010 amending Regulation (EC) No 1055/2008 defines the quality criteria, as well as the content and periodicity of the quality reports.

Every year, Member States are obliged to send to Eurostat quality reports on the transmitted data. These reports, called national reports, are pre-filled by Eurostat with information on indicators, reconciliation tables BOP/ITGS, bilateral asymmetries and values for national data. The Member States review and send back the completed national quality reports with comments and filled-in parts on plausibility, accessibility, clarity and comparability. On the basis of the collected information, Eurostat completes the quality assessment and consults the Balance of Payments Committee on the summary quality report. Eurostat also prepares a more concise version of the report which is sent to the European Parliament for information.

The quality report encompasses all quality criteria defined in the Regulation on European Statistics Regulation (EC) No 223/2009 of 11 March 2009 and is implemented by Commission Regulation (EU) No 1227/2010 of 20 December 2010.

The quality assessment is conducted according to Regulation (EU) 2015/759 of 29 April 2015 amending Regulation (EC) No 223/2009( 3 ) and the quality criteria defined in Article 12 of Regulation (EC) No 223/2009. These quality criteria are: relevance, accuracy, timeliness and punctuality, comparability, and coherence. The results of the assessment are presented in the quality report prepared in line with the Handbook of the European Statistical System for Quality Reports( 4 ).

Eurostat publishes separate quality reports for ITGS( 5 ).

3.Relevance

Relevance refers to the degree to which statistics meet current and potential users' needs. In the BOP, ITSS and FDI quality report this component of quality is measured in terms of the availability of the data required by the Commission Regulation (EU) No 555/2012 of 22 June 2012 amending Regulation (EC) No 184/2005 and disseminated to the final users.

3.1.    Data availability

3.1.1.    Completeness

For all BOP domains the method of calculating the availability for all requests considers the number of reported cells divided by the total number of requested cells according to the Commission Regulation (EU) No 555/2012 of 22 June 2012 amending Regulation (EC) No 184/2005.

Monthly BOP, Quarterly BOP and Quarterly IIP

For monthly BOP request 27 Member States were compliant with the requirements of the Commission Regulation (EU) No 555/2012 of 22 June 2012 amending Regulation (EC) No 184/2005 for reference months in 2014, while Croatia has not transmitted any monthly data and Iceland and Norway were granted derogation.

Figure 1: Quarterly BOP average data availability for 2013Q4 - 2014Q3
(as reported end of December 2014) compared to the EU average

For quarterly BOP, data availability was stable throughout the reference quarters of 2013 and 2014, reaching on average, for the EU Member States, 96% for the four quarters from 2013Q4 until 2014Q3 (see Figure 1). In most cases, the missing items concerned the manufacturing services on physical inputs owned by others, maintenance and repair services, portfolio investment income on investment fund shares and sector breakdown for the financial account and related income. Completeness was 100% for 20 Member States (Bulgaria, Czech Republic, Denmark, Germany, Estonia, Spain, France, Italy, Cyprus, Latvia, Lithuania, Luxembourg, Hungary, Malta, Netherlands, Austria, Poland, Portugal, Romania and Sweden) and over 90% for all remaining countries, except Croatia, Iceland and Norway.

The completeness indicators for monthly BOP and quarterly BOP by country are shown in Table 1.

For quarterly IIP mandatory data requirements of the Commission Regulation (EU) No 555/2012 of 22 June 2012 amending Regulation (EC) No 184/2005 differentiated between Euro Area and non-Euro Area countries, for the latter they were much less detailed. The completeness reached on average 98% for the four quarters from 2013Q4 until 2014Q3 for the EU Member States. 21 Member States (Belgium, Bulgaria, Czech Republic, Denmark, Germany, Estonia, Spain, Croatia, Italy, Cyprus, Latvia, Lithuania, Luxembourg, Hungary, Malta, Netherlands, Austria, Poland, Romania, Sweden and the United Kingdom) as well as for Norway submitted all requested data, while the remaining countries - over 90%. Transmission of data on Revaluations was mandatory only for Euro Area Member States. 15 out of 18 EA Member States provided fully complete data. Two EA Member States (Belgium and Austria) transmitted revaluations after completing of the quality reports’ pre-filling.

Table 2 presents the data availability for quarterly IIP and revaluations by country.

Table 1: Data availability for monthly bop for 2014M07-2014M10 and for quarterly BOP for 2013Q4-2014Q3 (%)

Table 2: Data availability for quarterly IIP and revaluations for
2013Q4-2014Q3 (%)

International Trade in Services Statistics (ITSS)

Completeness of data related to ITSS was 95% in average. Croatian data were reported according to the previous methodological standard BPM5; thus, since Croatia did not report any data according to the BPM6 for reference year 2013, its percentage of completeness is considered to be 0% 6 . Figure 2 and Table 3 show the percentages of data provided by the individual Member States for reference year 2013. Eleven Member States, namely Czech Republic, Estonia, Greece, Ireland, France, Lithuania, Luxembourg, Hungary, Netherlands, Poland and Slovenia, sent all data for services items and partners required by the Commission Regulation (EU) No 555/2012 of 22 June 2012 amending Regulation (EC) No 184/2005. Seven of the remaining countries reached a 99% of completeness rate and merely 3 countries were below the EU average. France, Lithuania, Malta and Poland attended this percentage of completeness after Eurostat's requests for complementary transmissions.

Figure 2: Data availability for ITSS items, per country, for reference year 2013 compared to the EU average

The most common omissions concerned the counterpart area D09 (extra EU not allocated) and R12 (Offshore financial centers). Moreover, there were cases where apparently zero values were reported as empty. Under the same perspective many countries reported empty the cells where the reference area equals the counterpart area. Although these cases were not taken into account for the calculation of the completeness percentages, Eurostat strongly encourages countries to fill in the relevant cells with zero instead of leaving them empty.

Table 3: Data availability for ITSS, FDI flows and stocks for reference
year 2013 (%)

(*) Norway sent annual ITSS data merely according to the BPM5 methodology and thus its percentage of completeness is considered as 0%.

Foreign Direct Investment

For reference period 2013 (due at t+9 months, end of September 2014), the EU average data availability was 91% both for FDI flows and stocks. The completeness of the first delivery according to the BPM6 was lower than before when the data were reported according to the BPM5.

The activity breakdown, with the deadline of t+21 months, was only requested according to the BPM5 methodology, and is not considered in the current quality assessment.

None of the Member State provided 100% coverage for FDI flows. This was mainly because of the relatively lower completeness for FDI income (89%), and despite the complete data reported by 7 Member States for FDI transactions (other than income).

Nine Member States (Germany, Estonia, Greece, Italy, Lithuania, the Netherlands, Slovenia, Slovakia and Sweden) reported full datasets for FDI stocks.

Half of the Member Stats reported values different from zero for resident special purpose entities (SPEs), which indicates that defining separately the SPEs in the Data Structure Definition may have encouraged the national compilers to identify and report FDI activities by such entities. For comparison in the past only 6 countries submitted two parallel datasets (Community concept, incl. SPEs, and national concept, excl. SPEs).

2013 Flows (t+9)

The average availability for EU-28 stood at 91% (see Figure 3). Croatia did not transmit data according to the new BPM6-based data structure. France's low completeness was mainly because data were not reported for reverse investment for equity flows. In addition, the directional principle was not yet applied on income components. Net inward flows data were missing for Bulgaria and Slovakia.

Figure 3: Completeness - FDI flows for 2013 (t+9)

A general problematic issue was the allocation of inward income on debt between fellows by residency of the UCP, where only 6 Member States reported figures.

2013 Stocks (t+9)

The EU overall availability was 91% (see Figure 4). Similar to flows, Croatia did not transmit stocks data according to the new data structure.

Belgium did not report outward equity stocks and inward equity stocks between fellows (UCP is resident), as well as aggregate partners in general. Similar to flows, France did not report data for reverse equity positions.

Belgium and Malta did not allocate the inward stocks between fellows by residency of the UCP.

Member States claiming not being concerned by resident SPEs applied different practices – some transmitted zero (0) values while the others did not include any SPE related codes in their data transmission, the latter contributing further to incompleteness.

Figure 4: Completeness - FDI stocks data availability – 2013 (t+9)

3.1.2.    Confidentiality

Due to national dissemination policies, full monthly BOP datasets were flagged as “non-publishable” or “confidential” by 7 Member States. For Monthly BOP 14 Member States, for Quarterly BOP 10 Member States, for Quarterly IIP eleven Member States, for ITSS nine Member States, and for FDI 10 Member States have made all their data available to users.

For Quarterly BOP for 17 EU Member States the share of free cells is higher than 90% and for 3 Member States higher than 80%, while for IIP for 18 EU Member States the share of free cells is higher than 90% and for 2 Member States higher than 80%. On the other hand, following national confidentiality rules, Iceland flagged all BPM6 data as “confidential”, while Spain and Italy extensively used "N" flags (for non-publishable data). For Spain the share of publishable cells extended in the first half of 2015. The period covered by this quality report was characterized by its transitory nature due to the significant changes on methodological and sources issues, which had impact on confidentiality policy of some Member States.

Table 4 provides information on the levels of disseminated data (all items) for monthly and quarterly BOP and IIP by country.

For ITSS reference year 2013, for 12 Member States this share is higher than 90% and for 7 higher than 80%. For 6 Member States, namely Spain, Malta, Netherlands, Portugal, Slovakia and the United Kingdom, the number of annual data made available to the users is particularly low (below or equal to 7%). More specifically the Netherlands, Spain, Portugal and Slovakia extensively flag the data as non-publishable while the United Kingdom and Malta as confidential. Finally, it is worth mentioning that some countries (Italy, Sweden, Luxembourg and Portugal) that had originally flagged their data as non-publishable, agreed to remove (entirely or partially) their flags after the national publication of their data.

Ten Member States did not apply any confidentiality flags to their national Foreign Direct Investment data transmitted to Eurostat – Belgium, Bulgaria, Greece, Italy, Latvia, Lithuania, Hungary, the Netherlands, Poland and Slovenia.

As expected, the new, more detailed data structure imposed, in general, more extensive confidentiality flagging of national data for 2013 than in previous years.

With a few exceptions, the remaining Member States have flagged relatively small figures, and this is confirmed by the higher share of the free-for-publication data in the total value than in the total number of cells.

For flows for reference year 2013, the above listed 10 Member States made all figures available for dissemination, and twelve Member States reached 100% in value terms.

Despite the larger scale, stocks data for 2013 were also flagged extensively, partially due to the newly requested data for resident SPEs. Whereas eleven Member States released 100% of their 2013 (t+9) stocks figures, the number is bigger (15 Member States) in value terms, confirming the tendency of suppressing the smaller figures.

The implementation of the new data standards also led to the extensive number of cells to be flagged as “non-publishable”. Two Member States (Austria and Slovakia) flagged all data (as required by the Regulation) as "non-publishable". In addition, Spain, Luxembourg and Portugal made less than 30% of their FDI figures publicly available. Anyhow, the situation improves, more visibly for Spain, considering the value of the flagged cells.

Data availability generally, although not in all cases, improves taking into account the values of the flagged cells in the total value of provided cells. Substantial differences can be observed between the share of flagged cells on total cells reported and the share of flagged values on the total value reported. For Quarterly BOP it was most visible for Spain, Luxembourg, Malta and the United Kingdom. This can be explained by the fact that countries generally flag cells with smaller values for more detailed breakdowns. Table 5 shows the confidentiality levels of disseminated data for ITSS and FDI for all items.

Table 4: Share of unflagged cells (available to final users), all items

Table 5: Share of unflagged cells (available to final users), all items for ITSS, FDI flows and stocks

A distinction is also made between flagging of main items and total flagging. Main items for monthly BOP include goods, services, primary income, secondary income, capital account and financial account, and for quarterly BOP they include goods, services, compensation of employees, direct and portfolio investment income on equity and debt, other investment income, secondary income, capital account, direct investment, portfolio investment and other investment (assets and liabilities), with partners Total World, EU 28, Extra-EU 28, Euro Area 18 and Extra-Euro Area 18. For annual ITSS, the main items are total services (S), manufacturing services on physical inputs owned by others (SA), maintenance and repair services n.i.e. (SB), transport (SC), travel (SD), construction services (SE), insurance and pension services (SF), financial services (SG), charges for the use of intellectual property n.i.e. (SH), telecommunications, computer and information services (SI), other business services (SJ), personal, cultural and recreational services (SK), and government goods and services n.i.e. (SL) with partners Total World (W1), EU28 (B5), Extra-EU 28 (D5), Euro Area 18 (I7), Extra-Euro Area 18 (J7), Switzerland, Russia, the USA, Canada, Brazil, Japan, India, China and Hong Kong. For FDI, main items are net outward/inward flows, positions and direct investment income, with the same geographical breakdown as for ITSS.

Looking only at main items (Tables 6 and 7), the availability of Quarterly BOP and ITSS data to the final users is, as could be expected, higher than for all items. For Quarterly BOP and Quarterly IIP improvement was most significant for Ireland, France, Luxembourg, Malta, Austria, Portugal and the United Kingdom. For ITSS, the availability of the main items reached 100% for 12 EU Member States, remaining low for the same countries that exhibited low percentages of availability in all service items as well. For FDI, calculating the share of unflagged data for the main items does not generally show better results neither in number nor value terms. Indeed, for many Member States the share of publicly available figures for main items is lower than the share for all items.

Table 6: Share of unflagged cells (available to final users), main items

Table 7: Share of unflagged cells (available to final users), Main items for ITSS, FDI flows and stocks

4.Accuracy

Accuracy refers to the closeness of estimates to the unknown true values. In the BOP, ITSS and FDI quality report this component of quality is measured looking at the stability of the data that can be assessed looking at the size of the revisions.

4.1.    Stability

For Monthly and Quarterly BOP Table 8, Table 9 and Table 10 present values for mean revisions in the last 36 months (October 2011 – September 2014) and 14 quarters (2011Q2-2014Q3) and Table 11 mean values for main items of IIP. Due to the short available vintages of BPM6 data, with only one data transmission of quarterly data, the analysis provided in this report must be considered as very preliminary. Due to lack of available vintages of annual ITSS and FDI, an analysis of revisions of annual data has not been included in the document.

Mean values of revisions should be interpreted with caution as values might be abnormally high when the initial estimates are low. In particular, indicators of small economies are very sensitive due to the latter factor and in some cases show extreme values despite the fact that the absolute amounts of both first estimates and revisions are small.

In the goods, services and secondary income accounts very small revisions were recorded on credit and debit side, for both Monthly and Quarterly BOP. The few exceptions in monthly BOP included Malta and Slovakia for services, and Hungary, Slovakia and Finland for secondary income where revised values were, however, quite low.

The primary income account was more affected by revisions, especially due to direct investment income. For monthly BOP higher values were recorded for income on equity for Belgium, Estonia (credits) and Malta (debits), while for income on debt instruments for Belgium, Czech Republic (credits) and Malta (credits).

Mean values of revisions are generally higher for financial account items than for current account items, being significant for direct investment assets and liabilities (Belgium, Ireland, France, Lithuania, Austria, Sweden and the United Kingdom), portfolio investment liabilities (Malta and the United Kingdom) and other investments liabilities (Belgium, Estonia and the United Kingdom). As could be expected, the size of revisions for main international investment position items is much less significant than in BOP, with the most substantial mean values of revisions recorded for Malta for direct investment and financial derivatives.

The indicator on directional reliability was introduced for the first time in this round of quality reporting. It measures the reliability of Monthly BOP statistics analysing how often the first assessments correctly predicted an increase or decrease in the final value in comparison with the previous observation. The indicator represents the percentage of cases where the initial series correctly predicts the month-to-month changes of the latest figures and equals 100% when the first and subsequent estimates of monthly BOP items have always the same sign. As shown in Table 12, the directional reliability was 100% for almost all Member States, except, for some items, for Estonia and the Netherlands, which revised their data more often than most of the other countries but still kept high values of directional reliability.

Table 8: Mean values of revisions for monthly bop (C: credits, D: debits)

Table 9: Mean values of revisions for main items of current account (C: credits, D: debits)

Table 10: Mean values of revisions for main items of financial account

Table 11: Mean values of revisions for main items of IIP

Table 12: Directional reliability

5.Timeliness and punctuality

The Commission Regulation (EU) No 555/2012 of 22 June 2012 amending Regulation (EC) No 184/2005 defines the timeliness and sets the deadlines for the data transmission to Eurostat. Punctuality is calculated as the actual date of data delivery minus the scheduled date of transmission to Eurostat. It shows how many calendar days the first data transmission was behind (positive value) or ahead (negative value) of the legal deadline.

5.1.    Punctuality

Timeliness of Monthly BOP, Quarterly BOP and Quarterly IIP kept gradually improving from the first preliminary BPM6 transmission in June 2014 to the last analysed datasets with deadlines on 15 and 23 December 2014. Eurostat would like to thank all Member States for sending the data on the deadline and in many cases even before. Especially for monthly data and for reference periods that include quarterly revisions for longer periods and imply a longer time for processing, it is helpful and very much appreciated to receive the data before the deadline.

For International Trade in Services punctuality of data transmission sustained a satisfactory level. More specifically, 24 Member States sent data before or on the deadline of 30 September 2014 (7 of them on the deadline) while merely 3 countries sent data shortly after the deadline. Eurostat would like to thank the countries for their efforts to report their data on time, taking into account the challenges introduced by the new methodology. Croatia did not send any data according to the BPM6. Nevertheless, it should also be mentioned that additional (corrected, revised or previously missing) data were sent, after the deadline by 8 countries (Bulgaria, Finland, France, Ireland, Lithuania, Malta, Poland and Portugal). As each new set of data multiplies the checks that Eurostat has to perform and makes the process of calculating the EU aggregates lengthier, Eurostat hopes that, in the future, countries will intensify their efforts to send complete data in their first transmissions in order to avoid complementary transmissions after the deadline.

Finally, Eurostat would like to thank the countries that sent "test" or "pilot" BPM6 datasets early before the deadline (Austria, Belgium, Estonia, Italy, the Netherlands, Portugal, Slovenia and Turkey). These datasets and the exchanges helped to test and improve the entire production system (from transmission to dissemination); and to anticipate some of the technical challenges imposed by the introduction of the new methodology.

With the implementation of the new data standards for reporting annual FDI, the punctuality expectedly deteriorated as compared to previous reporting cycles. 15 Member States transmitted data according to the deadline, 30 September 2014, while 8 others did so within 3 days after it. The 4 remaining countries had a delay longer than a half-month. Croatia did not send any data according to the BPM6.

The challenges related to the new, BPM6-based reporting were even more evident considering the large number of re-submissions deemed necessary due to the insufficient quality of the first transmission. In fact, more than one third of the countries were asked to do additional transmissions with corrected data due to inconsistencies irresolvable by Eurostat, which finally caused a delay in the EU aggregates compilation and publication.

Table 13: Overview of punctuality for monthly BOP, quarterly BOP and Quarterly IIP
(sum of 4 months/3 quarters)

Table 14: Overview of punctuality for annual ITSS and FDI data transmission

Table 15: Punctuality of monthly BOP, Quarterly BOP and quarterly IIP

Table 16: Punctuality of annual ITSS and FDI

6.Comparability

Comparability refers to the differences that can be observed when statistics related to the same domain are compared between geographical areas or over time. In the BOP, ITSS and FDI quality report this component of quality is measured by looking at asymmetries and the use of the FDI network aiming to reduce asymmetries in FDI.

6.1    Asymmetries

The Figure 5 below shows total Intra-EU asymmetries, based on Quarterly BOP figures for periods from 2007Q1 until 2015Q1. Asymmetries for total current account mainly reflect fluctuations in asymmetries in trade in goods having both positive imbalances (excess of recorded credits over debits). Asymmetries for services have been stable, also positive and relatively higher than for goods. For primary and secondary income signs of imbalances have been changing; being quite low and without clear pattern for primary income and relatively higher and usually negative for secondary income. Current account asymmetries recorded a maximum value in 2008Q1 (1.9% of the sum of credits and debits) and since then kept decreasing (with the lowest value in 2008Q4 - 0.2% of sum of credits and debits). Starting from 2011 asymmetries were around 1% of sum of credits and debits, with some growth in three last quarters of 2014, for which data can be still, however, considered as preliminary. Due to shorter available time series for the EU aggregates, it was not possible to prepare yet the similar analysis of the financial account data.

The individual quality reports include tables dedicated to bilateral asymmetries, which are object of a separate analysis.

6.2    FDI Network

The BOP, ITSS and FDI quality report monitors the number of transactions and positions exchanged within the FDI Network. During the period 1 October 2013 - 30 September 2014 the Member States exchanged information on a total of 105 FDI transactions, of which not even the half were already reconciled. However the number of reconciled transactions is just a provisional estimation as more exact figures on the reconciliation process will be reported by the Member States at a later stage.

The total value of the 105 transactions observed within the reference period of this Quality Reporting Exercise sums up to about EUR 430 billion. The total number of exchanged transactions since the start of the FDI Network in June 2009 until 30 September 2013 adds up to 596 transactions, representing a total amount of EUR 2,224 billion. Without the FDI Network, parts of these transactions could have been recorded in one country only, which could have affected the bilateral asymmetries in FDI significantly.

The Netherlands and Luxemburg were the initiators of about 76% of the total number of FDI Network exchanged transactions. These 2 countries plus Italy accounted for about 85% of all of the initiated transactions within the reference period. The Netherlands, Luxembourg, the United Kingdom and Germany were also the counterparts for almost 55% of the exchanges initiated by other Member States; with these 4 countries plus Belgium, and Ireland representing almost 75% of the counterparts.

Figure 5: EU28 total asymmetries for main current and capital account items – quarterly balances in millions of euro

While the Netherlands, Luxembourg and Belgium were main users of the FDI Network from the beginning, much more countries now are actively involved in the FDI Network. Looking at both, the initiator and counterpart role, apart from the 3 above mentioned countries now also the United Kingdom, Italy, Germany and Ireland play an important role within this bilateral exchange.

Table 17: Number of transactions exchanged through the FDI Network,
1 October 2013 – 30 September 2014

7.Coherence

Coherence focuses on consistency of statistics produced for different purposes. In the BOP, ITSS and FDI quality report this component of quality is measured looking at the internal consistency (respect of integrity rules, coherence between the quarterly and annual data, size of Errors and Omissions), and external consistency (coherence between BOP data and similar statistics belonging to different statistical frameworks); only external consistency related to BOP goods and international trade in goods statistics (ITGS) data (Intrastat and Extrastat) is monitored.

7.1    Internal consistency

7.1.1.    Consistency with integrity rules

Integrity rules state that the sum of the components should be equal to the aggregates. The integrity rules are defined by a set of equations included in the BOP Vademecum, which should be respected in the datasets transmitted to Eurostat.

Consistency is assessed as ‘excellent’ if no or one inconsistency was detected, and ‘good’ if from 2 to 5 small inconsistencies solvable by Eurostat were noticed. In case of resending of data (marked with an asterisk in the table) due to irresolvable inconsistencies, the last sending has been considered for assessment.

The overall internal consistency improved slightly. Although the need for second delivery diminished in comparison to the last reporting cycle, there were still cases where Eurostat was not able to fix the problems, so the national compilers were asked to resend corrected datasets.

It is reminded that the existence of serious inconsistencies or missing data in the datasets have consequences on the timeliness of Eurostat’s data treatment and dissemination. Therefore, Member States are again strongly encouraged to check the datasets against the integrity rules before transmission. IT tools were already developed in Eurostat and have been made available to the FDI and ITSS national compilers since mid-2012 to further facilitate these checks, such as the EDIT tool available according to the BPM6 methodology as of September 2014. Eurostat strongly encourages Member States to use this tool which can help to significantly improve the quality of their annual ITSS and FDI datasets.

Table 18: Consistency with integrity rules

* resent data

7.1.2.    Consistency between quarterly and annual data

In principle, when annual data are published, quarterly data should be adjusted accordingly. Each subsequent quarterly publication, which includes revisions of previous years, may introduce temporary discrepancies until the next delivery of annual data. Tables 19 and 20 monitor the progress made in aligning quarterly and annual data.

International Trade in Services Statistics

In the datasets delivered at the end of September 2014, there are almost no discrepancies for quarterly and annual ITSS data, except for the Netherlands.

Table 19: Consistency between ITSS quarterly and annual data

Foreign direct investment

The situation remains similar to the previous quality assessment round. Almost all Member States register zero or negligible discrepancies between the two datasets. For the presented reference periods, some higher single discrepancies are observed for FDI flows assets for Estonia, Finland and Poland (both assets and liabilities). For FDI income, credits, discrepancies (with low absolute values) could be observed for Latvia and Slovakia.

Table 20: Consistency between FDI flows quarterly and annual for reference year 2013

A possible reason for these inconsistencies is the delay in the update of the quarterly series following the annual surveys. Therefore, Member States are strongly encouraged to regularly check the consistency between the quarterly and annual datasets, and to provide any revisions to the BOP team in Eurostat on a timely basis.

7.1.3.    Consistency between monthly and quarterly data

The monthly BOP is the initial assessment of the BOP figures. Therefore, full consistency between monthly and quarterly data is not required, as while quarterly data are requested on a full accruals basis, best estimates (e.g. partly on a cash basis) are accepted for the monthly BOP. Table 21 and Table 22 show that, with only few exceptions, consistency between monthly and quarterly figures was ensured by the Member States.

Table 21: Consistency between monthly and quarterly data – goods and services

Table 22: Consistency between monthly and quarterly data – primary and secondary income

7.1.4.    Consistency between BOP and IIP data

Table 23 presents an analysis of consistency between BOP financial account transactions and IIP, which was introduced for the first time in the 2015 quality report. Consistency was to be ensured on voluntary basis, as data for other changes in the volume of assets/liabilities are not required by the Commission Regulation (EU) No 555/2012 of 22 June 2012 amending Regulation (EC) No 184/2005, and for non-Euro Area countries also data for revaluations due to exchange rate changes and other price changes are to be provided on a voluntary basis. Revaluations were provided by 21 Member States, while 2 countries (Belgium and Austria) sent these data only after pre-filling of the quality reports. For assets, consistent data were transmitted for Czech Republic, Germany, Estonia, Latvia and Hungary while for liabilities for the Czech Republic, Denmark, Germany, Estonia, Greece, Latvia, Hungary and Portugal. Data for financial derivatives are requested for BOP only on net basis and for assets and liabilities only for IIP; therefore full reconciliation of this item was not possible.

Table 23: Consistency between BOP and IIP data unexplained changes
(in millions of national currency)

7.1.5.    Errors and omissions

Errors and Omissions (E&O) is the residual BOP item, which in principle should be equal to zero. Thus, E&O is expected to be relatively small and not persistently positive or negative in the long run. E&O tend to be very volatile, in order to get an idea about its tendency, the Average Relative Error, ARE(EO) is calculated for each country. Errors and omissions can be caused by entries in current, capital and, increasingly frequently, financial account. Due to the lack of available data on gross financial flows the analysis below has been limited to the relation to the current account transactions, despite that financial flows are generally bigger in most EU Member States than current account transactions. It is important to note also that errors and omissions in the financial account of the balance of payments do not imply necessarily errors and omissions in international investment position statistics. Table 24 shows ARE(EO) in three different periods: 2009-2011, 2010-2012 and 2011-2013.

The indicator ARE (EO) is defined as follows:

Where EOt are errors and omissions in reference quarter t, N = is the number of the periods analyzed - 12 quarterly observations during 3 years), is the current account (BOP item CA) in reference quarter t, accounting entry - credit, partner World, and = current account in reference quarter t, accounting entry debit, partner World.

Significant efforts have been made in recent years by EU Member States to reduce the size of error and omissions and, as values of the median and of quartiles show, the situation improved compared with the last quality report according to the BPM5. Finland, Sweden and the United Kingdom have been EU Member States with the highest values of the ARE(EO) indicator. While Finland and Sweden recorded values lower than calculated under the BPM5, the figures increase for the United Kingdom. The indicator had results above the median also for Denmark, Spain, Italy, Cyprus, Poland and Romania, as well as Iceland and Norway. On the other hand the Average Relative Error indicator recorded values of 0% or 1% for all analyzed periods for Belgium, Czech Republic, Latvia, Lithuania, Luxembourg, Hungary, Portugal and Slovenia.

An additional indicator shows the cumulated relative sum of E&O and is computed for each year as the cumulated sum of errors and omissions in this reference year divided by the total current account (sum of credit and debit). This indicator assesses the persistency of the sign of errors and omissions, showing lower values for Member States for which E&O are changing signs, which was the case for all countries with higher values of errors and omissions. Table 25 present values of the indicator for years 2011 - 2013.

Table 24: Average Relative Error

Table 25: Cumulative relative errors and omissions

7.2.    External consistency

External consistency is related to the coherence between BOP data and similar statistics belonging to different statistical frameworks.

Only external consistency related to BOP goods and ITGS data (Intrastat and Extrastat) is monitored as part of the BOP, ITSS and FDI quality report.

7.2.1.    Overall consistency between ITGS and BOP

Overall consistency between ITGS and BOP can be quickly assessed looking at the time series of differences between the values of Exports/Credits and Imports/Debits as available from the two statistical frameworks. The national BOP, ITSS and FDI quality reports included a graph and tables for each country. Compilers were asked to provide explanations in case of unstable or increasing divergences. When comparing the two datasets, differences between the BOP and ITGS should be taken into account, which can be classified into three categories. Differences in concepts and definitions are generated by the fact that the BOP requires a 'change of ownership' in order to record a transaction, whereas ITGS record physical cross-border movements of goods. Differing treatment of specific transactions concern e.g. non-monetary gold that changes ownership without being physically transported to the country of the new owner; this gold is not included in ITGS but is included in the BOP. Transactions linked to merchanting are included only in BOP goods, as goods involved in transactions are not being present in the compiling economy and transactions linked to goods send abroad for processing are recorded in BOP as service and in ITGS as goods. Differences in valuation occur because imports/debits are valued free on board (f.o.b.) in the BOP, but are valued cost, insurance and freight (c.i.f.) in ITGS.

While increasing divergences could be detected in a few countries, these divergences do not necessarily imply increasing quality problems since there may be valid methodological causes. The overall EU situation is similar compared with the last quality report according to the BPM5 and remains good, as can be seen from graph 6. Discrepancies remain higher for imports/debits than exports/credits mainly due to the difference in valuation – free on board (f.o.b.) in the BOP and cost, insurance and freight (c.i.f.) for goods imports in ITGS. Member States provide reconciliation tables between BOP and ITGS as part of national quality reports, where all adjustments for transition from ITGS to BOP figures are explained in detail.

Figure 6: Goods – difference between ITGS and BOP, EU28, partner Extra-EU28

Exports and Imports (ITGS), Credits and Debits (BOP) – Source: Eurostat

7.2.2.    Directional consistency 

The Directional Consistency indicator (Qc) assessing consistency between BOP and ITGS is defined as follows:

Where, n11 is the number of cases where the positive trend indicated by the foreign trade statistics is confirmed by a positive trend in the BOP statistics, n22 is the number of cases where the negative trend indicated by the foreign trade statistics is confirmed by a negative trend in the BOP statistics and N is the number of periods analysed, that is 12 (quarterly data for 3 years).

This coefficient (Qc), when multiplied by 100, equals 100% when the changes in the BOP series and the changes in the external trade statistics follow the same pattern; it is equal to 0% when there is a total dissociation
( ).

Table 26 illustrates Qc for three different periods: 2009-2011, 2010-2012 and 2011-2013. In some cases, there is a perfect consistency (Qc =100%) between ITGS and BOP. Directional consistency slightly deteriorated compared with the last quality report according to the BPM5. For exports 4 and for imports 3 Member States have 100% consistency for all three periods. Additionally, 7 countries have 100% consistency for at least one period for exports and 8 for imports. On average, consistency for exports and for imports is on the similar level.

A limited number of countries show a weaker degree of consistency: the Netherlands, Hungary and Poland show low consistency for exports, while Greece, the Netherlands and the United Kingdom for imports.

Table 26: BOP/ITGS directional consistency

8.Conclusions

The results of the first quality assessment after the introduction of the BPM6 meet the expectations. All Member States except for Croatia coped with the changed data requirements and methodology. As shown in Annex 2, the situation regarding BOP data is the most satisfactory from a quality point of view while improvements are most often needed in the area of FDI statistics. In terms of quality criteria, the overall results are as follows:

Relevance	Data completeness was on average 96% for monthly and quarterly BOP, 98% for quarterly IIP, 83% for quarterly revaluations, 95% for ITSS and 91% for FDI flows and stocks. 27 Member States satisfied fully the requirements for monthly BOP, while 1 country has not provided monthly BOP datasets, while completeness was equal to or over 95% for 26 EU Member States for quarterly BOP and for 27 EU Member States and 1 EEA country for quarterly IIP data. Completeness was equal to or over 95% for 24 Member States for ITTS, for 23 Member States for FDI stocks and for 19 Member States for FDI flows. Data availability to the final users was satisfactory, with 95% of main items publishable, for 18 Member States for monthly BOP, 20 Member States for quarterly BOP, 22 Member States for quarterly IIP, 14 Member States for ITSS, 12 Member States and 2 EEA countries for FDI flows and 14 Member States and 2 EEA countries for FDI stocks. Some Member States exceedingly flag their data as 'non-publishable' or as 'confidential'. In sharp contrast to the overall good results is the case of Croatia and, to a certain extent, of Iceland and Norway.
Accuracy	The analysis of revisions in this report must be considered as very preliminary in view of the short availability of BPM6 data vintages, with only one data transmission of quarterly data. For monthly and quarterly current account items, small revisions were observed, indicating stable estimates. The relative magnitude of the revisions was larger for primary income and financial accounts items. The lower stability is mainly due to the difficulty, among others, in estimating reinvested earnings. The indicator on directional reliability shows a good predictability of the first assessments for monthly BOP. Due to a lack of available vintages of annual ITSS and FDI, revisions of annual data were not analysed in this quality assessment.
Timeliness and punctuality	The timeliness of monthly BOP, quarterly BOP and quarterly IIP kept gradually improving from the first preliminary BPM6 transmission in June 2014 to the last analysed datasets with deadlines on 15 and 23 December 2014. For International Trade in Services, 24 Member States sent data before or on the deadline of 30 September 2014, while for FDI, 23 Member States and Iceland made their first transmission before/on or within 3 days after the official deadline.
Comparability	The asymmetries for the current account components have improved during the reference periods. Due to shorter available time series for the EU aggregates, it was not possible to analyse the financial account. Experience in the European FDI Network shows that data exchange can actually help to solve asymmetries, but several preconditions must be met first.
Coherence	The overall consistency with integrity rules improved slightly. However, there were still cases where Eurostat was not able to fix the detected problems, and national compilers had to resend corrected datasets. There have been almost no discrepancies between the quarterly and annual ITSS data, while there have been some differences for FDI flows and income. These inconsistencies can be normally explained by delays in the revision of quarterly series following the annual surveys. Significant efforts have been made by the Member States to reduce the size of errors and omissions. Average Relative Error of errors and omissions (E&O) indicator recorded values equal to or higher than 5% for 6 EU Member States and two EEA countries, while was 0% or 1% for all analysed periods for 8 Member States. The indicator on cumulated relative sum of E&O showed lower values of errors and omissions for all countries with higher values of errors and omissions. The BOP/ITGS directional consistency remains high. However, few Members States have lower than the expected degree of consistency.

The overall quality of data transmitted according to Regulation (EC) No 184/2005 and Commission Regulation (EU) 555/2012 is good. However, all Member States as well as Iceland and Norway need to address the existing deficiencies while bearing in mind that this first assessment could not explore all the dimensions of the data quality. Moreover, the sound implementation of the new methodology is yet to be analysed and assessed. Eurostat in cooperation with the Member States has already started such an analysis within the remits of the Balance of Payments Working Group. The practices for recording of illegal economic activities, goods sent abroad for processing and non-monetary gold have been identified so far as areas for harmonisation and improvement. Eurostat and the ECB are also conducting missions to Member States to address, where relevant, country-specific issues. Particular attention and support will be provided to Croatia.

On the basis of the current quality assessment, Eurostat recommends to the Member States to address with priority the following areas:

Data completeness for the countries where progress remains to be made in order to achieve compliance with the requirements of Regulation (EC) No 184/2005,

The application of the confidentiality rules as defined in Regulation (EC) No 223/2009 for the countries which continue to flag the substantial share of the provided data as “confidential” or “non-publishable”. This decreases the value of statistical information that is provided to users and does not allow an appropriate policy analysis based on European statistics,

Reduction of asymmetries where they continue to persists. Asymmetries remain the issue on the European and global levels. Eurostat would like to encourage the increased use of FDI Network and the bilateral exercises to reconcile other BOP items,

Reduction of the size of errors and omissions. This size remains a substantial challenge for number of countries. Eurostat would like to encourage the investigations of the reasons for high error and omissions and efforts to eliminate them.

Annex 1: List of national institutions compiling BOP

Annex 2: Overall quality assessment results

(1) Considered as 'partially satisfied' when completeness < 95%.

(2) Main items. Considered as 'partially satisfied' in cases when share of unflagged cells < 95%.

(3) Considered as 'partially satisfied' in cases when the average relative error > 5% in all reference periods or average relative error > median in all reference periods.

(4) Considered as 'partially satisfied' in cases when directional consistency indicator < EU-median in all reference periods either for exports/goods credits or for imports/goods debits.

(5) Considered as 'partially satisfied' when completeness < 95%.

(6) Main Items. Considered as 'partially satisfied' in cases when share of unflagged cells < 95%.

(7) Member States of the Euro Area only.

Legend:

‘S’ – ’satisfied’ - quality criteria have been met within specified threshold, some minor improvements may still be needed in specific cases.

‘PS’ - ’partially satisfied’ - quality criteria have not been fully met and substantial improvement is needed.

‘NS’ – ‘not satisfied’ - quality criteria have not been met since no data have been available for the evaluation.

‘n.a.’ – evaluation of the indicator has been not applicable for the specific country.

Annex 3: List of abbreviations and codes

Abbreviations

BOP        Balance of Payments

BPM        Balance of Payments Manual of the International Monetary Fund

FDI        Foreign Direct Investment

ITGS        International Trade in Goods Statistics

IIP        International Investment Position

ITSS        International Trade in Services Statistics

MBOP        Monthly BOP

QBOP        Quarterly BOP

SPE        Special Purpose Entity

UCP        Ultimate Controlling Parent

Geographical aggregates and country codes

EEA        European Economic Area

EU28        European Union of 28 Member States

EU        European Union

EA        Euro Area

BE        Belgium

BG        Bulgaria

CZ        Czech Republic

DK        Denmark

DE        Germany

EE        Estonia

IE        Ireland

EL        Greece

ES        Spain

FR        France

HR        Croatia

IT        Italy

CY        Cyprus

LV        Latvia

LT        Lithuania

LU        Luxembourg

HU        Hungary

MT        Malta

NL        Netherlands

AT        Austria

PL        Poland

PT        Portugal

RO        Romania

SI        Slovenia

SK        Slovakia

FI        Finland

SE        Sweden

UK        United Kingdom

IS        Iceland

NO        Norway

Annex 4: Glossary

Current account	The current account shows flows of goods, services, primary and secondary income between residents and non-residents.
Goods	The goods component of BOP covers moveable goods for which a change of ownership occurs between residents and non-residents.
Services	Services are the result of a production activity that changes the conditions of the consuming units, or facilitates the exchange of products or financial assets. Services are not generally separate items over which ownership rights can be established and cannot generally be separated from their production.
Primary income	Primary income represents the return that accrues to institutional units for their contribution to the production process, or for the provision of financial assets or from renting natural resources to other institutional units. It comprises Compensation of employees, Investment income and Other primary income.
Compensation of employees	Compensation of employees is recorded when the employer (the producing unit) and the employee are resident in different economies.
Investment income	Investment income is derived from a resident’s ownership of an external financial asset (credit) and symmetrically, income derived from a non-resident’s ownership of a domestic financial asset (debit). Investment income includes income on equity (dividends, withdrawals from income of quasi-corporations, reinvested earnings) and on debt (interest), and investment income attributable to policyholders in insurance, pension schemes, and standardised guarantee schemes.
Secondary income	The secondary income account shows current transfers between residents and non-residents. A transfer is an entry that corresponds to the provision of a good, service, financial asset, or other non-produced asset by an institutional unit to another institutional unit where there is no corresponding return of an item of economic value. Current transfers consist of all transfers that are not capital transfers.
Capital account	The capital account covers the acquisition/disposal of non-produced non-financial assets (natural resources; contracts, leases and licenses; marketing assets, e.g. brand names, trademark; goodwill) and capital transfers (transfers of ownership of fixed assets; transfers of funds linked to, or conditional upon, the acquisition or disposal of fixed assets; the cancellation, without any consideration being received in return, of liabilities by creditors).
Financial account	Financial account records transactions that involve financial assets and liabilities that have taken place between residents and non-residents. The financial account shows transactions in net terms: net acquisitions of financial assets correspond to acquisitions of assets less reductions in assets.
Direct investment	Direct investment is 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). Following the international standards, the direct or indirect ownership of 10 % or more of the voting power of an enterprise resident in one economy by an investor resident in another economy is evidence of such a relationship.
Portfolio investment	Portfolio investment includes transactions and positions involving debt or equity securities, other than those included in direct investment or reserve assets. Portfolio investment includes equity securities, investment fund shares and debt securities, unless they are categorised either as direct investment or as reserve assets. Transactions as repurchase agreements and securities lending are excluded from portfolio investment.
Financial derivatives and employee stock options	A financial derivative contract is a financial instrument that is linked to another specific financial instrument or indicator or commodity and through which specific financial risks (such as interest rate risk, foreign exchange risk, equity and commodity price risks, credit risk, and so on) can be traded in their own right in financial markets. Employee stock options are options to buy the equity of a company offered to employees of the company as a form of remuneration.
Other investment	Other investment is a residual category that includes positions and transactions other than those included in direct investment, portfolio investment, financial derivatives and employee stock options or reserve assets. To the extent that the following classes of financial assets and liabilities are not included under direct investment or reserve assets, other investment includes: (a) Other equity; (b) Currency and deposits; (c) Loans (including use of IMF credit and loans from the IMF); (d) Insurance, pension and standardised guarantee schemes; (e) Trade credit and advances; (f) Other accounts receivable/payable; and (g) SDR allocations (SDR holdings are included in reserve assets).
International investment position	International investment position shows, at the end of each quarter, the value of financial assets of residents of an economy that are claims on non-residents, and the liabilities of residents of an economy to non-residents, plus gold bullion held as reserve assets.
Revaluations	Revaluations (holding gains or losses) on an asset or liability arise from changes in their prices and/or exchange rates. Revaluations are further classified into those that are due to exchange rate changes and those that are due to other price changes.

(1) () OJ L 35, 8.2.2005, p. 23–55 and OJ L 166, 27.6.2012, p. 22–66

(2) () http://ec.europa.eu/eurostat/statistics-explained/index.php/Main_Page

(3) () OJ L123, 19.5.2015, p. 90.

(4) () http://ec.europa.eu/eurostat/documents/64157/4373903/01-ESS-Handbook-for-Quality-Reports-2014.pdf/d6152567-a007-4949-a169-251e0ac7c655

(5)

() http://ec.europa.eu/eurostat/web/products-statistical-working-papers/-/KS-TC-15-002

(6) () The BPM5 data reported by Croatia were converted by Eurostat into BPM6 in order to calculate the relevant EU aggregates. These converted data do not appear in the Eurostat reference database.