UK FOREIGN DIRECT INVESTMENT (FDI)

SURVEYS

UK Data Archive Study Number 6664 - Annual Inquiry into Foreign Direct Investment, 1996-2011: Secure Access

Definition of FDI :• FDI refers to an investment that adds to, deducts

from, or acquires a lasting interest in an enterprise operating in an economy other than that of the investor.

• The investor’s purpose being to have an “effective voice” in the management of the enterprise.

• For purposes of the survey an effective voice is taken as equivalent to a holding of 10 per cent or more of the voting share capital of the enterprise.

UK FDI and M&A inquiries

Quarterly Cross Border Merger& Acquisitions Abroad

(UK>Abroad)(Outward M&A)

Quarterly DomesticMergers & Acquisitions

(UK>UK deals only)

Quarterly Cross Border Merger& Acquisitions in the UK

(Foreign>UK)(Inward M&A)

UK Annual FDI Abroad(Outward FDI)

UK Quarterly FDI Abroad(Outward FDI)

Foreign Annual FDI in the UK(Inward FDI)

Foreign Quarterly FDI in the UK(Inward FDI)

UK FDI and Mergers & Acquisitions surveys

UK FDI/M&A populations & samples(1)

(1) Counts cover 2003 survey periods

(2) Selection parameters are turnover for the inward sample and number of foreign subsidiaries for outward

IDBR = Inter Departmental Business Register

Survey UKpopulations

Selection criteria Tim etable UK sam ples Response rates(targets)

Sam ple coverageby Net BookValue

M &A n/a Self selecting sample Despatch:immediate

Publication:Quarter + 5weeks

Inward=531Outward=1272Domestic=756

Inward=93% (80%)Outward=96%(80% )Domestic=54% (80%)

N/A

FDIquarterly

IDBR (updatedannually byW orldBase )&existingstandalone FDIregister (13000[inwards] &3500 outward]companygroups)

Automatic selectionfor all companygroups with a netbook value more than£50 million (75million)

Random sampling forother sizebands

Despatch:End of quarter

Publication:Quarter + 13weeks

Inward=734Outward=415

Inward=94% (75%)Outward=95%(70% )

Inward=79%Outward=75%

FDIannual

IDBR (updatedannually byW orldBase )&existingstandalone FDIregister (13000[inward] &3500[outward]companygroups)

FDI Automatic selection

for all companygroups with a netbook value more than£50 million (75million)

Random sampling forother sizebands

W orldBase (2) Automatic selection

for top 100 companies Random sampling for

other sizeband

Despatch:End of year +2 months

Publication:Year +12months (firstrelease)

Year +13months (M A4)

Inward=2574Outward=1344

Inward=90% (90%)Outward=89%(85% )

Inward=84%Outward=78%

FDI Survey Processes• Selection and printing of forms (FDI systems & IDBR)• Despatch of paper forms and reminders for non-responders

(IDBR).• Manual data entry to FDI systems• Validation checks (intra period and extra period)

– query single large component values in current period (priority vets)– checks gross not net data entered– compares same component in different periods (annual/annual,

annual/quarterly, quarterly/quarterly)– compares different components in same period (eg: profits and dividends

or interest accrued and intercompany balances)– compares same components for same period in different inquiries (M&A)

• Analysis and comparison of data from the different surveys (Annual FDI/Quarterly FDI/M&A)

Imputation (including manual constructions for large non-responders) and estimation

annual use a mix of period on period ratio estimation (for contributors with previous returned data) and averages (for contributors with no previously

returned data) for imputation and estimationquarterly uses a imputation link to add to the previous quarter’s value (1

quarter only)

Aggregation - data is aggregated for annual, quarterly and monthly datasets

annual data is held within the Ingres/OpenRoad survey system by component/industry/country

quarterly data is initially estimated within the Ingres/OpenRoad survey system, but intermediate processing takes place in EXCEL and final

processing within Win CSDBmonthly data is aggregated using reports in the Ingres/OpenRoad survey

system then final processing takes place in WinCSDB

• Disclosure - standard ONS disclosure rules are followed– annual data is disclosure tested for all components/industries/ countries

combinations held– quarterly data is not disclosure tested, only World total all sector by

component data is published– Monthly M&A data is disclosure tested using a semi-automatic system of

OpenRoad reports backed by manual checks for the final stage

M&A Survey Processes• Monitoring of public sources for M&A information

– specialist publications & financial press– commercial M&A websites/large company websites/news websites– returned forms from the other FDI surveys

• Self selection from monitoring activities, forms despatched from Data Validation Unit directly (not IDBR)

• Manual entry to M&A systems• Validation checks

– compare funding splits and total value of deal– compare confirmed total with unconfirmed total– compare with same contributors in FDI inquiries for same period

• Analysis and comparison of data from different surveys (Quarterly FDI (>10% deals) and M&A (50% deals)

• Aggregation & disclosure (limited to first release data mainly)

UK FDI outputs (1)

• Quarterly FDI– quarterly first release “Balance of Payments”

• global FDI earnings, flows, positions data by component– annual publications :– “United Kingdom Balance of Payments” (The Pink Book)– “United Kingdom National Accounts” (The Blue Book)

• global FDI earnings, flows, positions data by component and sector

• Annual FDI– annual first release “Foreign Direct Investment”

• geographical breakdown of FDI earnings, flows, positions– annual business monitor MA4 “Foreign Direct Investment”

• geographical/component/industrial breakdown of FDI earnings, flows, positions

– UK & international returns (DTI, DFID, Inland Revenue, Eurostat, OECD, United Nations)

(1) first releases and publications are available free from the National Statistics website: www.statistics.gov.uk

UK M&A outputs (1)

• M&A (Cross Border & Domestic)– quarterly first release “Mergers and acquisitions involving UK

companies” • Cross Border and Domestic data• Numbers and values of deals• funding information• brief geographical breakdown• list of ‘significant’ deals (includes information from the public

domain only)

• Domestic Acquisitions & Mergers– Company and Personal sector branch

(1) This first release is available free from the National Statistics website: www.statistics.gov.uk

Monthly FDI estimation for European Central Bank (ECB)

• currently no monthly data collection is carried out for FDI data• real monthly M&A data is used to model equity capital monthly FDI

flows component (Large unconfirmed or moved deals are manually checked prior to estimation procedures)

• FDI flows components and FDI earnings data are forecasted on a quarterly basis using ARIMA models within the ONS’s WinCSDB system. The quarterly data is then splined to produce monthly estimates

• flows & earnings estimates are produced for World (A1) and Non-Euroarea (U4)

• monthly funding split information is also provided for portfolio investment estimation procedures, as well as briefing on large M&A deals for the current month or revisions to previous months data.

Title: FDI - Methodology 1. The purpose of the foreign direct investment (FDI) inquiries is to collect financial information on the relationship between UK companies and foreign parents and associates (inward FDI) and between UK companies and foreign subsidiaries, affiliates and branches (outward FDI). The information is primarily required for measuring the UK’s balance of payments and international investment position. It covers the investment flows into and out of the direct investment enterprises, the earnings attributable to and current remittances (dividends and interest) to and from the direct investors and the overall stock of direct investment at the end of the inquiry period. Data for the banking sector are collected by the Bank of England; other sectors are covered by ONS inquiries via sample surveys. These keep inquiry costs to a minimum both for the government in terms of processing and business in terms of form-filling. They also lead to a quicker delivery of inquiry results than might be the case with a census. 2. The direct investment inquiries have been conducted in one form or another since 1958 and since their inception the number of foreign owned businesses both in the UK and foreign has increased considerably. As a result, the number of contributors to the inquiries has increased in order to maintain acceptable coverage and accuracy of results. For example, in 1995, the sample design of the quarterly inquiry was improved and the number of forms despatched increased from 800 to 1100 per quarter. The inquiries are all conducted on a statutory basis under the Statistics of Trade Act 1947. 3. The surveys are based on paper forms which are generally sent to the head of the enterprise group within the UK and which request consolidated information for the group as a whole. Because of the nature of direct investment the questionnaires used are relatively complex. Therefore over the years there have been a number of changes to the forms to improve consistency and coherence. For example, as a result of a 1994 review, the inquiries no longer collect detailed balance sheet information. 4. The objective of the surveys is to provide timely and reliable quarterly and annual data on foreign direct investment. The results are used principally in the compilation of the balance of payments and the sector and financial accounts of the national accounts. They are also used by others both within and outside government. These include the Department of Trade and Industry, HM Treasury, Foreign Office, Bank of England, Ministry of Agriculture, Fisheries and Food, embassies of foreign governments and private sector economists, journalists and academics. International organisations such as the Statistical Office of The European Union (Eurostat), the Organisation for Economic Co-operation and Development (OECD), the International Monetary Fund and the European Central Bank also take the data. The statistics are used in a variety of ways: for example, the DTI use the data to measure the attractiveness of the UK to foreign investors. 5. Direct investment refers to investment that adds to, deducts from or acquires a lasting interest in an enterprise operating in an economy other than that of the investor, the investor's purpose being to have an effective voice in the management of the enterprise. For the

purposes of the statistical inquiry, an effective voice is taken as being equivalent to a holding of 10 per cent or more in the foreign enterprise. Other investments in which the investor does not have an effective voice in the management of the enterprise are mainly portfolio investments and these are not covered here. 6. Direct investment is a financial concept and is not the same as capital expenditure on fixed assets. It covers only the money invested in a related concern by the parent company and the concern will then decide how to use the money. A related concern may also raise money locally without reference to the parent company. 7. The inquiries cover investment in subsidiaries, associate companies and branches. They do not cover cross-border investment by public corporations or investment in property. These are regarded as direct investment in the national accounts however and figures for that activity are added to inquiry totals when they are used in the balance of payments and national accounts. 8. The definitions used to compile direct investment information are based on international guidelines. These include the IMF Balance of Payments Manual (fifth edition), the OECD Benchmark Definition of Foreign Direct Investment (third edition) and the 1995 European System of Accounts. They also take account of company accounting practice in the UK through periodic reviews which include consultation with contributors. 9. Statistics based on the latest versions of these international guidelines were first published with the 1998 United Kingdom Balance of Payments Pink Book. The changes from earlier data are as follows: · Previously for the measurement of direct investment, an effective voice in the management of an enterprise was taken as the equivalent of a 20 per cent shareholding. This is now 10 per cent. · The Channels Islands (Jersey, Guernsey etc.) and the Isle of Man are excluded from the definition of the economic territory of the UK. Previously these islands were considered to be part of the United Kingdom. · Interest received or paid was replaced by interest accrued in the figures on earnings from direct investment. · In the national accounts, local taxes payable on direct investment are included in the figures for earnings from direct investment. Previously earnings were net of tax. 10. The inquiries provide information on three main aggregates which can be broken down further into component detail. 10.1 The flows of investment could be in the form of a takeover of a business, a capital injection perhaps to fund the setting up of a new business or a loan to an existing subsidiary. Data are collected on: · Reinvested earnings · Net purchase of shares and capital injections

· Loans between related concerns · Change in the level of investment in branches 10.2 The level or stock of investment represents the net book value of the subsidiary to the parent at the end of the period. The levels are not simply the cumulative total of all the flows up to the end of the relevant period: they take into account goodwill write-offs, the revaluation of assets and exchange rate fluctuations. They are also affected by any cross-border takeover activity which involves local funding from within the country of the target company but which would not appear in the cross-border flow of investment. The levels of investment are at book values and these are likely to be significantly different from current market values as book values tend to reflect values at earlier periods when assets were acquired or subsequently revalued. Figures are collected for: · Share capital and reserves at end period · Balance of loans outstanding · Level of branch assets held 10.3 Earnings from the investment could for example be in the form of a dividend or interest on an outstanding loan. Profits and losses included in earnings are before accounting for capital gains and losses or other exceptional items such as environmental provisions. These would, of course, affect the levels of investment. Earnings data are collected on: · Profit or loss attributable to the parent · Interest payments and receipts between related concerns · Branch profits 11. Some of this detail is collected quarterly and some only annually. For example, information on share capital and reserves is collected annually. 12. A geographical breakdown of investment is collected as part of the annual inquiry. The analysis of inward investment is based on the country of ownership of the immediate parent company. Thus, inward investment in a UK company may be attributed to the country of the intervening foreign subsidiary, rather than the country of the ultimate parent. Similarly, the country analysis of outward investment is based on the country of ownership of the immediate subsidiary. For example, to the extent that foreign investment in the UK is channelled through holding companies in the Netherlands, the underlying flow of investment from this country is overstated and the inflow from originating countries is understated. 13. Annexe 1 contains a list of the countries used in the annual inquiry system together with their allocation to the broader regions used in the main inquiry publication. Although data are held for all of the countries listed, not all of the information is available publicly owing to the need to protect confidential data on individual enterprise groups. 14. Annual data are collected at the 2 digit level of the UK Standard Industrial Classification. This provides 60 possible industry codes for use in analysing the results. Owing to the need to protect confidential data on individual enterprise groups, the industrial sectors published in the results are generally amalgamations of the codes. Published headings and the codes included are described in annexe 2.

15. This level of industrial detail is not produced as part of the quarterly inquiry. Instead figures are produced for inclusion in the national accounts according to broad sectors of the UK economy. These are: Monetary financial institutions Insurance companies Security dealers Other financial intermediaries and financial auxiliaries Private non-financial corporations 16. Only underlined sectors are included in the quarterly and annual inquiries described in the remainder of this paper. Data on monetary financial institutions are provided by the Bank of England and on insurance companies and security dealers by other statistical inquiries into financial institutions conducted elsewhere within ONS. Figures for these sectors are added to the results of the standard inquiries before disseminating any direct investment aggregates. Additionally, data on foreign direct investment by public corporations and investment in property are included in national accounts totals for direct investment, as mentioned in paragraph 7 earlier. More information on the estimates for other sectors is available in the methodological notes of the United Kingdom Balance of Payments Pink Book. 17. The statistical methodology and processing systems used to produce direct investment statistics are continuously under review. Current developments include an investigation into the treatment of extreme returns in the annual inquiry, greater use of the ONS central register - the Inter Departmental Business Register - in the sampling procedures for both the annual and quarterly inquiries, more automated data capture techniques using scanners and image character readers and the use of electronic questionnaires. 18. Summary results of the annual inquiries are available in the foreign direct investment First Release which is published annually in the December following the year covered. It shows flows, levels and earnings by area and country for both inward and outward inquiries. The text of the Release is available on the ONS website at www.ons.gov.uk. The full Release including tables is available in hard copy from: The Press Office DG19 Office for National Statistics 1 Drummond Gate Pimlico London SW1V 2QQ Tel: 0171 533 6363 Fax: 0171 533 5719 19. More detailed information is published subsequently in the ONS Business Monitor MA4 - foreign Direct Investment. This additionally includes analyses by industry and detailed component, and is published around 14 months after the year covered. The Business Monitor is available in paper and electronic formats: as a Microsoft Word 97 ® file of the whole

publication or a Microsoft Excel 97 ® spreadsheet of the tables with separate Word 97 ® text. These are available from: ONS Direct Office for National Statistics Room D.140 Government Buildings Cardiff Road Newport South Wales NP9 1XG Tel: 01633 812078 Fax: 01633 812762 e-mail onsdirect@ons.gov.uk 20. Special analyses of foreign direct investment are available on request from: Simon Harrington Office for National Statistics foreign and Financial Division Room 1.075 Government Buildings Cardiff Road Newport South Wales NP9 1XG Tel: 01633 812406 Fax: 01633 813306 e-mail simon.harrington@ons.gov.uk A charge may be made for data not readily available but any possible charges will be agreed before the work is carried out. 21. The aggregate results from the inquiry are included in other ONS and international publications such as the UK Balance of Payments Pink Book, the UK National Accounts Blue Book, Eurostat's News Release on Direct Investment and OECD's International Direct Investment Statistics Yearbook. 22. Results from the quarterly inquiries are published in the Balance of Payments First Release. This is available quarterly (in March, June, September and December) from the ONS Sales Office at the address given in paragraph 18.

Annual Direct Investment Inquiries 23. The annual direct investment inquiries are the surveys from which detailed results are produced. They are the source for analyses by country and industry. They also collect component detail not included in the quarterly inquiries. This includes information on share capital and reserves and a breakdown of share and loan transactions. The production of annual direct investment figures is carried out towards the end of the calendar year following the year being estimated. This section describes the current survey methodology used to determine the estimates for both the inwards and outwards surveys. 24. The inquiries cover UK enterprise groups with foreign subsidiaries, affiliates and branches, and UK groups with an foreign parent. As described in paragraph 16, the sectors covered are private non-financial corporations and other financial intermediaries and financial auxiliaries. Other ONS inquiries collect direct investment information on insurance companies and security dealers, and the Bank of England covers monetary financial institutions. Figures for these are included in the published inquiry results. 25. Data for the sample selection emanate from an FDI register of enterprise groups, with reference to the ONS' Interdepartmental Business Register (IDBR) to pick up contact addresses. One of the main sources of information for the FDI register is a separate inquiry conducted into cross border acquisitions and mergers. This inquiry uses the financial press, specialist publications (such as Acquisitions Monthly and Mergers and Acquisitions International) and the internet to identify deals. Paper forms are then mailed to the UK company to confirm the transaction and pick up further detailed information to be used in the national accounts. Once the deal is completed, the business is added to the register. 26. The IDBR is compiled primarily from administrative information such as VAT details from HM Customs and Excise and PAYE from the Inland Revenue. It holds a number of variables including information on the country of ownership for each group and information on which UK groups have foreign affiliates. Until 1998 the main source of data on these foreign links was a publication called "Who Owns Whom" published by Dun and Bradstreet. 27. Sampling of enterprise groups is based on a stratified design. The strata are defined in terms of the level or stock of investment in the UK part of the group in the inward inquiry and the level of investment in the foreign affiliate in the outward inquiry. The level of investment is measured at net book value. Both the outward inquiry and the inward inquiry are based on six net book value strata. Due to the skewed distribution of investment among the groups, the sample is heavily weighted towards the larger concerns which account for the vast majority of direct investment assets and where the potential for sampling error is much larger. Procedures used to estimate for groups which were not selected ensure that inquiry results are not biased towards the larger groups however. 28. All groups in the top strata (containing the largest businesses) are sent forms but in strata containing smaller businesses only a proportion are selected. In effect all enterprises with net assets above £10m or net liabilities above £2m will be selected. In total around 900

forms are used in the outward inquiry and 1600 in the inward inquiry. Table 1 summarises the sampling fractions: Table 1: Sampling fractions for the annual direct investment inquiries OUTWARDS INWARDS Net book Sampling Net book Sampling value (£m) fraction value (£m) fraction >=10 1:1 >=10 1:1 5 - 10 3:4 5 - 10 2:3 2 - 5 3:5 2 - 5 2:3 0 - 2 3:5 0 - 2 2:3 -2 - 0 2:3 -2 - 0 2:3 <-2 1:1 <-2 1:1 29. Major changes were made during 1995 to the sampling procedures used in the FDI inquiries. This was partly in line with recommendations from a formal review of the inquiries and partly the result of ONS' programme to improve the statistical methodology across the range of its inquiries. This has affected both the quarterly and annual inquiries. The main change to the annual is to accelerate sample rotation to ensure no groups’ data are estimated for more than two consecutive years, and also that the majority of smaller groups are not approached for more than three consecutive years. 30. Forms are despatched to businesses in January after the end of the year covered with a return-by date of the end of April. Two written reminders are subsequently sent to non-responding groups and these are followed by telephone, fax and electronic mail reminders to groups that have still not replied. There is also the possibility of using the legal powers of the Statistics of Trade Act 1947 to force response, though the Office prefers to work closely with businesses to produce the necessary information. Once a sufficient number of responses have been received (by the end of September at the latest) the system is ‘shut down’ and the annual estimation process itself starts. Target response rates are 90 per cent by number of forms for both inward and outward inquiries. 31. The estimation methodology is essentially to project forward earlier annual returns to the latest year, using a comparison of the results for each component and sizeband. However, in precise methodological terms there are exceptions to this rule as follows: Large Enterprise Groups 31.1 Very large groups that do not respond have their results manually constructed. This relies heavily on the officer responsible obtaining information from a wide range of sources, such as the press, group publications and the internet. For each component, recourse is also taken to the previous annual or quarterly figures (if available) or data from the acquisitions & mergers inquiry or published group results. For example, annual accounts will be sought from Companies House and consolidated balance sheets / profit & loss accounts analysed. The

actual construction values require a detailed appraisal of the group’s performance and results over the past year, and the officer concerned is guided by his/her experience and investigative skills. Constructions with high values (for example, over £50m) are double-checked for accuracy. New Groups 31.2 Groups approached for the first time do not have previous data so if they fail to respond, their results are also manually constructed from the best information available to FDI (usually the acquisitions and mergers inquiry mentioned earlier, accounting data or press releases). 32. Values are estimated for other businesses which did not respond or which were not selected. Generally these are produced by applying factors to the relevant business’ responses (or estimates) in the previous year’s inquiry. Within each stratum the factor for each component is based on the aggregate growth rate of groups responding to both current and previous inquiries. Unless there are exceptional reasons these factors are expected to lie within the range 0.3 to 1.7. By viewing the factors, exceptional amendments can be made and used in the current estimation round. Responses or estimates that produce extreme or ‘outlier’ data are closely examined for accuracy and congruence with previously recorded data and compatibility with published national accounts statistics. 33. Once the database contains information for all groups in the population, results can be analysed according to component, country and industry simply by aggregating records. 34. Sample surveys are employed rather than censuses because to do otherwise would make the cost to business and government prohibitive and considerably lengthen the time taken to produce the results. However associated with sample surveys are sampling errors. Measurements of sampling errors illustrate the spread of estimates which would be expected from successive samples selected by chance from the same population using the same sample specification. Whilst each sample is designed to produce the “best” estimate of the true population value, a number of equal-sized samples covering the population would generally produce varying population estimates. The sampling error measures this variation. 35. The sampling errors are generally small for the aggregates of the main inquiry components and indeed the sample is specifically designed to achieve this. An example of the sampling errors is shown in the following table. UK direct investment

abroad foreign direct investment in UK

1997 Investment flows

Levels of investmen

t

Earnings Investment flows

Levels of investmen

t

Earnings

Standard error (£ million)

1091 2422 415 1379 1395 236

Coefficient of variation (%)

2.8 1.1 1.4 6.3 0.9 1.8

36. As an example of interpretation, there are approximately 19 chances in 20 that the real value of net earnings from direct investment in the UK is within two standard errors, that is, £472 million of the inquiry estimate. In other words, if the estimate of earnings from inward investment is £13,160 million, we can be confident that the real value lies within the range £12,688 million to £13,632 million. 37. However, in addition to sampling errors there is the potential for non-sampling errors which cannot be easily quantified. Examples where these may occur are deficiencies in the inquiry register and errors made by contributors that are not detected by the inquiry processing team.

Quarterly Direct Investment Inquiries 38. The quarterly inquiries are used to project forward estimates from the latest available annual inquiry results. The first quarterly figures are prepared nine weeks after the end of the reporting quarter, by which time response targets should have been achieved. A revised estimate is prepared with the following quarter’s estimate 13 weeks later. These estimates are replaced by the results of the annual inquiry when they become available. As already mentioned fewer questions are asked in the quarterlies than the annuals (for example, no geographical breakdown is collected), results are produced for a short sectoral breakdown only (see paragraph 15) and publication is with other balance of payments and national accounts information. 39. The quarterly inquiry sample design was improved in 1995 when the inquiries switched from using a cut-off sample (that is, only large groups above a specified size were included) to a stratified rotational approach similar to the annual inquiry. The inward inquiry includes all groups where the level or stock of investment in the UK part of the group is greater than or equal to £20 million at net book value, ignoring the sign. The outward inquiry includes all groups where the stock of investment in the foreign affiliate is greater than £50 million, again ignoring the sign. Table 2 below shows the sampling fractions used in the quarterly inquiries: Table 2: Sampling fractions for the quarterly direct investment inquiries

OUTWARDS INWARDS Net book value (£m) (ignoring sign)

Sampling fraction Net book value (£m) (ignoring sign)

Sampling fraction

5 - 10 1:8 5 - 10 1:6 10 - 20 1:3 10 - 20 1:2 20 - 50 1:2 20 - 50 1:1 > 50 1:1 > 50 1:1 40. The quarterly questionnaires are despatched at the end of the reporting quarter with a request that the data be returned within one month. After one month has elapsed, all non-responding groups are reminded by telephone of the need to reply and at the same time they

are sent a reminder letter. A second reminder is issued a further two weeks later. This reminder is in the form of a statutory notice instructing the group to comply. However, as the key objective is to obtain a high response by the end of week 8, extensive telephone fax and email contact occurs between weeks 5 and 7. Target response rates for the inquiry teams are 74 per cent for the outward inquiry and 85 per cent for the inward inquiry, both by number of forms. 41. Imputation for individual non-responders is carried out using a procedure developed within ONS. The box below shows the technical specification: For each component within each stratum for each contributor responding in both current and previous periods, Difference = (value in period) – (value in previous period) Discard the top 10% and bottom 5% of differences: Link value = 1 x (remaining differences) n [where n = number of contributors left] The link value is then added to the non-responding contributors’ values for the previous quarter. The link value itself may be positive or negative, but any negative values for dividends resulting from the addition are overwritten by a zero value. 42. Extreme values are removed from the procedure used to estimate for groups not selected in the inquiry. They are identified using the ‘I’ value method, which has been in use in ONS for several years. This methodology is suitable for the quarterly FDI inquiries because of its ability to calculate the impact of an extreme value in either the 'base' period or the current period. The base period value for each contributor is the average quarterly response in the year covered by the latest available annual inquiry results. The technical details of the I value calculation are shown in the box below.

Calculation of I-values The I-values are computed for each component in each stratum as: QR - qr QB ABS x - 1 x 100 QB – qb QR where QR = sum of responses for current quarter QB = sum of quarterly base values qr = individual contributor response for current quarter qb = individual contributor base value

If the result of the above formula is greater than 500/n, where n is the number in the quarterly sample with a ‘response’ value (good returns and imputed values), then the value is identified as an outlier. (This methodology was improved in 1998 to take account of stratum size; previously a value was considered to be an outlier if the result of the formula was greater than 2). 43. The procedure compares individual current values with the total value in the current period and also compares individual values in the base period with total base period response. The I value is a measure of the impact each contributor is having on the results. Once identified the outlier is excluded from the procedure used to estimate for groups not selected in the sample but it is added back into the result at the end of the calculation; it is thus given a weight of one (that is, its own weight only). 44. The estimation procedure calculates the growth of the quarterly contributors between the current and base periods and applies this to the whole of the population, allowing for outliers calculated previously. The quarterly estimation formula is as follows: QR – RE

Estimate = x ( TB - BE ) + RE QB - BE

QR = sum of responses for current quarter QB = sum of quarterly base values RE = sum of outliers in current quarter BE = sum of outliers in base TB = total base value for population COUNTRY DATA HELD AND ALLOCATION TO GEOGRAPHIC & ECONOMIC

ZONES EUROPE

EU Austria Belgium & Luxembourg Denmark Finland France Germany Greece Irish Republic Italy Netherlands Portugal

Spain Sweden EFTA Iceland *Liechtenstein Norway Switzerland OTHER EUROPEAN COUNTRIES *Albania *Andorra *Belarus *Bosnia and Herzegovina Bulgaria Channel Islands and Isle of Man *Croatia Cyprus Czech Republic *Estonia Gibraltar Hungary *Latvia *Lithuania *Macedonia,Fmr Yugo Rep Malta *Moldova Poland Romania Russia Serbia & Montenegro *Slovakia *Slovenia Turkey *Ukraine *Vatican City State AMERICA Antigua & Barbuda Argentina *Aruba Bahamas Barbados *Belize Bermuda Bolivia Brazil *British Virgin Islands & Montserrat Canada

Cayman Islands Chile Colombia *Costa Rica *Cuba Dominica *Dominican Republic Ecuador *El Salvador *Falkland Islands *Grenada *Guatemala Guyana *Haiti *Honduras Jamaica Mexico Netherlands Antilles Nicaragua Panama Paraguay Peru *Puerto Rico *St Kitts & Nevis and Anguilla *St Lucia *St Vincent *Surinam Trinidad & Tobago *Turks & Caicos Islands Uruguay *US Virgin Islands USA Venezuela ASIA NEAR AND MIDDLE EAST COUNTRIES *Armenia *Azerbaijan *Gaza & Jericho *Georgia Iran Israel Jordan Lebanon Syria GULF ARABIAN COUNTRIES *Abu Dhabi

*Bahrain *Dubai Iraq Kuwait *Oman Other Gulf States Qatar Saudi Arabia Yemen OTHER ASIAN COUNTRIES *Afghanistan Bangladesh *Bhutan *Brunei *Cambodia (Kampuchea) China Hong Kong India Indonesia Japan *Kazakhstan *Korea, North Korea, South *Kyrgyzstan Laos *Macao Malaysia *Maldives *Mongolia Myanmar (Burma) *Nepal Pakistan Philippines Singapore Sri Lanka Taiwan *Tajikistan Thailand *Turkmenistan *Uzbekistan *Vietnam AUSTRALASIA & OCEANIA *American Oceania Australia *Australian Oceania *Fiji

*Kiribati *Nauru New Zealand *New Zealand Oceania Other Oceania Papua New Guinea *Pitcairn *Polar Regions *Solomon Islands *Tonga *Tuvalu *Vanuatu *Western Samoa AFRICA Algeria Angola *Benin Botswana *British Indian Ocean Territory *Burkina Faso *Burundi Cameroon Canary Islands *Cape Verde *Central African Republic *Chad *Comoros *Congo Dem Rep of the Congo (Zaire) *Djibouti Egypt *Equatorial Guinea *Eritrea Ethiopia Gabon Gambia Ghana *Guinea *Guinea Bissau *Ivory Coast Kenya Lesotho Liberia Libya *Madagascar

Malawi *Mali *Mauritania Mauritius Morocco Mozambique Namibia *Niger Nigeria *Rwanda *Sao Tome & Principe Senegal Seychelles & dependencies Sierra Leone *Somalia South Africa St Helena & dependencies Sudan Swaziland Tanzania *Togo Tunisia Uganda Zambia Zimbabwe OECD Australia Austria Belgium & Luxembourg Canada Czech Republic Denmark Finland France Germany Greece Hungary Iceland Irish Republic Italy Japan Korea, South Mexico Netherlands New Zealand

Norway Poland Portugal Spain Sweden Switzerland Turkey USA CENTRAL & EASTERN EUROPE *Albania *Bosnia and Herzegovina Bulgaria *Croatia Czech Republic *Estonia Hungary *Latvia *Lithuania *Macedonia, Fmr Yugo Rep Poland Romania Serbia & Montenegro *Slovakia *Slovenia Data are available for the countries listed in this annexe subject to restrictions to protect confidential data relating to individual enterprise groups. * separate country level data not available prior to 1995. data not available prior to 1997. INDUSTRIAL DETAIL COLLECTED IN THE ANNUAL INQUIRIES Agriculture, forestry & fishing 01 Agriculture 02 Forestry 05 Fishing Mining & quarrying (including oil & gas production) 10 Coal/peat mining 11 Oil/gas production/distribution 12 Mining of uranium/thorium ores 13 Mining of metal ores 14 Other mining & quarrying

Food products 15 Food & drink production 16 Tobacco products Textile & wood, printing & publishing 17 Textile products 18 Manufacture of clothing 20 Wood/timber products 21 Paper/cardboard products 22 Printing & publishing Chemicals, plastic & fuel products 23 Manufacture of coke, refined petroleum products & nuclear fuel 24 Chemical products 25 Rubber & plastic products Metal & mechanical products 27 Manufacture of basic metals 28 Manufacture of fabricated metal products 29 Manufacture of machinery/equipment nes Office, IT & communications equipment 30 Office, IT equipment 32 Communications, radio, TV equipment Transport equipment 34 Manufacture of vehicles 35 Manufacture of other transport equipment Other manufacturing 19 Leather products 26 Glass, cement & ceramic products 31 Electrical products 33 Instrument & control products 36 Furniture & other manu nes 37 Recycling of waste materials Electricity, gas & water 40 Electricity/gas production/distribution 41 Water collection, purification, distribution Construction 45 Construction services Retail/wholesale trade & repairs 50 Sale & repair of vehicles 51 Wholesale distribution

52 Retail distribution Hotels & restaurants 55 Hotels, restaurants & catering services Transport & communications 60 Land transport services 61 Water transport services 62 Air transport services 63 Transport support services 64 Postal & telecommunications services Financial services 65 Financial intermediation services 66 Insurance services 67 Other financial intermediation services Real estate & business services 70 Real estate services 71 Equipment rental services 72 Data processing & consultancy services 73 Research & development services 74 Business services Other services 75 General public service activities 80 Educational services 85 Medical & social services 90 Sanitation services 91 Activities of membership organisations 92 Entertainment & recreation 93 Other services nes 95 Domestic services 99 Extra territorial organisations Data are available for all the 2 digit SIC(92) codes above, as well as more detailed industry level data for the following codes: 65-Financial intermediation services 651 Non banking intermediation services 652 Banks 66-Insurance services 661 Life (long term) insurance services 662 Non life (general) insurance services

67-Other financial intermediation services 671 Security dealers 672 Other auxiliary financial services 74- Business services 741 Holding companies 742 Other business services Data are available for the industrial breakdown listed in this annexe subject to restrictions to protect confidential data relating to individual enterprise groups.

1

Tuesday 20th January 2015 Issued by Office for National Statistics, Government Buildings, Cardiff Road, Newport, NP10 8XG Media Office 08456 041858 Business Area 01633 455923

Lead Statistician: Michael Hardie

Quality and Methodology Information

General details Title of output: Foreign Direct Investment

Abbreviated title: FDI Designation: National Statistics Geographic coverage: UK and the World

Date of last SQR or QMI: 29 July 2013 Contact details: FDI public enquiry team

1

Executive summary

The purpose of the Foreign Direct Investment (FDI) survey is to collect financial data on assets and liabilities. The survey collects information on an annual and quarterly basis between foreign parent companies and their UK affiliates (Inward FDI) as well as between UK parent companies and their foreign affiliates and branches (Outward FDI). FDI is a major driver of globalisation. As investment patterns of multinational enterprises become more complex, reliable and internally comparable FDI statistics are necessary for sound policy making. The following components are collected: a) Earnings: income from investments, eg profits, interest and tax; b) Flows: changes within investments, eg reinvested earnings, acquisitions, disposals and loan movements; and c) Positions: these data represent an enterprise’s total investment value at a point in time, of equity capital and other capital. Positions data are then used to calculate the international investment position. The FDI outputs were previously published on the ONS website in two annual publications; the Annual Foreign Direct Investment Statistical Bulletin

2 and the Foreign Direct Investment MA4

Business Monitor3. From January 2015, these releases have been consolidated into one Foreign

Direct Investment publication.

Quality and Methodology Information' (QMI) replaced 'Summary Quality Reports' (SQR) from 04/11

Inward FDI Outward FDI

2

This information is primarily required for measuring the UK’s Balance of Payments and it’s International Investment Position (IIP). It is also used to measure the international investment patterns between the UK and the rest of the world, either by country or groups of countries. This document contains the following sections:

Output quality;

About the output;

How the output is created;

Validation and quality assurance;

Concepts and definitions;

Other information, relating to quality trade-offs and user needs; and,

Sources for further information or advice. Output quality (This document provides a range of information that describes the quality of the output and details any points that should be noted when using the output.) ONS has developed Guidelines for Measuring Statistical Quality

4 these are based upon the five

European Statistical System (ESS) quality dimensions. This document addresses these quality dimensions and other important quality characteristics, which are:

Relevance;

Timeliness and punctuality;

Coherence and comparability;

Accuracy;

Output quality trade-offs;

Assessment of user needs and perceptions, and;

Accessibility and clarity. More information is provided about these quality dimensions in the sections below.

About the output Relevance (The degree to which the statistical outputs meet users’ needs) The FDI survey collects financial information relating to direct investment in the UK by enterprises located abroad (inward FDI) and direct investment abroad by enterprises located in the UK (outward FDI). In either case the foreign investment must be at least 10% of the ordinary shares or voting power. Data on investments less than 10%, which are classed as portfolio investments, are collected as part of the Balance of Payments by the Structural and International statistics division within ONS. These data serve a range of purposes for example:

Inward FDI data indicate which sectors of the economy are of interest to foreign competitors and where long-term growth is expected to be; and

Outward FDI data indicates which countries of the world are expected to be growth areas and where a company might wish to invest.

The current FDI database holds full country/industry aggregate data from 1972 to the current published year for both the Inward and Outward FDI surveys. Limited country and industry level data are available for the years 1958 to 1971 from within historical publications produced by the Department of Trade in Industry (DTI) now known as the Department for Business Innovation and Skills (BIS

5).

For more details on the Foreign Direct Investment Survey Statistical Bulletins, the UK National Statistics Publication Hub

6 is available online and provides 12 months advance notice of release

dates. In the unlikely event of a change to the pre-announced release schedule, public attention are drawn to the change and the reasons for the change should be explained fully at the same time, as set out in the Code of Practice for Official Statistics

7.

For additional information on FDI see the International Monetary Funds’ (IMF) Balance of Payments manual 6 (BPM6)

8.

3

Data for the banking sector are collected by the Bank of England; all other sectors are covered by ONS surveys. The banking surveys collect information from all banks, whereas ONS surveys are based on samples. This Quality and Methodology Information concentrates on these sample surveys. These surveys are statutory, collecting information under the Statistics of Trade Act 1947

9.

Change in measurement Survey History

1990 - surveys became statutory. 1997 - ESA95

10/BPM5

11 requirements adopted.

2002 - sample redesigned to include enterprises identified from Dun & Bradstreet (D&B)12

data. 2008 - methodological review following a discrepancy with the 2006 data. 2009 - the IMF introduced a set of data requirements as part of the Co-ordinated Direct Investment

Survey13

. The purpose of these new data requirements was to provide comprehensive and

comparable data across countries’ Inward and Outward direct investment.

2010 - changes to the Standard Industrial Classifications (SIC’s), converting from Standard

Industrial Classification 2003 (SIC 2003) to Standard Industrial Classification 2007 (SIC 2007), were

implemented to the FDI annual 2009 and 2010 quarterly surveys. Detailed SIC 2007 information is

available on the ONS Website14

.

2010 - the FDI quarterly and annual questionnaires were enhanced to include additional questions, implementing the 2009 IMF data requirements. The new requirements required the collection of debt between Fellow Enterprises. These changes represented the first steps towards the introduction of the Balance of Payments manual 6 (BPM6)

8 and the Organisation for Economic Co-Operation and

Development (OECD) Benchmark definition of Foreign Direct Investment (4th edition)15

. 2011-2012 - work began to ensure that the surveys became compliant with ESA(10)

16,

Benchmark definition of Foreign Direct Investment (4th edition)15

and the Balance of Payments

manual 6 (BPM6)8 as required by Eurostat, the IMF and OECD.

2014 - first annual FDI publication compliant with BPM68.

Timeliness and punctuality (Timeliness refers to the lapse of time between publication and the period to which the data refer. Punctuality refers to the gap between planned and actual publication dates.) The Annual Foreign Direct Investment (FDI) Statistical Bulletin

2 publication is published 12 months after

the reference period (in December of the following year). For more details on related releases, the UK National Statistics Publication Hub 6 is available online and provides 12 months’ advance notice of release dates. If there are any changes to the pre-announced release schedule, public attention are drawn to the change and the reasons for the change will be explained fully at the same time, as set out in the Code of Practice for Official Statistics

7.

How the output is created The processes for creating the outputs for the FDI surveys are detailed in the following table:

Sample frame There are two separate sampling frames for the FDI surveys. These are explained as follows: Sample Frame 1 - Outward population - approximately 6,000 enterprise groups. This population contains the details of all wholly owned UK Businesses with ownership of subsidiaries and branches in foreign countries. Sample Frame 2 - Inward population - approximately 20,900 enterprise groups. This population contains the details of UK businesses, who themselves are subsidiaries and or branches of a foreign company. The sampling frames are produced from three sources:

4

the historical FDI inward and outward registers of known and proved businesses;

IDBR - which is primarily compiled from administrative information, such as VAT and PAYE details from HM Customs and Excise . The register holds a number of variables including information on the country of ownership for each enterprise group. It also holds information on which UK groups have affiliates. The main source of information for these foreign links is the Dun & Bradstreet (D&B)

12 ‘WorldBase’® information. This is a commercial data

source purchased by ONS;

the Cross Border Mergers and Acquisitions surveys - which are conducted on a continuous basis, collecting UK company information on acquisitions and disposals of a foreign company and vice versa when a foreign company acquires or disposes of a UK company.

Sample design Sampling of enterprise groups for inclusion in the FDI surveys is based on stratified designs for the known FDI historical population and the source provided by the D&B

12 ‘WorldBase’® population.

For the known FDI population, the outward and inward strata are defined in terms of the value of the net international investment position (Net Book Value).

For the ‘WorldBase’® population, the outward survey strata are defined in terms of the number of foreign affiliates held by the UK parent company.

For the inward ‘WorldBase’® population, the strata are defined in terms of the turnover of the UK group.

For the annual population except for inwards ‘WorldBase’® population, all groups in the top stratum (containing the largest businesses) are sent questionnaires in order to maximise the survey coverage of foreign direct investment assets

For annual inwards and quarterly inwards and outwards for the ‘WorldBase’® population no top strata are fully enumerated.

In strata containing smaller businesses, only a proportion are selected. Additionally, the sample of smaller businesses is rotated to minimise burden on the respondents.

Sample size The sample sizes for the Foreign Direct Investment surveys are as follows: Annual Outward - approximately 2,100 enterprise groups; Annual Inward - approximately 3,500 enterprise groups; Quarterly Outward - approximately 680 enterprise groups; Quarterly Inward - approximately 970 enterprise groups. All businesses in the quarterly sample are included in the annual sample.

Data collection FDI surveys are collected using paper questionnaires or Secure Electronic File Transfer (SEFT) - Excel spreadsheets. These are generally sent to the head of the enterprise group within the UK and request consolidated information for the group as a whole. Both data collection instruments are despatched to businesses in March/April after the end of the reference year.

Response Two written reminders are subsequently sent to non-responding groups and are followed by telephone, fax and electronic mail reminders to try to minimise non-response and thereby any non-response bias. There is also the possibility of

using the legal powers of the Statistics of Trade Act 19479 to force response,

though ONS prefers to work together with businesses to produce the necessary information. Response rates are analysed on a weekly basis for both the quarterly and annual surveys. The number of questionnaires received and cleared after validation is documented. Response rates are also analysed by proportion of Net Book Value received and by industry sector and sizeband. See Standard

Industrial Classification (SIC)14

.

Response Rate Targets by Net Book Value (NBV):

5

Annual Outward - 75%; Annual Inward - 85%; Quarterly Outward - 50%; Quarterly Inward - 75%. These targets will be reviewed in future with respect to the new, larger sample sizes.

Validation Returned information is run through a series of checks to identify errors. These tests ensure that: 1. all questions are completed; 2. responses to individual questions are consistent within the questionnaire as

a whole; 3. the return is consistent with historical data from the business or differences

are explained; 4. acquisition and disposals data for equity capital are compared with data

produced by the Mergers and Acquisitions’ survey.

Data clearance Any errors in FDI data are fully investigated and corrected in order to ensure that the data outputs are as accurate as possible before final results are produced. This process is known as ‘cleaning’ of the data. Therefore realistic clearance target levels are set to achieve good quality data. The clearance target levels, by Net Book Value (NBV)are listed below: Annual Outward - 40%; Annual Inward - 50%; Quarterly Outward - 30%; Quarterly Inward - 60%.

Outliers The distance from the mean methodology is used. This involves calculating the mean and standard deviation within each stratum. The data are trimmed first so that any extreme values do not influence the calculation of the mean or standard deviation. The top 15% and bottom 5% are trimmed where there are more than 8 observations with non-zero data within each stratum. The remaining data within each stratum is then used to calculate the mean and standard deviation. An interval is then formed, +/- 4 standard deviations either side of the mean, and then applied to all returned data. Anything that falls outside the range is identified as an outlier. The identified outliers are kept in the sample data and only represent themselves in the estimation. They are excluded from the calculation of the mean for the estimation.

Imputation The imputation methods used vary depending on the variable being measured. The methods used include:

a ratio of means method for calculating the imputation link. This involves multiplying the previous period value by an “imputation link”, a value calculated to give an estimate of the ratio between the previous value and the current value. In a ratio of means method, this imputation link is derived from the ratio between the stratum mean for the current and previous period, for the stratum containing the value to be imputed. In some cases, if the previous value was negative, the imputed value is set to zero;

a trimmed stratum mean method, where the top 15% and bottom 5% of values are not included in the calculation of the stratum mean, are used to calculate some imputed values, particularly where no previous period data was available;

a median value within a given stratum;

copying forward imputing previous period values; and

Imputing zero values. The different methods were chosen based on their suitability for the particular variable. For example, a ratio of means estimator typically performs well for variables which have a strong correlation with their previous values, whereas for

6

variables that are strongly correlated with other values in their stratum, a trimmed mean or median could be used.

Estimation Estimation for FDI assigns a trimmed mean value to each of the non-sampled businesses within a given stratum. These are then added up, along with the sampled data, to give the population estimate. The reason for this approach is that the estimation takes in to account any clustering of responses (more so in outwards). If distance from the mean is used for outliering, then the outlier is excluded from the calculation of the mean. The mean is calculated for each stratum and then applied to every non-sampled unit in the population, and these values are added together to find an estimate for the total population.

Disclosure Statistical disclosure control methodology is applied to FDI data. This ensures that information attributable to an individual or individual organisation is not identifiable in any published outputs. The Code of Practice for Official Statistics

7,

and specifically the Principle on Confidentiality17

set out practices for how we protect data from being disclosed. The Principle includes the statement that ONS outputs should ‘ensure that official statistics do not reveal the identity of an individual or organisation, or any private information relating to them, taking into account other relevant sources of information.’ More information can be found in National Statistician’s Guidance: Confidentiality of Official Statistics and also on the Statistical Disclosure Control Methodology page on the ONS website

18.

Revisions to annual data on the ONS Web page

19.

Further details of the coverage of the FDI surveys are covered in the FDI Statistical Bulletin

2.

Annual time cycle

Selection February/

March

Despatch forms April

Close for

Provisional

results

end July

Construct for non-

responders early

September Close for

final results

Mid September

Bottom-line

Approved

Early October

Re-bench

marking for

quarterly

Mid October

FDI

Statistical

Bulletin

Published

December

A cycle illustrating processing

of the Annual FDI

surveys

7

Validation and quality assurance Accuracy (The degree of closeness between an estimate and the true value.) Sampling error Sampling error is the error caused by observing a sample instead of the whole population. While each sample is designed to produce the ‘best’ estimate of the true population value, a number of equal sized samples covering the population would generally produce varying population estimates. The sampling error measures this variation. Sample surveys are employed rather than censuses in order to balance the burden on businesses and cost of collection, processing and quality, against the relative accuracy of an estimate instead of a population value. Standard errors are an estimate of the sampling error and provide a measure of the precision of the estimate, and a low standard error indicates a precise estimate. To aid comparison, the standard error is also expressed as a percentage of the total value. This quantity is called the coefficient of variation and it allows the standard errors to be put into context. A high coefficient of variation indicates a greater relative variation between the true population value and the estimated value from the sample. Non-sampling error

In addition to sampling errors, there is the potential for non-sampling error that cannot be easily quantified. One potential source of non-sampling error is non-response, which relates to the failure to obtain data from some businesses selected in the sample. Another source of non-sampling error may be as a result of undetected deficiencies that may occur in the survey register and errors or inconsistencies may be submitted by the contributors when completing the survey questionnaires. The FDI revisions, response rates and estimated standard errors are available in the Basic Quality Information in the background notes of the FDI Statistical Bulletin

2 and

the FDI MA4 Business Monitor

3.

Another aspect of accuracy is reliability, which refers to the closeness of the initial estimated value(s) to the revised estimated value(s), and can be thought of as a measure of consistency between the initial and revised estimates. The reliability of key indicators in this release can be obtained by monitoring these revisions. Comparability and coherence (Comparability is the degree to which data can be compared over time and domain eg geographic level. Coherence is the degree to which data that are derived from different sources or methods, but refer to the same topic, are similar.) Comparability Since 1997, various changes have occurred in the coverage of the survey. The European System of Accounts (ESA(95))

10 definitions were introduced to the 1997 MA4 Business Monitor

3, allowing

comparison across the European Union. The changes were as follows:

prior to 1997 for the measurement of direct investment, an effective voice in the management of an enterprise was taken as the equivalent of a 20% shareholding. Since 1997, this is now 10%;

the Channel Islands (Jersey, Guernsey etc.) and the Isle of Man have been excluded from the definition of the economic territory of the UK. Prior to 1997, these islands were considered to be part of the UK. Their data are published under the UK Offshore Islands title in the data tables within the FDI Statistical Bulletin

2; and

since 1997, interest received or paid was replaced by interest accrued in the figures on earnings from foreign direct investment. There is deemed to be little or no impact arising from this definitional change on the estimates.

The FDI international investment position data have been collected annually since 1987. A further change caused by the move to ESA(95)

10 was how withholding taxes payable on direct

investment earnings were measured. Earnings were first shown gross of these taxes in the FDI First Release 2004 and the Balance of Payments 2004 Pink Book

20 and were calculated net of tax until

2006. Recently the updated European System of Accounts (ESA(10)16

) definitions have also been introduced. All of the above changes from ESA(95)

10 are still included in it, in addition to some other

changes which have been incorporated into the relevant sections of this document. More recently the FDI survey has undergone changes as a result of new requirements laid out in the

8

latest BPM68 manual. The main changes to the survey include:

the move to an Extended Directional Principle where loans to fellows are taken into consideration when calculating both inward and outward positions. These were previously excluded under BPM5.

publishing data on an asset and liability basis where previously only directional FDI was available i.e. inward, outward.

producing estimates both including and excluding special purpose entities (SPE’s) which are defined as legal entities with little or no employment and are ultimately controlled by an overseas parent company.

To implement these changes, the questionnaire was thoroughly reviewed with the addition of new questions and a number of existing questions were revised. The first survey year in which data has been collected using the revised questionnaire is for 2012. Both the 2013 estimates and revised 2012 estimates will be published in January 2015. Further information can be found in an article titled - Impact on the Balance of Payments as a result of the introduction of BPM6.

21

Coherence

The ONS FDI surveys are the only source of UK FDI data that complies with FDI international definitions.

Mergers and Acquisitions Statistical Bulletin22

provides cross border equity transaction data for the

respondents selected on the annual and quarterly FDI surveys. FDI equity data includes transactions

above a 10% ownership threshold. The FDI surveys report those deals, or parts of deals, funded

directly by parent companies and exclude any deals, or parts of deals, funded by locally raised funds

(either in the UK for Inward FDI or abroad for Outward FDI).

Concepts and definitions (Concepts and definitions describe the legislation governing the output, and a description of the classifications used in the output.) Statistics of Trade Act 1947

9.

ESA(95)10

. ESA(10)

16

IMF’s BPM68.

Eurostat23

. OECD

15.

Standard Industrial Classification14

.

Other information Assessment of user needs and perceptions (The processes for finding out about users and uses, and their views on the statistical products.) Data from the FDI surveys are used by a wide range of users. The key users are:

ONS Balance of Payments and National Accounts teams;

Other Government Departments eg Department for Business Innovation and Skills5, Her

Majesty’s Treasury (HMT)24

, Department For International Development (DfID)25

, Her Majesty’s Revenue and Customs (HMRC)

26, Cabinet Office

27, British Foreign and Commonwealth Office

(FCO)28

;

Eurostat23

;

Organisation for Economic Cooperation and Development15

; United Nations Committee on Trade and Development (UNCTAD)

29; and

International Monetary Fund30

.

Additional users are UK and foreign academics and students, UK & foreign Embassies and Trade Missions, businesses, UK and foreign press and the general public. The annual survey data are published initially in an FDI Statistical Bulletin

2 which is freely available

9

from the National Statistics website (text and spreadsheet versions).

Further queries can be addressed to the FDI public enquiry team1.

Sources for further information or advice Accessibility and clarity (Accessibility is the ease with which users are able to access the data, also reflecting the format in which the data are available and the availability of supporting information. Clarity refers to the quality and sufficiency of the release details, illustrations and accompanying advice.)

ONS's recommended format for accessible content is a combination of HTML web pages for narrative,

charts and graphs, with data being provided in usable formats such as CSV and Excel. The ONS

website also offers users the option to download the narrative in PDF format. In some instances other

software may be used, or may be available on request. Available formats for content published on the

ONS website but not produced by the ONS, or referenced on the ONS website but stored elsewhere,

may vary. For further information please refer to the contact details at the beginning of this document.

For information regarding conditions of access to data, please refer to the links below:

Terms and conditions (for data on the website)31

;

Copyright and reuse of published data32

;

Pre-release access(including conditions of access)33

;

Basic Quality Information relevant to each release is available in the background notes of the relevant Statistical Bulletin FDI Statistical Bulletin

2.

Further queries can be addressed to the Foreign Direct Investment public enquiry team at: fdi@ons.gsi.gov.uk

1.

Useful links

National Statistician’s Guidance : Confidentiality of Official Statistics34

.

Basic Quality Information in the Foreign Direct Statistical Bulletin35

.

FDI International definitions36

.

Balance of Payments and National Accounts37

.

National Statistics Website38

. References

Reference No Link to Website

1. FDI public enquiry team fdi@ons.gsi.gov.uk

2. Annual Foreign Direct Investment Statistical Bulletin

http://www.ons.gov.uk/ons/rel/fdi/foreign-direct-investment/index.html

3. MA4 Business Monitor http://www.ons.gov.uk/ons/rel/fdi/foreign-direct-investment/index.html

4. Guidelines for Measuring Statistical Quality

http://www.ons.gov.uk/ons/guide-method/method-quality/quality/guidelines-for-measuring-statistical-quality/index.html

5. Department for Business Innovation and Skills

http://www.bis.gov.uk/

6. UK National Statistics Publication Hub

http://www.statistics.gov.uk/hub/index.html

7. Code of Practice for Official Statistics

http://www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html

8. International Monetary Funds’ Balance of Payments Manual 6

http://www.imf.org/external/pubs/ft/bop/2007/bopman6.htm

9. Statistics of Trade Act 1947 http://www.legislation.gov.uk/ukpga/Geo6/10-11/39/contents

10. The European System of Accounts (ESA(95))

http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/Glossary:European_system_of_national_and_regional_accou

10

nts_(ESA95) 11. International Monetary

Funds’ Balance of Payments Manual 5

http://www.imf.org/external/np/sta/bop/bopman5.htm

12. Dun & Bradstreet http://dnb.com.au/Credit_Reporting/The_quality_of_DandBs_data/WorldBase/index.aspx

13. Co-ordinated Direct Investment Survey

http://www.imf.org/external/np/sta/cdis/index.htm

14. ONS Website http://www.ons.gov.uk/ons/guide-method/classifications/current-standard-classifications/standard-industrial-classification/index.html

15. Organisation for Economic Co-Operation and Development Benchmark definition of Foreign Direct Investment (4th edition)

http://www.oecd.org/document/33/0,3343,en_2649_33763_33742497_1_1_1_1,00.html)

16. The European System of Accounts (ESA(10))

http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/Glossary:ESA_2010

17. Principle on Confidentiality http://www.ons.gov.uk/ons/guide-method/census/2011/confidentiality/index.html

18. Statistical Disclosure Control Methodology page on the ONS website

http://www.ons.gov.uk/ons/guide-method/best-practice/disclosure-control-policy-for-tables/index.html

19. Revisions to annual data on the ONS Web page

http://www.ons.gov.uk/ons/guide-method/revisions/revisions-and-corrections-policy/index.html

20. Balance of Payments 2004 Pink Book

http://www.ons.gov.uk/ons/search/index.html?newquery=Balance+of+Payments+2004+Pink+Book

21. Impact on the Balance of Payments as a result of the introduction of BPM6

http://www.ons.gov.uk/ons/rel/naa1-rd/national-accounts-articles/impacts-of-changes-to-the-measurement-of-the-balance-of-payments-as-a-result-of-the-introduction-of-bpm6/index.html

22. Mergers and Acquisitions Statistical Bulletin

http://www.ons.gov.uk/ons/search/index.html?newquery=Mergers+and+Acquisitions

23. Eurostat http://epp.eurostat.ec.europa.eu/portal/page/portal/eurostat/home/

24. Her Majesty’s Treasury http://www.hm-treasury.gov.uk/ 25. Department For

International Development http://www.dfid.gov.uk/

26. Her Majesty’s Revenue and Customs

http://www.hmrc.gov.uk/index.html

27. Cabinet Office http://www.cabinetoffice.gov.uk/ 28. British Foreign and

Commonwealth Office http://www.fco.gov.uk/en/

29. United Nations Committee on Trade and Development

http://unctad.org/en/Pages/Home.aspx

30. International Monetary Fund http://www.imf.org/external/index.htm 31. Terms and conditions (for

data on the website) http://www.ons.gov.uk/ons/site-information/information/terms-and-conditions/index.html

32. Copyright and reuse of published data

http://www.ons.gov.uk/ons/site-information/information/creative-commons-license/index.html

33. Pre-release access(including conditions of access)

http://www.ons.gov.uk/ons/guide-method/the-national-statistics-standard/code-of-practice/pre-release-access/index.html

34. National Statistician’s Guidance : Confidentiality of Official Statistics

http://www.statisticsauthority.gov.uk/national-statistician/ns-reports--reviews-and-guidance/national-statistician-s-guidance/index.html

35. Basic Quality Information in the Foreign Direct Statistical Bulletin

http://www.ons.gov.uk/ons/rel/fdi/foreign-direct-investment/2010-release/index.html

36. FDI International definitions http://www.oecd.org/document/33/0,3746,en_2649_33763_33742497_1_1_1_1,00.html

37. Balance of Payments and http://www.ons.gov.uk/ons/taxonomy/index.html?nscl=Balance

11

National Accounts +of+Payments

38. National Statistics Website http://www.ons.gov.uk/ons/index.html

Summary quality report for Foreign Direct Investment Annual surveys releases 1 Introduction

ffice for ical Quality

This report is part of a rolling programme of quality reports being introduced by ONational Statistics (ONS). The full programme of work being carried out on Statist 1

ary Quality Reports are overview notes which pull together key qualitative information on the various dimensions of quality as well as

by ONS, measure the flows of investment, earnings from investment and the international investment

) companies' affiliates abroad (outward) and conversely in the UK (inward).

is available on the National Statistics website. Summ

providing a summary of methods used to compile the output. This report relates to Foreign Direct Investment (FDI). FDI surveys, conducted

positions for United Kingdom (UKforeign companies' affiliates 2 Summary of quality 2.1 Relevance The degree to which the statistical product meets user needs for both coverage and content Foreign Direct Investment refers to investment that adds to, deducts from, or acquireinterest in an enterprise operating in an economy other than that of the invespurpose being to have an "effective voice" in the management of

.

s a lasting tor; the investor's

rise. For the an effective voice is taken as equivalent to a holding of 10 per

cent or more of the ordinary shares or voting power for any incorporated subsidiary or associate nally

the enterppurposes of the statistical survey,

enterprises or the equivalent for unincorporated branch enterprises. (This internatiorecognised definition is taken from the International Monetary Fund’s (IMF) Balance of Payments Manual2 version 5 (BPM5)). Balance of Payments investments consist of Foreign Direct Investments, Portfolioand Other Investments. Investments in which the investor does not have an effective the management of the e

Investments voice in

nterprise (i.e. they hold less than 10 per cent of the ordinary shares or voting power) are mainly portfolio investments; these are out of scope of the FDI surveys. The

d “reserve t of scope of

the IMF’s BPM5

Data for the banking sector are collected by the Bank of England; all other sectors are , whereas

Quality ing

information under the Statistics of Trade Act 1947. Detailed Standard Industrial Classification (SIC) information is available on the National Statistics website

other components of balance of payments investments are “other investment” anassets”; these cover investments where no ownership is held. These are also outhe FDI surveys. Definitions of all of these types of investment are available inpublication mentioned above.

covered by ONS surveys. The banking surveys collect information from all banksONS surveys are based on samples and cover all other industries. This Summary Report concentrates on these sample surveys which are statutory surveys, collect

3 The surveys are conducted using questionnaires which are generally sent to the head of the enterprise group within the UK and which request consolidated information for the group as a whole. The survey is divided into two parts: one measures FDI by UK groups abroad (outward survey) and the other measures FDI in the UK by foreign groups (inward survey).

FDI surveys What it measure ent companies

Investment Flows, estment Positions (i.e. Stocks or Levels) and

Investment Earnings.

s Financial information relating to transactions between parand their affiliates. Headline data components are International Inv

Frequency Annual

Sample size Outward survey approximately 1,400 enterprise groups Inward survey approximately 2,500 enterprise groups

Periods available egate data urrent published year for both the inward and outward

ta are available for

The current FDI database holds full country/industry aggrfrom 1972 to the cFDI surveys. Limited country and industry level dathe years 1958 to 1971

Sampling frame e Inter

oups ups

Historical standalone FDI register supplemented by thDepartmental Business Register (IDBR).

Outward sampling frame approximately 2,700 enterprises grInward sampling frame approximately 14,000 enterprises gro

Sample design andom sampling sitions values (Net

The additional IDBR sample (known as the WorldBase sample) is ign

er’ for inward FDI. These samples are mutually exclusive.

The FDI population samples are selected by stratified rusing previously returned International Investment PoBook values)

selected using stratified random sampling on ‘number of foresubsidiaries’ for outward FDI and ‘UK turnov

Weighting umber

sinesses using business Net Book values.

Each respondent in the stratified samples is used to represent a nof similar bu

Imputation on period as 1.7 and

previous data

are manually rvey or

accounts in preference to imputed data based on sectoral variations. Otherwise the data is imputed for using average

All non-responders are imputed for individually using period ratio imputation. Imputation links larger than 1.7 are takenthose smaller than 0.3 are taken as 0.3. (where there arevalues available) If previous returns are not available, large non-respondersconstructed using data sources from the matching quarterly sutheir published

returned values.

Estimation n-sampled estimation for the entire population at

detailed data levels.

After imputation average returned values are applied to nobusiness data to provide full

O l ypical or extreme returns are treated as outliers

from the average value calculations. The top and rise group returns are excluded as outliers

lations.

ut iers Any respondents with atand are excludedbottom 5 per cent of enterpfrom the average value calcu

Data from the FDI surveys are used by a wide range of users. The key users are:

• Balance of Payments and National Accounts • Other Government Departments e.g.

o Department of Trade and Industry (DTI) o Foreign and Commonwealth Office (FCO) o Department For International Development (DfID)

o Her Majesty’s Revenue and Customs (HMRC) • Eurostat4 • Organisation for Economic Cooperation and Development (OECD)5 • United Nations Committee on Trade and Development (UNCTAD)6

& foreign Embassies and Trade Missions, businesses, UK and foreign press and the general public.

Additional users are UK and foreign academics and students, UK

2.2 Accuracy

ss between an estimated result and the (unknown) true value. The closene Sampling Error Sample surveys are employed rather than censuses because the census processlengthy and costly to be viable in this monitor. Standard errors illustrate the spwhich would be expected from estimates derived from successive samples selectchance from the same population using the same sample specification. While eacdesigned to produce the

is too read of results,

ed by h sample is

‘best’ estimate of the true population value, different equal-sized samples covering the population would generally produce varying population estimates. The

easures this variation. In the FDI surveys this standard error is estimated for

Non Sampling Error

sampling error mthe Worldbase sample only.

r example,

the

ference year. nt to non-responding groups and are followed by

sponse and owers of the

gether with ecessary information.

ese include questions

historical data from the business.

nitial estimated evised estimated value(s). An indication of reliability of the key indicators in

this release can be obtained by monitoring these revisions. Please note: FDI is subject to a revisions policy.

the Basic tional

There is the potential for non-sampling errors that cannot be easily quantified. Foundetected deficiencies may occur in the survey register and errors may be made by respondents when completing the survey questionnaires. Questionnaires are despatched to businesses in March after the end of the reTwo written reminders are subsequently setelephone, fax and electronic mail reminders to groups to try to minimise non-rethereby any non-response bias. There is also the possibility of using the legal pStatistics of Trade Act to force response, though the Office prefers to work tobusinesses to produce the n

Returned information is run through a series of checks to identify any errors. Thtests to ensure that all questions are completed, that the responses to individualare consistent within the questionnaire as a whole, and that the return is consistent with

Another aspect of accuracy is reliability. This refers to the closeness of the ivalue(s) to the r

The FDI revisions, response rates and estimated standard errors are available inQuality Information in the first release. For other FDI accuracy information see NaStatistics website7. 2.3 Timeliness and Punctuality

ich the data nned dates of publication.

The National Statistics Release Calendar

Timeliness refers to the lapse of time between publication and the period to whrefer. Punctuality refers to the time lag between the actual and pla

8 is available on the National Statistics website and provides twelve months advance notice of releases. The FDI surveys consistently meet publication deadlines. A First Release publication is published 12 months after the reference period, and a more detailed Business Monitor publication 14 months after the reference period.

2.4 Accessibility and Clarity Accessibility is the ease with which users are able to access the data, also reflectiformat(s) in which the data are available and the availability of supporting inform

ng the ation. Clarity

refers to the quality and sufficiency of the metadata, illustrations and accompanying advice.

Release The annual survey data are published initially in a Foreign Direct Investment First 9

t versions).

More detailed data are available in the MA4 Business Monitor

which is freely available from the National Statistics website (text and spreadshee

10 which is freely available from

Limited country and industry level data for the years 1958 to 1971 are available from historical D cations. These publications are:

• 1968 to 1971, DTI Trade & Industry extracts.

be addressed to the FDI public enquiry team at: fdi@ons.gov

the National Statistics website (text and spreadsheet versions).

epartment of Trade and Industry (DTI) paper publi

• 1958 to 1967, DTI Board of Trade Journal extracts; and

Further queries can 11

2.5 Comparability The degree to which data can be compared over time and domain. Since 1997 various changes have occurred in the coverage of the survey. The new European System of Accounts (ESA(95))12 definitions were introduced with the 1997 M

itor, allowing comparison across the EuropeaA4 business

mon n Union. The changes were as follows:

n the

this is now 10 per cent.

een excluded e islands were

heir data are published under the UK Offshore Islands title in the data tables.

the figures on earnings ising from

on data have only been collected annually since 1987,

d by the move to ESA(95) is that withholding taxes payable on direct se taxes in the

monitor

s available from 1998 have led to revisions to the figures from that year onwards. These give an improved estimate of the population satisfying the criteria for Foreign Direct Investment.

The definitional changes were introduced from 1997 only and the new register source changes were introduced from 1998. The data prior to these years have not been reworked in the MA4 Business Monitor

i) Prior to 1997 for the measurement of direct investment, an effective voice imanagement of an enterprise was taken as the equivalent of a 20 per centshareholding. Since 1997,

ii) The Channel Islands (Jersey, Guernsey etc.) and the Isle of Man have bfrom the definition of the economic territory of the UK. Prior to 1997 thesconsidered to be part of the UK. T

iii) Interest received or paid was replaced by interest accrued in

from foreign direct investment. There is deemed to be little or no impact arthis definitional change on the estimates.

The FDI international investment positiprior to that date these data were only collected via a triennial census survey. A further change causeinvestment earnings are now measured. Earnings were shown gross of theBalance of Payments 2004 Pink Book. However, for the purposes of this businessearnings are calculated net of tax, as before.

New FDI register source

10. For clarity, the breaks in the series are identified separately in Chapter 1, tables 1.1 to 1.3.

2.6 Coherence The degree to which data that are derived from different sources or methods, but which refer to the same phenomenon, are similar.

FDI survey is the only source of UK FDI data that complies with FDI international

team also carry out quarterly surveys covering the same survey populations. The rst

The ONSdefinitions. The FDIresults from these surveys are published in the quarterly Balance of Payments FiRelease13. There are key differences in the coverage between the annual FDI surveys andFDI surveys as the balance of payments compilers have access to sources ounsuitable for use

the quarterly f data that are

in the annual surveys due to their lack of geographic detail. For example Public data on private investment in property in the UK or abroad and investment by UK

Corporations abroad. Also the Mergers and Acquisitions14 survey which provides data for the annual aFDI surveys treats the data it collects differently from the datasets published insurvey. FDI covers all transactions above a 10% ownership threshold whereas M&A covers only those transactions where majority ownership is involved (i.e. where the 50barrier is crossed during a transaction. Minority deals (above or below the 50% obarrier) are not published by the M&A surve

nd quarterly either FDI

% ownership wnership

y but would be included in the FDI survey. Also ectly by parent

(either in inward FDI or abroad for outward FDI).

the UK’s national

accounts) are not covered by the FDI annual survey, but are shown in the Balance of Payments

the FDI surveys would only report those deals, or parts of deals, funded dircompanies and exclude any deals, or parts of deals, funded by locally raised fundsthe UK for

FDI by Public Corporations or in Property (which is regarded as FDI in

Pink Book15.

3 Summary of methods used to compile the output Coverage FDI data are measured by industrial sector, sub components of the main data components and the country of the immediate foreign parent investor (inwards FDI) or the country of the

s A, B and C in

immediate foreign affiliate (outwards FDI). Further details of the coverage of the FDI surveys are covered in Appendicethe MA4 Annual Business Monitor10 Sample Design The FDI population samples are selected using previously returned InternationaPositions values (Net Book values). All respondents with an absolute Net Bookthan £50 million are selected automatically and the rest of the sample arstratified random sampling using Net Book values.

l Investment Value of more

e selected by

ubsidiaries’ for ch survey

the relevant

ation. Statistical Disclosure

The IDBR sample (WorldBase sample) is selected using ‘number of foreign soutward FDI and ‘UK turnover’ for inward FDI. The largest 100 are selected for eaand the rest of the sample is selected using stratified random sampling usingstratification variable. The IDBR foreign links information is updated annually using information from Dun & Bradstreet’s “WorldBase”® public

Statistical disclosure control methodology is applied to the FDI annual survey data. This ensures that data attributable to an individual respondent are not disclosed in any publication or dataset. The National Statistics Code of Practice and specifically the protocol on Data Access and Confidentiality set out the principles of how we protect data from being disclosed. For further information, see: National Statistics Code of Practice16.

The protocol includes a guarantee to survey respondents that “no statisticsthat are likely to identify an individual unless specifically a

will be produced or more

tion on ONS statistical disclosure control methodology see: Statistical Disclosure greed with them”. F

informaControl17. 4 Refe

ce ion

rences

Title of Referen Website locat1 istical Quality Programme http://www.statistics.gov.uk/about/data/methodology/quality/default.asp Stat 2 ments ?sk=157 IMF Balance of Pay

BMP5 http://www.imf.org/external/pubs/cat/longres.cfm

3 003 Classification umentation

4012SIC 2Doc

http://www.statistics.gov.uk/statbase/Product.asp?vlnk=1

4 at 1090,30070682,1090 Eurost http://epp.eurostat.cec.eu.int/portal/page?_pageid=_33076576&_dad=portal&_schema=PORTAL

5 _1_1_1,00.html OECD http://www.oecd.org/home/0,2987,en_2649_201185_1_1 6 D http://www.unctad.org/Templates/StartPage.asp?intItemID=2068 UNCTA 7 Investment lnk=179&Pos=&ColR Foreign Direct http://www.statistics.gov.uk/StatBase/Source.asp?

surveys web source document

vank=1&Rank=1000

8 ases.asp Release Calendar http://www.statistics.gov.uk/ReleaseCalendar/currentrele 9 gn Direct Investment

irst Release http://www.statistics.gov.uk/StatBase/Product.asp?vlnk=728&Pos=1&Col Forei

surveys F Rank=1&Rank=272 10 Foreign Direct In

surveys MAvestment

4 Business 614&Pos=6&C http://www.statistics.gov.uk/StatBase/Product.asp?vlnk=9

Monitor olRank=1&Rank=272

11 ntacts FDI co fdi@ons.gov 12 European System of

Accounts http://forum.europa.eu.int/irc/dsis/nfaccount/info/data/esa95/en/esa95en.htm

13 lance of Payments First tatistics.gov.uk/StatBase/Product.asp?vlnk=1118&Pos=1&C Ba http://www

Release .s

olRank=1&Rank=272 14 rs and Acquisitions

t http://www.statistics.gov.uk/about/data/methodology/quality/information_ Merge

Summary Quality Repor business_statistics.asp 15 Balance of Payments Pink

Book p://www.statistics.gov.uk/StatBase/Product.asp?vlnk=1140&Pos=6&Chtt

olRank=1&Rank=272 16 National Statistics Code of

Practice http://www.statistics.gov.uk/about/national_statistics/cop

17 Statistical Disclosure Control http://www.statistics.gov.uk/about/data/methodology/general_methodology/sdc.asp

How to avoid inflated data in FDI statistics and improve on their geographical breakdown: better meeting user needs

Frank E.M. Ouddeken Balance of Payments and Financial Accounts Department

De Nederlandsche Bank

July 2006

______________________________________________________________________________

The opinions amd arguments brought forward in this paper are those of the auther only. No single statement in this paper necessarily reflects the opinion and/or position of the Netherlands Bank, nor of any other institution.

Abstract: Current FDI statistics show ever increasing flows and stocks of cross border investment, clearly representing the process of globalization. However, multinational enterprises establish ever more entities for sub-holding and financing activities across an increasing number of countries. Real global networks are created. Through these networks investments are channelled through a number of countries, ever more evaporating the 'direct' character of FDI. Funding operations, organised by the ultimate controling owner (UCO) from somewhere within its global network of enterprises, have modified the investment character of FDI in an ever more financing character. As a result, current FDI statistics show to a large extent Foreign Indirect Financing operations. In many countries large FDI inflows and outflows are recorded that are strongly related to each other due to passing-through and subholding operations. To an extent of which users of the statistics are often totally unaware. Due to the network through which the investments flow, sight on the ultimate controler and ultimate beneficary is also lost for the user. Consequently, the relation between FDI statistics and the real economy has disappeared. Nevertheless, most users still use FDI data assuming that FDI statistics provide useful information for the measurement of the impact of FDI on the real economy. In order to better meet user needs a proposal is made for a slight modification of some of the basic concepts of FDI in order to compile more sensible FDI statistics. As a result the inflation in the FDI data will be ‘reduced’ tremendously and more sensible information about the ultimate origin and destination of FDI can be provided also.

1. Introduction Currently the major international statistical manuals are being reviewed and updated. The process is well underway, as drafts are already circulating and being commented on. Both the concepts and methodology of the System of National Accounts (SNA) of the UN, the Balance of Payments Manual (BPM) of the IMF, and the Benchmark Definition of Foreign Direct Investment (BMD) of the OECD are reconsidered, although not in a fundamental way. The manuals are, so to speak, “refreshed”, on the basis of recent experiences and contemporary problems encountered by the compilers and users. The responsible international organisations, UN, IMF and OECD, have organised a delicate structure of working, experts and advisory groups Reference is made to: the AEG of the UN, IMF’s BOPCOM and BOPTEG, the OECD’s WIIS and BAG, and the joint group DITEG on FDI statistics. for the revision processes, in which also the European Central Bank (ECB), National Central Banks and National Statistical Institutes have been actively participating, with the aim of harmonising these statistical manuals to the largest extent possible. One striking feature of the whole revision process of FDI statistics is the relatively limited involvement of the users in these discussions. Although large weight was given to the perceived user needs, problems still remain in defining statistical concepts that would be closely in line with the concept of FDI as intuitively understood by the users. So far, these deliberations have resulted in a decision not to change fundamentally the ‘core accounts’, but to provide the compilers the opportunity to compile various types of so-called supplementary presentations of FDI statistics, potentially better meeting user needs. This paper aims at providing some background information on the issue of the user needs of FDI statistics and the problems users are confronted with in analysing FDI by using the statistics as they are currently compiled. Various “solutions” have been suggested by a large number of participants of the debate. So far, no solutions have yet been found that were convincing to the majority of the participants of the debate. This paper gives a short overview of some of the proposed “solutions” and an evaluation of them. It will end with a proposal for a modification in the way FDI statistics are presented, which may prevent users from drawing misleading conclusions from FDI statistics in economic analyses. 2. User needs and user problems in the area of FDI statistics In its contribution to the revision process the ECB undertook, on behalf of its Working Group on External Statistics (WG-ES), an investigation to the user needs of FDI statistics. Based on a large inventory of recent scientific economic studies, discussions with specialists and consultations of its business areas, the ECB concluded that FDI statistics as such play only a limited role in the monetary analysis of the Balance of Payments. The same applies to the analysis of the relationship between the exchange rate and the Balance of Payments. FDI statistics were, however, used to some extent for studying the relationship between FDI and taxation policy and the financial policy of multinational groups. Not surprisingly, the vast majority of uses of FDI statistics are related to the analysis of the impact of FDI on economic growth, employment and productivity. In other words: FDI statistics are mainly used for real economic analyses instead of financial analyses. It is therefore highly relevant to answer the question to what extent current FDI statistics suit these needs. In January 2005 the ECB organised an ad-hoc workshop on FDI statistics. The workshop identified three major problems in current FDI statistics See also DITEG outcome paper 11B (http://www.imf.org/External/NP/sta/bop/2005/op11b.pdf) .:

1. FDI data are substantially increased on both the inward and the outward investment side due to sub-holding and financing operations of resident entities, often established by non-residents. This is especially true for countries that host so-called Special Purpose Entities (SPEs Special Purpose Entities consists of many different types of entities established (mostly in another country) for specific purposes like: funding, subholding, financing, treasury operations, factoring, securitization, etc. ); 2. Sight on the final destination and the ultimate origin of direct investment is lost due to ownership chains that cross a number of countries; 3. Some data in non-equity FDI (intercompany debt) are ‘distorted’ Distorted means: resulting in strongly reduced or even negative (net) positions! due to the funding activities of various kinds of specialized entities in the form of reversal loans to the parent company.

It was concluded that for each of these problems different solutions would be necessary and a number of recommendations was made, such as: Separate identification of SPEs (as a sub-sector); Supplementary geographical breakdowns based on the concept of the Ultimate Beneficial Owner (UBO) and Ultimate Beneficial Affiliate (UBA); Exclusion from FDI of reverse investments made by a financial affiliate to the parent. These recommendations look promising and provide to some extent solutions to current problems. Nevertheless, they also raise some questions with regard to the interrelatedness of the various proposals. Do these recommendations sufficiently solve the current problems of the users of FDI statistics? How does a UBO/UBA geographical breakdown of FDI data relate to the standard presentation of FDI? Is there a fundamental difference between transactions of SPEs and FDI operations of ‘regular’ (i.e. non-SPE) companies? Is the concept of FDI still defined clearly enough for the users after having excluded various types of investment from FDI?

3. Current treatment of FDI in statistics – from a historical birds-eye view In the early years of BOP compilation (1950s and 1960s), certain financial transactions took place directly between a resident enterprise and a related non-resident enterprise, its affiliate. Or better: between, what one started to call, a direct investor in one country and a direct investment enterprise in a second country. The concept of Foreign Direct Investment was born. In those days FDI was mainly related to “one-to-one investments”, which meant that by the investment a direct relation was established between (only) two economies. Initially, these transactions were mainly restricted to equity investment. But after the gradual relieve of capital restrictions these types of operations were steadily broadened to include the direct provision of various kinds of intercompany loans also. From a conceptual point of view, at that time, there was no sound argument for limiting these non-equity transactions to certain types of loans only. Therefore, the concept of FDI came to encompass all types of intercompany equity and debt operations. However, in the perception of many users (and also compilers) FDI statistics were still based on a single one-to-one relationship, which was not ‘wide of the mark’ with reality at that time. During the 1970s and 1980s multinational enterprises became a real widespread and global phenomenon. Many multinational enterprises set up regional headquarters in the form of sub-holding companies in the process of structuring their growing global networks. Sometimes this was done in the form of the establishment of Special Purpose Entities with hardly any real presence in the host economy. But this was far from being the rule. Many regular sub-holdings, like regional headquarters, having real presence in the host economy also, were established as well; for exactly the same purposes and performing exactly the same operations as SPEs. Tax-treaty shopping found its start in those days in order to optimize tax liabilities of the entire group. The further removal of restrictions on cross border financial transactions allowed the free flow of intra-group credit on a worldwide scale. It induced the establishment of financing companies in other countries than the head-office, because of comparative advantages of access to capital markets, and in order to exploit regulatory differences between countries to the benefit of the entire group. During these decades the nature of FDI progressively changed from “one-to-one investments” to “chain investments”, with many indirect links between related enterprises. This was equally true for both equity and debt operations. In the early 1990s capital market liberalisation and the abolition of cross border capital restrictions was almost completed in the industrialized world. To optimize their financial performance further many multinational enterprises established treasury centres operating for their worldwide group. Deregulation stimulated the development of many specific banking products like “netting”, “pooling” and “zero-balancing”, further boosting the gross flows between related enterprises. At the end of that decade this process resulted in the development of so-called “in-house banks” and “payment factories” responsible for the entire financing and settlement of all payments of the entire group. Also tax-treaty shopping reached ever higher levels of sophistication in the form of “round-tripping”. FDI was now circling around across a number of countries, creating a dichotomy between the real and financial economy. 4. Concerns and shortcomings of the current FDI statistics This evolution of investment and financing behaviour of multinational enterprises has fundamentally changed the character of the financial flows between related enterprises. However, both the measurement and presentation of FDI statistics has not really been adjusted to reflect this change One exception (to some extent) that is worth mentioning, is the exclusion from FDI of non-equity operations of SPEs with a sole purpose of serving in a financial intermediary capacity (BPM5; §365). As most SPEs serve many purposes at the same time this exclusion hardly solves the problems

related to the distortion in FDI statistics created by SPEs in practice.. The single “one-to-one” relation between the direct investor and the direct investment enterprise seems to have become an exception in FDI operations nowadays. Most transactions, in both equity and debt, are currently undertaken on a “hop, step and jump” basis through a chain of investments, passing through many entities established across a number of countries, which are used as stepping stones. Funding operations are preferably organised from foreign affiliates that are located in the most liquid and profitable/deregulated markets, and the collected funds are passed through the global network of the related enterprises. As a result, nowadays indirect investment seems to dominate the FDI statistics of many countries. What statisticians currently measure as Foreign Direct Investment has to a large extent become Foreign Indirect Investment. With it, even the character of these FDI operations has fundamentally changed. The “investment” character of FDI (in its meaning of being an active ‘stakeholder’ of another country) has eroded and is to a large extent exchanged for a “financing” character (in the meaning of passively holding assets and liabilities). So, FDI statistics could be regarded ever more as statistics on “Foreign Indirect Financing”. As a result, current data on direct investment hardly provide the users with sensible results; the data no longer have resemblance with the mere intuition of the users of what FDI really is. The data are, therefore, hardly applicable to sound (real) economic analysis. All in all, the “raison d’être” of the whole concept of FDI is at stake. A very clear example in this case is Luxembourg. In 2004 its FDI positions reached levels of € 650 bln (outward) and € 700 bln (inward). 95% of both these positions was related to entities that were identified as (foreign owned) SPEs. A second example is the Netherlands. End of period FDI stocks in 2005 amounted to € 1,600 bln on the outward investment side and some € 1,225 bln on the inward side, reaching levels comparable to that of the USA! Approximately 1/3 of the Dutch FDI positions is related to entities that are classified as SPEs, as they hardly intervene with the production process of the domestic economy. Most compilers in other countries are unable to discern the specific nature of the entities that are established in other countries. So, they cannot distinguish between foreign SPEs and regular enterprises. As a result these countries show large FDI positions with Luxembourg and the Netherlands. However, the real ‘vested interests’ are much smaller in reality. A clear illustration of its consequences was the exclusion of Luxembourg and the Netherlands from a FDI study undertaken by the ECB, due to “the seemingly disproportionate share” of the two countries in the euro area FDI. 5. Possible alternative treatments A lot of proposals have been made by many participants of the revision debate in trying to solve the problems in FDI statistics and especially the problems related to SPEs. The debate has been lively and fruitful in a sense of sharing a growing and more thorough understanding of the problems. Nevertheless, the results obtained so far do not seem to have created a general feeling by the participants of the debate that the real problems in FDI statistics will really be solved. To some extent this seems to be caused by the partial character of the approach followed. We may therefor, run the risk of concluding on partial solutions. This could either easily jeopardize the basic concept of FDI, due to the erosion of its intuitive content, or it may result in various different presentations of FDI data, which may cause confusion by the users of what the real data on FDI are (describing reality!). This would most certainly be to the detriment of the compilers also! In these discussions many ideas have been brought forward for solving the problems of the users. In the following rough overview some of the most important ideas will be briefly discussed and evaluated in general terms:

A. Transferring certain types of investment from FDI to Portfolio and/or Other Investment: - Short term debt: The idea behind this proposal is that only long term debt is aimed at having a lasting influence on the affiliated direct investment enterprise and its hosting economy. Short term assets and liabilities are supposed to distore the real picture of FDI. However, as most of intercompany debt is extended on a short-term (even daily!) basis nowadays, this would almost result in reducing FDI to equity operations only. Moreover, many credits are extended to an affiliated enterprise on a revolving basis, which effectively results in long lasting financing/investment by means of a short term instrument. - Reversal loans: These types of proposals are meant to provide a solution to the problem of obtaining strongly reduced or even negative (net) positions due to funding operations performed by (financial) SPEs that have been established in a foreign country (funding entity, e.g. a conduit). If the proceeds of a loan issue are transferred to the parent company it is proposed to exclude it from FDI; if however the proceeds are onlend to other affiliates they would still included in FDI. This would impair the clearness of the concept of FDI. B. “Looking Though” certain types of entities: - Financing companies and/or conduits: It has been proposed to reclassify intercompany loans received from a financing company that is especially established for that kind of activity as either Portfolio or Other Investment, depending on how the entity has financed itself (by issuing bonds or non-tradable loans). A drawback of this solution is the need to reclassify the assets of the “conduit” according to the type of instument on its liabilities side. Moreover, looking through techniques almost always result in large bilateral discrepancies between the countries concerned, because of a lack of information on the specific nature of the entities concerned and the type of transactions ‘behind’ it. C. Exclusion of certain types of entities: - Exclusion of all types of Other Financial Institutions: Debt operations of banks are traditionally excluded from FDI. This was extended to transactions and positions between all types of affiliated financial intermediaries. Further narrowing the range of FDI by excluding other financial intermediaries (like in-house banks, specialised entities to obtain financing and conduits) from FDI was considered to be to harmfull for the concept of FDI. D. Separate presentation of certain types of entities (SPEs): - In all discussions this proposal survived as having great appeal and is regarded as at least a part of ‘the’ solution. A solution already followed by some countries in their current publications (like e.g. Luxembourg and the Netherlands). The major drawback often expressed, is the lack of reciprocity on a bilateral level: the counterparty often lacks sufficient information for the identification of SPEs. E. Identification of “capital in transit” - Attempts are undertaken to define “capital in transit” in order to capture the passing through operations of all types of entities, both regular entities and SPEs, at the same time. Key elements in the proposed definition of the concept of capital in transit are the non-transformation of the capital in the passing-through process (e.g in terms of maturity, currency denomination, risk base, etc). So far, it has not yet resulted in a concept that is clear-cut and that can be applied in real pratice of data collection. The major reason for this disappointing outcome is that the very nature of all financial operations and the ‘raison-d’être’ of the establishment of specialized entities is the explicit desire to transform one financial title into the other! F. Netting FDI operations: - A kind of netting procedure makes sense if both assets and liabilities are inflated due to the overwhelming phenomenon of passing-through operations. In fact, netting may also be considered as a form of “consolidation”. Consolidation is closely in line with one of the most basic concepts in the Benchmark Definition of the OECD, in which the Fully Consolidated System (FCS) is at the core of the whole framework of FDI statistics.

- Netting can be done at the level of the individual enterprises and at the level of the individual transactions. This presupposes the existence of a one-to-one relationship between the assets and liabilities of an entity engaged in FDI. This perception of netting shows in fact great similarity with the concept of capital in transit and therefor also shares the drawbacks of it. - Netting can also be done at the macro level. It has been proposed to redefine the concepts of inward and outward investment. Inward investment would be redefined as all FDI operations of resident entities having a non-resident Ultimate Beneficial Owner (UBO); outward investment would be redefined likewise as all FDI operations of all resident entities having a resident UBO. A perceived drawback of this netting procedure is the risk of “netting away” too much from FDI statistics; even running the risk of creating negative positions that will be difficult to interpret. A common characteristic of all these proposals is the following: they all try to link in some way assets and liabilities of an entity that performs a transformation function in the process of intercompany financing. This is done by separating either entities that are specialised in this function, or by the exclusion of certain instruments that are most suited for that purpose, or by defining concepts that are explicitly based on the perception of a link between assets and liabilties. However, in reality that link does not really exist! In the end, each asset is to some extent financed by all liabilties (including the equity) of the entity. So, strictly speaking, we can only arrive at one sad conclusion: the statisticians will be unable to solve the problems of the users of FDI statistics entirely! Therefor, one has to accept that FDI statistics will necessarily capture more than the users intuitively expect. However, it does not imply that nothing can be done to make major improvements to FDI statistics. 6. Towards a kind of ‘solution’ All proposed solutions so far have been evaluated in isolation. However, a combination of two specific elements may be of great help to the users of FDI statistics: 1. A breakdown by Ultimate Control As both inward FDI and outward FDI can be undertaken by entities of “foreign based” and “domestically based” multinational enterprises, the concept of inward and outward direct investment has become rather difficult to interpret by the users, due to the passing-through of capital. As said before, passing through of capital is not at all resticted to SPE-like entities only. Regular production enterprises, either foreign or domestically based, may show exactly the same behaviour as foreign SPEs do. Of primary importance for the correct understanding of FDI data by the users, seems to be a breakdown of FDI between investment ultimately decided on by a resident Ultimate Controling Owner The Ultimate Controling Owner of a company is the entity that is on ‘top’ of the hiërarchy of a network of enterprises to which that company belangs and has the majority of voting rights (either directly or indirectly); it can be supposed to exercise at least in theory its powers to manage the company and can therefor be held responsible for all decisions taken by that company. (UCO) and investments made by entities having a non-resident UCO Of course not all direct investment enterprises will have an Ultimate Controling Owner. In case of minority ownerships only (up to 50:50 joint-ventures) a third category should be identified as “Uncontroled Ownership”. In most countries this category will be negligibly small.. This separation should consistently be made in all presentations of FDI for both inward and outward investment. The concept of the UCO is very close that of the UBO. UCOs are those UBOs that have a majority ownership (or possess majority voting rights), both direcly and indirectly. 2. Separation of SPEs The second step is the consistent separation of FDI undertaken by foreign owned SPEs that are established in the compiling country. The purpose of this separation is to show the magnitude of inward and outward FDI that is performed by entities that are specialized in the passing-through of capital.

The combination of these two elements is quite powerful. Reference is made to tables 2 and 3 of the annex of this paper for an example. The introduction of the split in FDI between FDI of entities having a resident UCO and those having a non-resident UCO will re-establish the link between the concept of FDI and the intuitively understood content of the concept by the users. As both inward and outward FDI of entities having a resident UCO will be shown explicitly, the magnitude of their passing-through operations will become apparent on the inward side of their FDI operations. The origin of the funds collected can be made apparent by e.g. the geographical breakdown, etc. At the same time, FDI undertaken by entities having a non-resident UCO will still be ‘distorted’ by the operations of the SPEs that they have established in the compiling country SPEs are supposed to be established by non-resident UCOs only. . Separate presentation of these operations is of great value for the users in all circumstances. The better the compiler is able to identify SPEs, the more informative these FDI data will be. However, non-SPE entities of foreign origing might still engage in passing through operations, together with ‘real’ investments. It will be impossible to distinguish between the two comprehensively. The result of these two adjustments in the presentation of FDI statistics will be that the inflation of FDI data will be made explicit to the users by splitting the data to various aspects. With it the users will get information on the ‘real’ investments made and the magnitude of the ‘financial’ passing-through operations at the same time. Both aspects of FDI seem to be of importance to analists. In the annex to this paper an example is worked out for the proposed breakdowns. For countries not having SPEs it suffices to publish table 2 only. For countries that host SPEs table 3 is a necessary step to provide usefull statistics to the users on a macro level. These two breakdowns wil restore to a large extent the usablity of FDI statistcs for the users. With it the necessary preparations are made at the same time for a consistent application of the third element that aims at improving the sight on the origin and destination of FDI: 3. A refined geographical breakdown As the users have lost sight on the ultimate origin and final destination of FDI, it will be easy to restore this insight on a conceptual level, once the breakdown by ultimate control has been introduced in FDI statistics: a. For Inward FDI operations of entities having a non-resident UCO both the country of the UBO and the Direct Counter Party (DCP) should be identified. For Inward FDI of resident UCOs only the DCP is unnecessary. b. For Outward FDI operations the compiler should try to collect the DCP and the country of the Ultimate Beneficial Affiliate; for investments made by non-resident UCOs the country of the UBO should be collected in addition.

With this triplet of changes the compilers of FDI statistics will meet the user needs quite close, without having to change the most basic concepts of FDI. FDI statistics will provide almost full information to the users of the financial structures behind the global investment networks in which a country takes part. Even negative net numbers due to reverse investments can be accepted, because almost full information is provided to the user of the origin behind it. The user needs will be met much closer, although not fully, as the identification of SPEs, UCOs/UBOs and UBAs will not allways be easy and because some foreign based entities wil combine real investment and passing-through operations within the same entity. 7. Concluding remarks This paper dealt with some of the most difficult and most intensively debated problems in the compilation of statistics nowadays. An attempt was made to bring the discussion on step further in the direction of better meeting the user needs in statistics on Foreign Direct

Investment. Many proposals for solutions have been made in the discussions during the last three years in the international working and expert groups dealing with these issues. At the end of this paper it was concluded that a final solution for the problems that have been identified seems to be nonexistent, as the perceived ‘distortion’ of FDI statistics seems to be caused by the very nature of the investment process itself: the transformation of one asset into another asset by routing it through a separate entity that constitutes a part of the global network of related enterprises. As statisticians cannot step away from that reality in their hunt for better meeting user needs, compilers will be forced to show the details of the global investment processes in order to restore the understanding and usability of the statistics they compile. This result can be obtained by breaking down FDI on an Ultimate Controling Ownership basis together with a separation of SPE operations. As current FDI statistics are to a large extent misleading, it should be given serious consideration to include these breakdowns even in the core accounts.

AnnexTable 1: Various types of FDI

Entities with a Entities with a

Non-Resident UCO Resident UCO

INWARD OUTWARD INWARD OUTWARD

E: 140 E: 23 E: 21 E: 500A: Eq: regular inv

140 23 21 500

D: 800 D: 100 D: 250B: Db: regular inv

800 100 250

E: 1600 E: 1600C: Eq: subholding Not applic

D: 1970 D: 1970 D: 2240 D: 2240D: Db: onlending

(intra-group)

4575 2380E: 2235 E: 910D: 2140 D: 1380

E: Eq/Db: funding

200 90

E: 66 E: 17 E: 32F: Eq: cross partic

66 17 32

D: 456 D: 123 D: 567G: Db: reversal loans

456 123 567

Legend

External economy

Eq = Equity entityDb = Debt

Domestic economy

= SPE

Virtual Micro Data Laboratory Data Brief 4: Winter 2007

Globalisation and Multinational Enterprises: Insights from the Annual Inquiry into Foreign Direct Investment

Robert Gilhooly The Annual Inquiry into Foreign Direct Investment (AFDI) is concerned with the investment flows of multinational enterprises between their UK and international operations. Multinationals and the increased interdependence of firms in the global economy have been at the forefront of academic and policy debate for several years. While migration often grabs the newspaper headlines, Foreign Direct Investment is a primary driver changing both developed and developing economies. Global FDI flows have grown at least twice as fast as trade, now well exceeding $500 billion and resulting in a total stock of more than $8 trillion (United Nations, 2004). This AFDI data set has been used for an extensive range of research, much of which has directly contributed to the globalisation literature. This data brief provides an overview of the AFDI data set that is accessible to researchers through the Virtual Microdata Laboratory (VML). We firstly summarize the content of the AFDI survey including: the years in which the survey was conducted, the main topics covered by the survey, the years for which data is available, the available sample size and the nature of links that can be made to other ONS data sets. We then provide a brief summary of previous research undertaken using this data. Finally, as the AFDI data set is used in research primarily as a marker to identify multinationals we present some descriptive statistics on these firms.

1. The AFDI survey

The AFDI is conducted in two parts: an inward inquiry and an outward inquiry. The inward inquiry concerns the subsidiaries/associates of foreign firms operating in the UK, while the outward inquiry covers the investment made by UK firms in their overseas operations. Firms are asked to provide information on a variety of aspects of their business. Notable areas include: country of ownership/investment, profit & loss, earnings, tax credits, sales/purchases of shares/loans, and gains/losses resulting from movements in exchange rates. The information collected by the AFDI measures direct investment as a financial concept, and is not the same as capital expenditure on fixed assets. Direct investment is only recorded when it is made in a “lasting interest”, and this deemed to be the case

when the firm owns more than a 10% equity stake in the company in which it is investing. Direct investment is defined in this way to avoid measuring international portfolio flows.

2. Sample Size The survey forms are sent to the head of enterprise groups in the UK requesting information for the group as a whole. The register from which the firms are sampled comes from sources including HM Customs & Revenue, Dunn & Bradstreet’s “Worldbase” system, and ONS inquiries on Acquisitions & Mergers. The sampling is based on a stratified design; the largest firms all receive the survey form, while only a proportion of the smaller firms are sent the survey forms. Below is the actual number of firms that are in the data files 1996 to 2004 and the number of observations. The sum of the observations is greater than the number of firms, primarily, because in the “out” files enterprise groups may have many subsidiaries/branches. It is also possible that each enterprise group in the “in” files may have more than one owner. At its conception the AFDI survey was relatively small; only 2000 and 1000 enterprises were surveyed in the inward and outward inquiries respectively. Changes in the registry system in 1998 and again in 2001 resulted in a large rise in the number of enterprises covered – over 13,000 firms are now covered by the inward inquiry.

Year 1996 1997 1998 1999 2000 2001 2002 2003 2004 Observations “inwards”

2255 2359 6664 8022 8767 14238 13999 13411 13873

Observations “outwards”

8420 8814 11168 12963 12917 13393 12375 11798 10580

Total observations

10675 11173 17832 20985 21684 27631 26374 25209 24453

Total firms per year (in+out)

1989+ 949= 2938

2097+ 981= 3078

6169+2387=8556

7681+2956=10637

8641+3236=11877

14080+3302= 17382

13788+ 3072= 16860

13190+2680= 15870

13643+2307= 15950

The AFDI survey is exceptionally consistent across years in terms of the questions which it asks, and this provides a particularly useful platform for the construction of panel data sets. However, the references assigned to companies’ branches/subsidiaries within the AFDI change over time, which makes it harder to track firm performance. The VML team has constructed consistent references at the enterprise group level, and these can be used to, more accurately, gauge the number of firms which appear in multiple survey years. The tables below show the number of occurrences with which enterprise groups appear within the survey for both the inward and outward inquiries. For example, a firm appearing within the survey during 1996 may appear in each of the eight successive surveys. These counts do not necessarily relate to sequential observations, but can also relate to units who appear in the sample intermittently. The use of consistent enterprise group references marginally reduces the number of firms in the files per year; hence, the totals in the tables below are less than those presented above.

Number of times an Enterprise group appears in the survey: Inward Inquiry

Year 1 2 3 4 5 6 7 8 9 Total

1996 182 588 86 81 58 104 150 73 599 1,099 1997 692 110 108 64 112 176 98 655 1,262 1998 347 800 767 445 960 611 2,120 3,930 1999 1,340 1,121 529 1,055 834 2,676 7,555 2000 1,804 573 1,474 970 3,709 8,530 2001 1,092 3,443 2,134 7,205 13,874 2002 4,189 2,131 7,274 13,594 2003 2,643 10,335 12,978 2004 13,375 13,375

Number of times an Enterprise group appears in the survey: Outward InquiryYear 1 2 3 4 5 6 7 8 9 Total 1996 76 78 46 47 43 55 99 45 369 345 1997 95 51 54 48 63 111 60 415 422 1998 143 312 261 160 376 312 704 1,564 1999 466 407 217 500 421 805 2,816 2000 629 231 654 510 1,092 3,116 2001 262 979 681 1,299 3,221 2002 1,015 678 1,327 3,020 2003 912 1,686 2,598 2004 2,273 2,273

3. Links to Other Business Surveys

The IDBR is the key sampling frame for business surveys within ONS. Enterprises appearing within ONS surveys are assigned a unique IDBR reference number which can facilitate linking of information on the same enterprise between surveys. Such linking provides the opportunity to explore additional research questions that otherwise would not have been possible. The largest and most comprehensive ONS business survey is the Annual Business Inquiry. This survey includes information on turnover, costs, employment and investment. Due to the size and content of this survey, the ABI generally forms the spine against which most linking activity takes place. Responses to the AFDI survey can therefore be linked to information on these organisations collected from the ABI. Within the VML, information from the ABI is held in the Annual Respondents Database (ARD). To reduce compliance costs, the ABI is not a census of all businesses, with smaller reporting units being sampled. Within the ARD there are therefore two types of enterprise. Information collected directly from the survey returns of the ABI are held on the ‘selected files’ of the ARD. Information on those organisations included within the ABI survey universe but which are not included within the actual survey during a given year are held on the `non-selected’ files. By including information from the ‘non-selected’ ARD files, the coverage of the ARD is broadened considerably. However, the range of data items held on the non-selected files is more limited. Measures of employment and turnover derived directly from the data sources used to construct the IDBR are available. The quality of this information is however inferior to that collected directly from the survey returns of the ABI.

The number of firms which can be linked from the AFDI to the ARD is considerably improved using the consistent enterprise group references compared to the original AFDI references. Matches improve from 65-75% to 80-90% for the inward files, and from 20-30% to 70-80% for the outward files. In examining the nature of links, it is important to take into account whether these links can be made to the selected or non-selected files. It can be seen from the tables below that approximately 20% of inwards inquiry firms can be linked to the selected files, while 40% of firms in the outwards inquiry can be linked to the selected files. Links to the ARD

year AFDI:

inwards Links:

selected Links: non-

selected Total links

1998 6,168 1419 3969 48421999 7,681 1510 4777 58302000 8,641 1563 5872 69772001 14,080 2648 10561 122672002 13,788 2381 9612 111402003 13,189 2502 10396 119562004 13,643 2675 10725 12406

Links to the ARD

year AFDI:

outwards Links:

selected Links: non-

selected Total links

1998 2,387 941 1560 17691999 2,956 1126 1868 21892000 3,236 1146 2038 23682001 3,302 1375 2349 27282002 3,072 1212 2130 24442003 2,680 1078 1904 21592004 2,307 940 1641 1877

Total links do not sum to selected + non-selected due to overlap of references 4. Overview of Previous Research

Much of the academic work on FDI has focused on the question of detecting “productivity spillovers” from multinationals, i.e. whether or not domestic firms increase their productivity through learning and competition from MNEs. Related work has examined the macroeconomic link between FDI and growth in cross-country growth regressions, as well as the potential wage effects of multinationals in the domestic economy. The following provides an overview of the work completed in the VML:

• Girma & Wakelin 2001 examine the regional impact of foreign-owned establishments on the performance of domestic establishments in the electronics sector in the UK.

• Barnes & Martin 2002 examine the productivity differences between UK and foreign firms. They find evidence that the superiority of foreign owned firms is a consequence of their multinational nature; UK MNEs are only marginally less productive than their foreign counterparts.

• Haskel, Pereira & Slaughter 2002 investigate potential productivity spillovers from FDI to domestic firms. They find positive spillovers where foreign firms have a share of industry activity, but not a geographical affect.

• Martin 2003 shows that subsidiaries of US multinationals (MNEs) are the productivity leaders in the UK. Strong evidence emerges that the US advantage lies in the ability to takeover already productive plants not through the use of superior shared assets.

• Griffith, Redding & Simpson 2004 examine the relationship between foreign ownership and productivity, paying particular attention to the role of multinationals in service sectors and the importance of R&D activity conducted by foreign multinationals.

• Maioli, Görg & Girma 2005 investigate the competitive discipline effect exerted by both imports and FDI on plant-level price-cost margins in the UK. The authors consider both total FDI and its decomposition into greenfield and non-greenfield investment.

• Criscuolo, Hagsten, Hanley & Karpaty 2006 compare the affects of offshoring in Ireland, Sweden and the UK.

• Criscuolo & Leaver 2006 analyse the importance of offshoring and its relationship with firms’ performance for both the manufacturing and service sectors. The heterogeneity of service sector firms appears to be important for analysis of the service sector.

• Girma & Gorg 2006 focus on the role of the efficiency gap in determining whether or not domestic firms benefit from productivity spillovers from FDI.

• Driffield, Henry & Love 2006 classify inward FDI into various types (using R&D Intensity differentials and unit labour costs) to obtain differing effects of FDI on domestic TFP compared to when inward FDI is treated as a homogenous variable.

• Girma & Wakelin 2006 seek to identify the causal effect of foreign acquisitions on the wages of skilled and unskilled workers.

• Burke, Görg & Hanley 2006 examine the impact of Foreign Direct Investment (FDI) on the survival of business start-ups; FDI has potential for both negative displacement/competition effects as well as positive knowledge spillover and linkage effects on new ventures.

• Simpson 2007 examines how investment in low-wage economies affects firms’ home-country operations. As predicted by models of vertical multinationals, findings support the notion that investment in relatively low-wage economies is associated with plant closures in relatively low-skill, low-wage industries in the UK.

• P.Criscuolo & Salter investigate the link between foreign subsidiary “embeddedness” (i.e. their links to domestic firms) and their survival patterns in the UK.

5. Descriptive statistics on identified Multinationals

The AFDI data sets have, so far, only been used as a marker to identify firms with multinational operations. The other variables held in the data sets have not been examined in their own right. This is, in part, because these variables tend to have a financial and accounting basis and are not easy reconciled with the variables held in other VML data sets such as the ARD. This section presents some basic descriptive statistics on firms identified as MNEs using the AFDI markers.

There are 3 separate types of MNEs that can be identified using the AFDI data set: firstly, UK owned; secondly, Foreign Subsidiaries/Branches of non-UK MNEs; thirdly, the least common, UK MNEs with foreign owners. This split allows researchers to explore

whether the type of MNE alters results (e.g. Barnes & Martin, 2002). The following table shows the number of each MNE type by year.

Year UK MNE Foreign

Subsidiaries Foreign owned UK

based MNE 1996 871 1,911 781997 887 2,003 941998 2,274 6,056 1131999 2,841 7,566 1152000 2,985 8,390 2512001 3,044 13,822 2582002 2,806 13,522 2662003 2,377 12,887 3032004 1,999 13,335 308

Linking the AFDI marker panel to the ARD we can examine how the characteristics of UK and Foreign owned MNEs vary. The following tables provide a comparative selection of firm characteristics by MNE type, sector and year.

The descriptive statistics on firm type appear to indicate that foreign-owned MNEs are larger (measured by mean employment/number of local units) than UK MNEs, while UK MNEs are of a similar size, or slightly larger, than foreign subsidiaries operating in the UK. The following charts, for a selection of industries, illustrate this.

Motor sector MNE characteristics: Employment

0

2000

4000

6000

8000

2000 2001 2002 2003 2004

Year

Mea

n E

mp

loym

ent

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Motor sector MNE characteristics: Local units

0123456

2000 2001 2002 2003 2004

Year

Mea

n c

ou

nt

per

fir

m

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Chemical sector MNE characteristics: Employment

0200400600800

100012001400

1998 1999 2000 2001 2002 2003 2004

Year

Mea

n E

mp

loym

ent

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Chemical sector MNE characteristics: Local units

02468

101214

1998 1999 2000 2001 2002 2003 2004

Year

Mea

n c

ou

nt

per

fir

m

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Electrical & Optical sector MNE characteristics: Employment

0200400600800

100012001400

1998 1999 2000 2001 2002 2003 2004

Year

Mea

n E

mp

loym

ent

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Electrical & Optical sector MNE characteristics: Local units

0123456

1998 1999 2000 2001 2002 2003 2004

Year

Mea

n c

ou

nt

per

fir

m

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Although it is very sector specific, it appears that on average foreign-owned MNEs invest more in software, and are often more productive (when measured by GVA per employee). The following charts illustrate these trends for the same sectors as above:

Chemicals sector MNE characteristics: Mean GVA per employee

0

50

100

150

200

250

1998 1999 2000 2001 2002 2003 2004

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Chemicals sector MNE characteristics: Mean Software Investment

0200400600800

10001200

1998 1999 2000 2001 2002

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Electrical & Optical sector MNE characteristics: Mean GVA per employee

-100

-50

0

50

100

1998 1999 2000 2001 2002 2003 2004

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Motor sector MNE characteristics: Mean GVA per employee

0

20

40

60

80

100

1998 1999 2000 2001 2002 2003 2004

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Electrical & Optical sector MNE characteristics: Mean Software Investment

050

100150200250300350

1998 1999 2000 2001 2002

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Motor sector MNE characteristics: Mean Software Investment

0500

10001500200025003000

1998 1999 2000 2001 2002

UK MNE

Foreign Subs/Branch

Foreign ow ned MNE (UK based)

Bibliography Barnes & Martin (2002), “Business Data Linking: An introduction”. Economic Trends 581: p34-41, available at: http://www.statistics.gov.uk/about/platforms/et/articles2002.asp Simpson (2007), “Investment abroad and adjustment at home: evidence from UK multinational firms”. IFS draft paper. Griffith, Redding & Simpson (2004), “Foreign Ownership and Productivity: New evidence from the Service Sector and the R&D Lab” Published in 'Oxford Review of Economic Policy,' Vol. 20, No. 3, pp. 440-456 Criscuolo, Hagsten, Hanley & Karpaty (2006), “Offshoring and productivity: the case of Ireland, Sweden and the United Kingdom.” Criscuolo & Leaver (2006), “Offshore Outsourcing and Productivity”. Haskel, Pereira & Slaughter (2002), “DOES INWARD FOREIGN DIRECT INVESTMENT BOOST THE PRODUCTIVITY OF DOMESTIC FIRMS?” NBER working paper 8724, available at: http://www.nber.org/papers/w8724 P.Criscuolo & Salter, “EASY-COME EASY-GO: EMBEDDEDNESS AND SURVIVAL OF FOREIGN SUBSIDIARIES IN THE UK” Martin (2003), “US productivity leadership: Evidence from the UK”. Royal Economic Society Annual Conference 2003 Maioli, Görg & Girma (2005), “Trade, FDI and Plant-Level Price-Cost Margins in the UK”. Working paper. Girma & Gorg (2006), “The role of the efficiency gap for spillovers from FDI: Evidence from quantile regressions”. Open Economics Review. Girma & Wakelin (2001), “Regional Underdevelopment: Is FDI the Solution? A Semi-parametric analysis” University of Nottingham Research Paper Series 2001/11. Driffield, Henry & Love (2006), “Linking Motivation and Effect: The Nature of Inward FDI and its Impact on Productivity Growth in the UK.” Draft paper.

Girma & Gorg (2006), “Evaluating the foreign ownership wage premium using a difference-in-differences matching approach”. Journal of International Economics. Burke, Görg & Hanley (2006), “The Impact of Foreign Direct Investment on New Firm Survival in the UK: Evidence for Static vs. Dynamic Industries”. Draft paper.

Matching ARD and AFDI

Chiara Criscuolo and Ralf Martin

28 January 2011

Abstract

This paper describes a dataset which was created by merging the Annual Register Database (ARD) and the Annual Inquiry into Foreign Direct Investment (AFDI). The purpose of the dataset is to identify multinational enterprises (MNEs) within the population of ARD enterprises. The novelty of the dataset is that it also allows to identify enterprises which are owned by UK multinationals. Previously it has only been possible to identify foreign owned enterprises.

CeRiBADataGuide

1/11 Centre for Research into Business Activity –www.CeRiBA.org.uk

CeRiBA

1 Introduction

We have created an identifier of UK multinationals in the Annual Respondents Database

(ARD) by merging the Annual Inquiry into Foreign Direct investment (AFDI) for the years

1996 to 2000. This document first describes the AFDI and then explains in detail how AFDI

and ARD have been merged1. The ARD comprises the production sector only before 1998.

For consistency we confine ourselves for the years including and after 1998 to the

production as well. For more Information on the ARD please refer to Barnes and Martin

(2002). The multinational identifier is made available to the public in a Stata dataset called

ARD_AFDI01.dta whose variables are described at the end of this document.

2 The Annual Inquiry into Foregin Direct Investment

The AFDI is an annual survey among businesses which requests a detailed breakdown of

the financial flows between UK firms and their overseas parents or subsidiaries. The AFDI is

thus a survey run at the enterprise group and not at the enterprise2 level as the ARD. The

inquiry has an `outward’ part that measures foreign direct investment (FDI) by UK

enterpriese groups abroad and an `inward' part that measures FDI in the UK by foreign

corporations.

To conduct the AFDI, the ONS maintains a register which holds information on the country

of ownership of each firm and on which UK firm has foreign subsidiaries or branches3. This

register is designed to capture the universe of firms that are involved in foreign direct

1 See Criscuolo and Martin (2002) for details about why this is an interesting thing to do. 2 In ARD terminology the term “enterprise” refers to an autonomous business unit. Larger enterprise

groups can therefore consist of several enterprises. In 80 percent of all cases an enterprise consists of

one plant only. In our other work we therefore usually refer to this level as the plant level for

simplicity. Enterprise groups we usually call “firms”. 3 In the following we refer to subsidiaries and branches jointly as affiliates. The ONS distinguishes

between subsidiaries and branches as follows: a “subsidiary” is mainly a company where the parent

company holds more than 50% of the equity share capital; a ”branch” is a permanent plant as defined

for UK corporation tax and double taxation relief purposes; companies where the investing company

holds between 10% and 50% of the equity share capital, i.e. does not have a controlling interest but

participates in the management, are defined `associates'. ONS (2002) p.120.

1/11

CeRiBA

investment abroad and in the UK4. The AFDI register is drawn from (and continuously

updated) using a variety of sources including administrative records, (from HM Customs

and Excise and from Inland Revenue), Dun and Bradstreet's `Worldbase' system and ONS

inquiries on acquisitions and mergers involving UK companies.

3 The basic challenge

The reporting unit level of the ARD is the enterprise level5. The AFDI on the other hand

inquires about issues which relate to the enterprise group. The reporting unit of the AFDI is

thus an enterprise group. An enterprise group is included in the AFDI if it has significant

foreign ownership or engages into foreign direct investment. In the current context we use

this to determine whether an enterprise in the ARD is part of a multinational enterprise

group. To do this we have to find enterprise group numbers which are consistent between

the two surveys. Section 3.1 explains the problems encountered thereby in detail. An

additional complication arises for the years 96 and 97 because of a coding change of the

enterprise group reference number. Section 3.2 elaborates on this.

4 The annual inquiry regards direct investment as an investment made abroad in order to have an

effective voice in the management of a foreign firm. For practical purposes this is defined, since 1997,

as holding a share of at least 10% (20% before 1997) in the foreign company, whereas holdings below

this threshold are considered portfolio investment. 5 Large enterprises have sometimes more than one reporting unit in the ARD. For a more elaborate

discussion of the complicated reporting structure of the ARD see Barnes and Martin (2002).

2/11

CeRiBA

3.1 Finding consistent enterprise groups references

Table 1 outlines how enterprise group numbers and reporting unit numbers typically look

like in the ARD and AFDI for a set of exemplary enterprise groups. In the example we have

3 enterprises which belong to two enterprise groups. One enterprise group – Smith Vehicles

– owns two enterprises. The other enterprise group – Jones Bakery - consists of only one

enterprise. Column 3 shows the ARD reporting unit number which refers to enterprises.

Column 4 reports enterprise group numbers. The two Smith enterprises have a common 10

digit enterprise number not related to their 11 digit reporting unit numbers. Singleton

enterprise groups6, such as Jones, typically have enterprise group numbers which are equal

to the reporting unit number without the leading 47.

Column 5 shows how these enterprise groups are represented in the AFDI. In the case of

Jones Bakery reporting on the enterprise group and enterprise level is equivalent.

Consequently it has the same reporting unit reference in the AFDI as well. Smith vehicles on

the other hand has an reporting unit number which is called in IDBR languge a “survey

specific reporting unit”. However in practice this is nothing else but the enterprise group

reference number with the two digits “60” added in front. About 70% of the enterprise

groups contained in the AFDI have such reporting unit references and thus consist of several

enterprises in the sense of the ARD.

Table 1: Reporting units in ARD versus AFDI (An example for clarification)

Representation

in AFDI Representation in ARD

ARD reporting

unit number

(ARD RUREF)

Enterprise

group number Enterprise

group name

Enterprise

Name (EGRPREF)

AFDI reporting

unit reference

(AFDI RUREF)

Smith Vehicles Smith trucks 49912345678 212345678 60212345678

Smith Vehicles Smith bicycles 49912345679 212345678

Jones Bakery Jones Bakery 49912345680 9912345690 49912345680

6 i.e. enterprise groups which consist of only one enterprise. 7 Often the enterprise group reference number is missing altogether, as well.

3/11

CeRiBA

3.2 The change in enterprise group numbers after 1997 in the ARD

3.2.1 Two lookup tables

After 1997 the coding of enterprise group numbers in the ARD changed from an earlier

register system to the coding used by Dun&Bradstreet. This affected also the coding of

reporting units in the AFDI because they are made up from the enterprise group references

for some the units, as outlined in the previous section. Contrary to the ARD, in the AFDI the

coding change has been applied for waves before 1998; i.e. we find in the AFDI for 1996, say,

not the enterprise groups numbers we find in ARD for 1996 but the enterprise group

numbers the same units had after 1997. Units which died before that date along with a small

number of units which survived beyond 1997 have still have the old number however.

IDBR provides a lookup table for the post 1997 coding change. Unfortunately this table

seems incomplete: there is a number of units which make it beyond 1997 that have no entry

in the lookup table. There is a way to expand this lookup table however. In fall 2001 we

made a query to the IDBR which provided for each reporting unit stored in the IDBR at that

moment in time the corresponding enterprise group reference. Since all the enterprise

groups numbers appearing in the AFDI become by this very fact reporting units they will

also appear in our query8. Our query therefore is an alternative lookup table, at least for

enterprise groups which survived until fall 2001.

3.2.2 Singletons 1

Consider next Table 2. It shows several exemplary problems we encountered for a large

group of singletons after updating the EGRPREFs with the two lookup tables9 sequentially.

Columns 1 to 3 of the first panel show an enterprise belonging to an enterprise group

consisting of only this one enterprise in 1997 and 98. Such singleton’s would normally either

have a missing EGRPREF or it would be equal to the RUREF. In this particular case the

EGRPREF is equal to a number different from the RUREF after 1998 and missing before.

Taking this case as it is would imply accounting for an ownership change in 1998. However

as the enterprise group apparently consists only of this enterprise for an ownership change

it would have to change from itself. We therefore assume that it belongs already before 1997

to the very same enterprise group A.

8 Provided the enterprise group is still alive in fall 2001, that is. 9 i.e. the one published by the ONS and the table based on our 2001 IDBR snapshot.

4/11

CeRiBA

Table 2: Further enterprise group ref problem cases (1) (2) (3) (4) (5)

year egrp_ref ru_ref no of ru in

firma

egrp_ref ru_ref

1996 . 1 1 A 3

1997 . 1 1 A 3

1998 A 1 1 D 3

1999 A 1 1 D 3

2000 A 1 1 D 3

1996 B 2 1

1997 B 2 1

1998 . 2 1

1999 . 2 1

2000 . 2 1

1996 E 4 4

1997 E 4 5

1998 F 4 6

1999 F 4 5

2000 E 4 7 (a): (pre 1998 the figure refers to manufacturing only)

A situation in which this might be problematic is introduced in Columns 4 and 5. The

enterprise 3 which is introduced there actually belonged to firm A in 1997. Setting EGRPREF

for 1 to A before 1998 would thus show that A was not really a singleton in 1997. Thus our

argument for changing 1’s EGRPREF breaks. Possibly what really happened was that A sold

3 and bought 1 at the same time in 1997 and thus demonstrates that it has a firm identity

independent from establishment 1 or 2. Therefore in these cases we do not change 1’s

EGRPREF.

3.2.3 Singletons 2

A similar but slightliy different situation is shown in the next panel. Here the

establishment/firm is a genuine singleton after 1998 but not before. If the apparent firm

before 1998 consists only of this establishment we change the pre 1998 EGRPREF again to

the post 1998 one, which would be in this case the RUREF. A possible error we cannot

control for here is that the pre 1998 firm might have establishments in non manufacturing

sectors.

5/11

CeRiBA

3.2.4 Random changes

The last panel of Table 2 shows another frequent problem: An establishment has a changes

to another EGRPREF for only 1 or 2 years and then changes back. In this case we would set

the EGRPREF in 1998 and 99 to E.

3.2.5 The effect of all changes

We apply all these changes in the following order

1. Sequential updating using the lookup tables

2. Singlton type 1

3. Singleton type 2

4. Random changes

The effect of this on the number of EGRPREF changes from year to year is reported in Table

3

Table 3: Updating the enterprise group reference in the ARD

. Column 1 reports the number of changes if EGRPREF is left unchanged. The other

columns show the same figure after various updating steps.

(Number of ARD reporting units whose egrp_ref changes before and after updating

(1) (2) (3) (4) (5)year raw lookups singleton1 singleton2 random changes

1996 11991 11903 5063 4776 42261997 3445 3310 2812 2500 21651998 23309 13134 10019 7965 70861999 5825 4642 4642 4642 37622000 6416 5040 5040 5040 4385

4 The success of the merging exercise

4.1 Preparing the AFDI

The AFDI dataset is divided in two parts: An “in” dataset containing information about

foreign direct investment into the UK and and “out” dataset containing information about

foreign direct investment activities of UK companies. The “out” dataset has entries for each

foreign subsidary or branch of a UK company. For merging purposes we have to aggregate

this back to the enterprise group level. The next step assembles both “in” and “out” data in

one dataset. Table 4 shows how many enterprise groups by ownership type are covered by

the AFDI for each year. Although in principle the AFDI survey register should have an

entry for every multinational business in the country there is no mechanism which ensures

6/11

CeRiBA

that each and every company which does or receives FDI will automatically be in the

register. Consequently the register entries will vary depending on the ONS success and

effort in identifying such companies from a variety of information sources. This problem is

behind the apparent jump in FDI in 1998 in row 3 of Table 4: in this year the ONS used for

the first time data from Dun&Bradstreet “Who owns whom” database to identify FDI firms.

As we will see below, annoying as this variation might be, it does not seem to affect our

matched sample of selected reporting units too much.

Before we can match the AFDI to the ARD we have to extract consistent enterprise group

references from the AFDI. In the case of singletons this is done simply by deleting the

leading “4” and in the case of multiplant firms by deleting the leading “60”(see Table 1).

An additional complication is the following: As mentioned earlier, contrary to the ARD, in

the AFDI the identifiers are kept up to date also for older waves; i.e. the enterprise group

numbers found in the 1996 AFDI wave are not the pre 1998 ones but actually the most up

date ones at the time the dataset was assembled which is 2002. In practice this is, however,

not true for all observations. For some observations we found that the numbers were not

updated. To make sure that we use consistent coding in ARD and AFDI we apply the

lookup table created to update EGRPREFs (see Section 3.1) to the AFDI as well.

4.2 Preparing ARD

This section takes the existence of an intertemporally consistent ARD panel for 1996-2000 as

given. The details on the creation of such a panel are described in Martin (2002). Once a

reporting unit panel is created the problem of preparing the ARD reduces to updating the

enterprise group references for 1996 and 1997. How to do this was described in Section 3.2.

4.3 Merging ARD andAFDI

The two datasets are merged on the updated EGRPREF.

Table 5 reports how many (enterprise-)units by ownership type can be found in the ARD

after merging for each year. The table reports figures for the sample as a whole, for the

selected units only and for selected unit. For the population as a whole notice the jump in

UK multinationals in 1998 (Row 4). This might well be the consequence of the increased

AFDI sample size (see Section 4.1). Looking at the numbers for selected units in Table 5 we

do not find this jump, however. So any analysis based on these should not be affected by the

changes.

7/11

CeRiBA

Table 4 – Enterprise groups in the AFDI 1996 1997 1998 1999 2000

1129 1212 1203 2470 2371foreign_headuk 67 83 100 104 115ukmult 1 2213Source: Author's calculation b

foreign=foreign owned enterprise groups

m is undertaking foreign direct investment

foreign_headuk=a foreign owned enterprise oreign direct investment from the

foreign

692 739 1803 217ased on AFDI

uk ult=a UK owned enterprise group that

group that is undertaking f UK

The success of the merging operation

1998 1999 2000UK non MNE 165175 164090 161234foreign 2429 2202 3120 3059foreign_headuk 454 03 440UK MNE 2743UK non MNE 0.967foreign 0.014foreign_headuk 0.003UK MNE 0.016UK non MNE 9793foreign 1352foreign_headuk 284UK MNE 1507 1343UK non MNE 0.058foreign 0.008 0.008 0.009 0.009foreign_headuk 0.002UK MNE 0.009

Source: Author's Calculation based on ARD and AFDI

shares

Selected

res

numbers

Table 5:(Absolute numbers and shares 1996 – 2000)

1996 1997162937 166312

2468452 337 42624 3255 3257 29190.968 0.966 0.960 0.9620.014 0.013 0.018 0.0180.003 0.002 0.002 0.0030.015 0.019 0.019 0.0179527 9805 9596 92571260 1299 1455 1450261 191 231 2351332 1527 15290.055 0.057 0.056 0.0550.0070.002 0.001 0.001 0.0010.008 0.009 0.009 0.008

numbers

All

sha

5 The provided datasets

.dta

e description

5.1 multinats_panel01

This dataset contains the AFDI information from both “IN” and “OUT” part combined and

assembled in to a panel. The variables therein are as follows:

variable nam

new_egrp_ref The newly created egrp_ref max_idbr_ref_out min_idbr_ref_out min_idbr_ref_in min_idbr_ref_in

of

The maximum and minimum idbr_ref fromthe `in’ and the `out’ part of the AFDI respectively. This allows to keep track how our procdure derives the new_egrp

8/11

9/11

CeRiBA

variable name description

from the orginal AFDI key which is called idbr_ref. Taking minimum and maximum allows to track down if units which had different idbr_refs originally were allocated the same new_egrp_ref eventually.

ukmult equal to one if an observation appears in the `out’ part

f_own_in country code of a foreign owned unit in the `in’ part. Note: the country codes are different from the ARD ones.

out_type* foreign

Set of dummy variables indicating if a firm has an affiliate in a particularcountry. The numbers are ARD country codes.

I01.dta

tains a merge between ARD and A

rting unit. The variables included i

section are

me

5.2 ARD_AFD

This file con FDI. Thus the level of observation file is an

ARD repo n addition to the ones described in the last

variable na description

dlink_ref2 the ARD reporting unit identifier (in later

ARD rawdata files this is also called ruref)

new_egrp_ref The newly created egrp_ref ma _idbr_ref_out The maximum axmin_idbr_ref_out

idbr_ref_in

nd minimum idbr_ref from the `in’ and the `out’ part of the AFDI

This allows to keep track of

re allocated ly.

min_min_idbr_ref_in

respectively.how our procdure derives the new_egrp from the orginal AFDI key which is called idbr_ref. Taking minimum and maximum allows to track down if units which had different idbr_refs originally wethe same new_egrp_ref eventual

ukmult equal to one if an observation appears in the `out’ part

f_own_in country code of a foreign owned unit in the `in’ part. Note: the country codes are different from the ARD ones.

sel_id ARD selection ID sic92 Industry classifier egrp_ref orginial ARD egrp_ref f_own_abi foreign ownership code derived from ARD f_own_x combined foreign ownership code. Uses

ARD coding. Precedence is given to ARD

10/11

CeRiBA

variable name description

code. If ARD classifies firm as not foreign owned however AFDI classification is used.

out_type* Set of dummy variables indicating if a firm has an affiliate in a particular foreign country. The numbers are ARD country codes.

6 5BOther Files in the multinats_public directory

file description

multinats_public_dome.xls Collects the various data preparation and descriptive statistics steps.

multinats_public_results.xls Collects various result table displayed in this document

country_codes.xls Table with ARD country codes.

other *.xls files Result tables

*.do files Stata do files to create the data and statistics called by multinats_public_dome.xls

6BReferences Criscuolo, Chiara and Ralf Martin (2003), “Multinational, foreign ownership and US productivity leadership: Evidence from the UK”, Ceriba Discussion Paper. Office for National Statistics (2002), “Foreign Direct Investment 2000”, Business Monitor MA4. HMSO, London

FDI company types:

Type Code

Description Explanation

A fully completed return

sampled contributor that has returned form.

B non-responder to be imputed

sampled contributor that has not returned form, data imputed by FDI or WorldBase methodologies.

C estimated (non-approach)

unsampled contributor, data imputed by FDI or WorldBase methodologies.

D constructed (from accounts)

sampled contributor that has not returned form, no inquiry data for period available, form constructed from company accounts for that period.

G dead-nil transactions

contributor out of scope of inquiry & no transactions in the period.

J dead-transactions included on another form

contributor in scope of inquiry, but reported in consolidated group figures by another contributor. May hold data if group contributor cannot supply data until next period.

K dead-direct investment link ceased to exist

contributor out of scope of inquiry, contributor no longer has foreign links (eg disposal or company reorganisation), may have partial period data if link ceased within the current period.

L sampled-not yet responded

sampled contributor who has not yet responded to inquiry, will eventually become another type or type B imputation at end of inquiry.

N not approached for this period

out of scope of the inquiries, used to remove contributors added in error to the samples.

O dead-ceased trading

contributor has ceased trading , may have data if cessation occurred part way thru the period.

Q dead-nill off contributor has ceased trading for next period but supplied no data for this period, enter nil data returns if no other data available.

R M&A-construction-data-select next period

new contributor identified from the Cross Border Acquisition & Merger (CBAM) inquiries for this period , construct partial form (profits, inter-company balances, acquisition or disposals data, share capital & reserves) based upon the data from the relevant CBAM deal. automatic proving selection for next annual inquiry period.

S M&A-inward-no data-select next period

new contributor identified for next FDI period, no data entered for this period but entered into population with no data to force onto selection for next FDI period.

T M&A-outward-no data-select next period

as for S

U constructed from quarterly data

sampled contributor who has not returned form, form is constructed using previously returned quarterly data covering the same period from the FDI quarterly inquires. geographical breakdown based on previous years return (usually only large companies with 4 quarters of good returns (type A's).

V receipted temporary type required for IDBR receipt datafile, replaced by one of others asap Z external sources unsampled data from external sources, eg aggregate level Bank of England data for

banking sector, quarterly HMT database data for inward branch insurance companies.

The purpose of this glossary is to provide a comprehensive list of terms associated with the Foreign 

Direct Investment (FDI) survey which will aid users understanding and interpretation of the FDI 

statistics. 

Accuracy ‐ The degree of closeness between an estimate and the actual value. 

Acquisitions‐ a business transaction between unrelated parties based on terms established by the market where each enterprise acts in its own interest.  The acquiring enterprise purchases the assets and liabilities of the target enterprise.  In some cases, the target enterprise becomes a subsidiary or part of a subsidiary of the acquiring enterprise.    Additive imputation approach (Quarterly) ‐ Average difference taken between responders from the 

current and previous period, which is then added to the previous period value to impute for a non‐

responder.    

Affiliated enterprises – Affiliated enterprises are enterprises in a direct investment relationship.  Thus, a given direct investor, its direct investors, its subsidiaries, its associates, and its branches, including all fellow enterprises are affiliated enterprises.  It is possible for a given enterprise to be a member of two or more groups of affiliated enterprises.    Balance of Payments‐ a statistical system through which economic transactions occurring during 

specific time periods between an economy and the rest of the world can be summarised in a 

systematic way.  The IMF Balance of Payments and International Investment Manual provides 

conceptual guidelines for compiling balance of payments statistics according to international 

standards.   

Branch, Direct Investment Enterprise‐ A branch is any unincorporated direct investment enterprise in the host country fully owned by its direct investor. Thus, this term encompasses branches as commonly defined – i.e. formally organised business operations and activities conducted by an investor in its own name – as well as other types of unincorporated operations and activities.   All or most of the following features should be present for a branch to be recognised:   (i) Undertaking or intending to undertake production on a significant scale based in the territory for one year or more in a territory other than that of its head office:   (a) If the production process involves physical presence, then the operations should be physically located in that territory. Some indicators of an intention to locate in the territory include purchasing or renting business premises, acquiring capital equipment, and recruiting local staff;  (b) if the production does not involve physical presence, such as in some cases of banking, insurance, or other financial services, the operations should be recognised as being in the territory by virtue of the registration or legal domicile of those operations in that territory;   (ii) the recognition of the operations as being subject to the income tax system, if any, of the economy in which it is located even if it may have a tax‐exempt status.   

Book Value – Book value is a term that broadly encompasses many different accounting methods.  It 

represents the values that appear on the books of an enterprise.  It could represent the values on 

the books of direct investors or on the books of direct investment enterprises.   

Coefficient of variation‐ standard error expressed as percentage of total value.   

Creditor/Debtor principle‐ A debtor is a person or an entity which has a financial obligation to another person or entity. Conversely, a creditor is a person or entity which has a financial claim on another person or entity. Therefore, a debtor has a financial liability to a creditor and a creditor has a financial claim (an asset) on a debtor. For FDI statistical purposes, under the debtor/creditor principle, the FDI assets (both transactions and positions) of the compiling economy are allocated to the economies of residence of the non‐resident debtors; its FDI liabilities are allocated to the economies of residence of the non‐resident creditors allocated on the basis of the debtor/creditor principle. This principle, recommended by the Benchmark Definition as the basis for geographical allocation, differs from the transactor principle.   Data cleaning‐ a process by which any errors in FDI data are fully investigated and corrected to 

ensure that data outputs are as accurate as possible before final results are produced.   

Disposals‐  Disposals of assets (inventories, fixed assets or land or other non‐produced assets) by 

institutional units occur when one of those units sells or transfers any of the assets to another 

institutional unit; when the ownership of an existing fixed asset is transferred from one resident 

producer to another, the value of the asset sold, bartered or transferred is recorded as negative 

gross fixed capital formation by the former and as positive gross fixed capital formation by the latter. 

Dividends‐ Dividends are earnings distributed to shareholders from common and participating 

preferred stock, whether voting or non‐voting, according to the contractual relationship between 

the enterprise and the various types of shareholders, before deduction for withholding taxes.  

Dividends exclude liquidating dividends and bonus shares (which are dividends in the form of 

additional shares of stock). These are recorded on the date they are paid.   

Difference value‐ A difference value is derived by subtracting the previous value from the current 

value.   

Earnings‐income from investments e.g. profits, interest and tax 

Enterprise‐ An enterprise is an institutional unit engaged in production. An enterprise may be a corporation, a non‐profit institution, or an unincorporated enterprise. Corporate enterprises and non‐profit institutions are complete institutional units. An unincorporated enterprise, however, refers to an institutional unit – a household or government unit – only in its capacity as a producer of goods and services.   Exchange rate changes‐ Exchange rate changes reflect the impact that changes in exchange rates have on instruments that are denominated in a currency other than that in which the accounts are compiled. Exchange rate changes may be referred to by enterprises as realised or unrealised exchange rate or foreign   Equity capital‐ Equity capital comprises: (i) equity in branches; (ii) all shares in subsidiaries and associates (except non‐participating, preferred shares that are treated as debt securities and included under direct investment, debt instruments); and (iii) other contributions of an equity nature. Ownership of equity is usually evidenced by shares, stocks, participations, depositary receipts or similar documents. Shares and stocks have the same meaning while depositary receipts are securities that represent ownership of securities by a depositary. This category includes proprietors’ net equity in quasi‐corporations, as well as shares and equity in corporations. It also includes preferred stocks or shares that provide for participation in the residual value on dissolution 

of an incorporated enterprise. Reinvestment of earnings comprises the claim of direct investors (in proportion to equity held) on the retained earnings of direct investment enterprises. Reinvestment of earnings represents financial account transactions that contribute to the equity position of a direct investor in a direct investment enterprise.  Foreign Direct Investment‐ FDI is a category of investment that reflects the objective of establishing 

a lasting interest by a resident enterprise in one economy (direct investor) in an enterprise (direct 

investment enterprise) that is resident in an economy other than that of the direct investor.  The 

lasting interest implies the existence of a long term relationship between the direct investor and the 

direct investment enterprise and a significant degree of influence on the management of the 

enterprise.  The direct or indirect ownership of 10% or more of the voting power of an enterprise 

resident in one economy by an investor in another economy is evidence of such a relationship.  

Some compliers may argue that in some cases an ownership of as little as 10% of the voting power 

may not lead to the exercise of any significant influence while on the other hand, an investor may 

own less than 10% but have an effective voice in the management.  Nevertheless, the recommended 

methodology does not allow any qualification of the 10% threshold and recommends its strict 

application to ensure statistical consistency across countries.   

Flows‐ changes within investments e.g. reinvested earnings, acquisitions, disposals and loan movements. 

Fellow Enterprises‐ An enterprise in one economy may be related through the Framework of Direct 

Investment Relationships (FDIR) to another enterprise in the same economy, or in a different 

economy, without either being a direct investor in the other, but through both being directly or 

indirectly influenced by the same enterprise in the ownership hierarchy.  This common parent must 

be a direct investor in at least one of the enterprises in question.  Such enterprises can be 

considered to be related through horizontal linkage within the FDIR‐ not involving FDI voting power 

of 10% or more‐ and are called fellow enterprises.  It should be noted, however, that for FDI 

statistics, only cross‐border transactions and positions between FDI related enterprises should be 

recorded.   

Globalisation‐ the world wide movement towards economic, financial, trade and communications 

integration.   

Gross Domestic Product (GDP)‐ represents the total monetary value of all finished goods and 

services produced within a country’s borders during a specific period of time.   

Holding companies‐ a holding company is established to hold participation interests in other 

enterprises on behalf of its owner.  Some holding companies may have substantial physical presence 

as evidenced by, for example, office buildings, equipment and employees.  Others may have little or 

no physical presence and may exist only as shell companies.   

Inward FDI – represents direct investments made by non‐resident investors in the reporting 

country’s economy.   

Imputed Value‐ an imputed value is derived by adding the previous value to the link value. 

International Investment Position (IIP) ‐ is a statistical statement, compiled at a specified date such 

as year end, of (i) the value and composition of the stock of an economy’s financial assets or the 

economy’s claims on the rest of the world, and (ii) the value and composition of the stock of an 

economy’s liabilities to the rest of the world.   

 

Levels (positions) – enterprises’ total investment worth at a point in time e.g. closing balances and 

closing equity capital.  

Market value‐ market value is a conceptually ideal basis for valuing direct investment transactions 

and positions.  Market valuation places all assets at current prices rather than when last purchased 

or re‐valued and promotes consistency in the value of assets of different vintages.  It also promotes 

consistency when comparing stocks, transactions and other flows of different enterprises, industries 

and countries.   

Mean imputation‐ mean imputation is the replacement of a missing observation with the mean of 

the non‐missing observations for that variable. 

Multiplicative imputation approach (Annual) ‐ Ratio of means imputation is applied using sector and 

size band as imputation classes.  

Net International Investment Position‐ the stock of external assets minus the stock of external 

liabilities. 

Non‐voting stocks‐ equity/shares that do not give voting rights to the holder are called non‐voting 

stocks.  The category includes participating preference shares.   

Outlier – an observation point that is distant from the other values in the data set.   

Outward FDI‐ represents direct investments made by resident investors within foreign countries’ 

economies 

Reliability‐ refers to the closeness of the initial estimated value to the revised estimated value.  

Sampling error‐ The error caused by observing a sample instead of the whole population.   

Simple expansion estimator‐ The annual estimation is based on a mean estimation whereby the 

average returned value within each strata is used as an estimate for every non‐sampled business in 

the population.   

Special Purpose Entities – Multinational enterprises (MNEs) often diversify their investments 

geographically through various organisational structures.  These may include certain types of Special 

Purpose Entities.  Examples are financing subsidiaries, conduits, holding companies, shell companies, 

shelf companies and brass‐plate companies.  Although there is no universal definition of SPEs, they 

do share a number of features.  They are all legal entities that have little or no employment  

Standard Deviation ‐ a measure of the dispersion of a set of data from its mean.  The more spread 

apart the data, the higher the deviation.   

Strata‐ Stratification consists of dividing the population into subsets (called strata) within each of 

which an independent sample is selected. 

Subsidiaries‐ a company that is wholly or partially controlled by another through the ownership of 

50% of more of the voting stock.   

Transactor Principle ‐ According to the transactor principle, transactions resulting from changes in claims and liabilities are allocated to the country of residence of the non‐resident party to the transaction (the transactor) even if this is not the country of residence of the direct investment enterprise or direct investor. This principle differs from the debtor/creditor principle that is recommended by the Benchmark Definition as the basis for geographical allocation.   Trimming – a method used to remove any outliers (values that fell outside a certain standard 

deviation distance from the mean) from the calculation of the mean.   

Trimmed Mean‐ the value of the mean calculated after the implementation of the trimming method.   

Voting stocks ‐ equity/shares that give voting rights to the holder.  They can either be ‘listed voting 

stocks’ which are listed on an official stock exchange or ‘unlisted voting stock’ which are shares not 

listed on an official stock exchange.   

Winsorisation ‐ an imputation rule limiting the influence of the largest and smallest observations in the 

available data. 

 

 

  

 

 

MA4 Foreign Direct Investment 2005, Crown copyright 2007 134

Appendix C Definitions of industry sectors (SIC2003 two digit codes)

Resource industries Agriculture, forestry & fishing

01 Agriculture/hunting and trapping 02 Forestry/logging 05 Fishing/fish hatcheries

Mining & quarrying (including oil & gas

production) 10 Coal/peat mining 11 Oil/gas extraction and related services 12 Mining of uranium/thorium ores 13 Mining of metal ores 14 Other mining and quarrying

Manufacturing industries Food products

15 Food/tea/coffee/beverage production 16 Tobacco products

Textile & wood, printing & publishing

17 Textiles/weaving/textile products 18 Manufacture of clothing/hats/furs 20 Wood/timber/cork/straw products 21 Pulp/paper/cardboard/other paper products 22 Publishing/printing/other recorded media

Chemicals, plastic & fuel products

23 Manufacture of coke/refined petroleum products/ processing of nuclear fuel

24 Chemical/agrochemical/pharmaceutical products 25 Rubber/plastic products

Metal & mechanical products

27 Manufacture of basic metals/wire/tubes/strip products 28 Manufacture of fabricated metal products 29 Manufacture of machinery and equipment not

elsewhere classified Office, IT & communications equipment

30 Office/computer/other IT equipment 32 Communications/radio/TV equipment/components

Transport equipment

34 Manufacture of vehicles/coachwork/accessories/ spare parts/caravans/trailers

35 Manufacture of other transport equipment/ships/ trains/aircraft/bicycles/motorcycles

Other manufacturing

19 Leather tanning/leather products/footwear 26 Glass/cement/ceramic/stone/other mineral products 31 Electrical machinery (motors, generators,

transformers)

33 Medical instruments/industrial control products/ optical products/timepieces/measuring instruments

36 Furniture/jewellery/sports goods/musical instruments/ toys and games/other non classified manufacturing

37 Recycling of scrap/waste materials Service industries Electricity, gas & water

40 Electricity/gas/steam production/distribution 41 Water collection/purification/ distribution

Construction

45 Construction/building installation and completion/ civil engineering/rental of construction equipment with operator/ demolition services

Retail/wholesale trade & repairs

50 Sale/repair of vehicles/parts/accessories and fuel 51 Wholesale/commission trading (excl motor vehicles) 52 Retail trading & non classified repairs

Hotels & restaurants

55 Hotels/holiday sites/restaurants/bars/catering Transport & communications

60 Land transport/rail/road/pipelines 61 Water transport/sea/coastal/inland 62 Air transport/passenger/freight/space 63 Transport support/cargo storage/travel services 64 Postal/courier/telecommunications services

Financial services

65 Financial intermediation services/banks/building societies/investment trusts/venture capital companies

66 Insurance services/life/non-life/pensions 67 Other financial intermediation services

Real estate & business services

70 Real estate/buying/selling/developing/managing/rent 71 Equipment rental services (excl operator) 72 Data processing/IT maintenance and repair /hardware

and software consultancies/other related activities 73 Research & development services 74 Business services/advertising/accounting/industrial/

technical/recruitment/security/legal/architects etc Other services

75 General public service activities 80 Education/primary/secondary/higher/adult 85 Medical/dental/veterinary/social services 90 Sewage/sanitation/refuse services 91 Activities of membership organisations 92 Recreational/cultural/sports activities/news agencies 93 Other services not elsewhere classified 95 Domestic services in private households 96 Undifferentiated goods producing activities*

MA4 Foreign Direct Investment 2005, Crown copyright 2007 135

97 Undifferentiated services producing activities* 99 Extra territorial organisations & bodies

Split 2 digit SIC2003 industry codes

65 Financial intermediation services 651 Other non banking intermediation services 652 Banks [65.121 only] 66 Insurance services

661 Life (long term) insurance services [66.01] 662 Non-life (general) insurance & pension

services [66.02 and 66.03] 67 Other financial intermediation services

671 Security dealers [67.122 only] 672 Other financial intermediation services 74 Business services

741 Holding companies [74.15 only] 742 Other business services *Only difference between SIC(92) and SIC(2003) at two digit level is the addition of these new codes. (see article outlining changes and reasons from link to National Statistics website below) National Statistics Online A full listing of the SIC2003 classification is also available from the National Statistics website National Statistics Online - Product - Standard Industrial Classification (SIC)

file:///F|/eattree/sds/marker_panel_info.txt[18/01/2016 09:28:59]

The marker panel data set has been created to allow researchers an easymethod of linking Multinational Enterprise (MNE) information to otherdata sets through reporting unit reference numbers (RU_ref).

The marker variable identifies the following types of company withinthe AFDI:

1= UK based MNE (i.e. UK based company with subsidiaries based outside the UK)2= Foreign owned subsidiary based in the UK3= Foreign owned UK based MNE (i.e. UK based company with subsidiaries based outside the UK, but which is ultimately owned by a non-UK firm)

The construction of this variable can be seen in the file AFDI_marker.do whichis held in the source drive (V:\AFDI\do files for AFDI).

The reference numbers in the AFDI are not consistent with those in the ARD.A description of this issue is provided by the paper Matching ARD and AFDI(see X:\documentation\afdi\variables and notes).

Consistent reference numbers within the AFDI can be created on the basis of the following:

1 - single site enterprises in the AFDI have enterprise group reference numbers that are the same as the ARD reporing unit reference number - 11 digits.

2 - multi enterprise enterprise groups in the AFDI have enterprise group referencenumbers that are the same as the ARD enterprise group with the two digits '60'added in front.

The second AFDI marker panel "AFDI_marker_panel with consistent egrpref" contains transformedenterprise group reference numbers which allow matching to ARD on basis of enterprise groupreference number (see X:\documentation\afdi\variables and notes).

This panel was created by Bob Gilhooly and Rhys Davies (02/2007).