Chapter

IV. Survey Notes of Individual Countries

Author(s):
International Monetary Fund
Published Date:
August 1996
Share
  • ShareShare
Show Summary Details

Argentina

Institutions Responsible for Data Collection and Compilation

CPIS data were compiled and estimated by the National Directorate of International Accounts of the Argentine Ministry of Economy and Public Works and Services. The directorate is also responsible for compiling balance of payments statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

Some data were obtained from the records of domestic supervisory agencies having legal authority to collect information from supervised entities.

Survey Approach

The Argentine National Directorate of International Accounts conducted an aggregate survey of end-investors and supplemented survey data with estimates. The information consisted of data reported by selected categories of end-investors (banks, insurance companies, pension funds, and investment funds) to relevant supervisory agencies. (These investors generally submit supplementary data on investments in nonresident securities by means of appendices to balance sheet reports provided to supervisory agencies.) The information also included data—which were reported by other directorates—for the general government sector and estimates of the portfolio investment assets of the private nonfinancial sector. (Most of these assets were thought to be directly owned by entities located abroad and entrusted to nonresident custodians. The CPIS was the first instance in which such data were compiled for statistical purposes, and some types of data were collected for the first time. IMF-designed survey forms were not used. Records of supervisory agencies proved effective as existing sources of data.) The estimates were based mainly on the results of a survey of international financial experts, but BIS and partner country data were also used. The estimated component amounted to nearly US$24 billion and could not be allocated geographically. (Estimation methodologies are described in the Metodología de Estimación del Balance de Pagos of the Ministry of Economy and Public Works and Services.)

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks138EI100%
Other Financial Institutionsc302EI100%
General Government1EI100%
Nonfinancial SectordN/AEIN/A

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Pension funds, investment funds, and insurance companies

The nonfinancial sector was not surveyed directly. Sector holdings of nonresident securities were estimated.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Pension funds, investment funds, and insurance companies

The nonfinancial sector was not surveyed directly. Sector holdings of nonresident securities were estimated.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Data on direct investment were reported and collected separately; an attempt was made to ensure that data for the CPIS did not include direct investment positions.

Market Price Valuation

Respondents reported portfolio investments at market prices.

Treatment of Accrued Interest

Most respondents included accrued interest in market values.

Views of National Compilers on Accuracy of Reported Data

A major effort was made to cover portfolio investment assets of resident households. The prevalence of high risk during past decades encouraged the development of long-term positions in financial assets issued abroad and held through nonresident intermediaries. The mix of survey data and sectoral estimates of portfolio investment was thought to result in more reliable data. However, estimated components could not be geographically disaggregated, and this lack of detail diminished the quality of the data.

Australia

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Australian Bureau of Statistics (ABS), which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The Census and Statistics Act of 1905 provided the authority to conduct the CPIS on a compulsory basis. This legislation permitted Australian statisticians (or their delegates) to issue, if necessary, a “Notice of Direction” that instructs a person or organization to provide the information requested by a specific date. If compliance does not result, the matter may be referred to the director of public prosecutions for settlement through the courts. However, the ABS has always preferred to seek the cooperation of people and businesses being surveyed, and the vast majority of respondents recognized the importance of official statistics to the community and provided the required information without recourse to the compulsory powers available under the law.

Survey Approach

The ABS collected aggregate data from custodians and end-investors. Information for the CPIS was collected through the regular, quarterly Survey of International Investment (SII) covering large units that collectively account for approximately 90 percent of the nonresident investment in Australia and Australian investment abroad. In addition, a special collection of data was taken from all smaller resident enterprises known to have relevant portfolio investment activity. (In keeping with the SII methodology prevailing at the time, such enterprises were usually surveyed annually at the June 30 fiscal year end.) The regular, aggregate SII collection was deemed to provide, in terms of coverage and accuracy, satisfactory quarterly outputs. Some minor adjustments were made to comply with CPIS requirements. Collection, on a security-by-security basis, of information from custodians was investigated but proved to be impractical.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authorities1EI100%
Banks85EI96%
Other Financial Institutions249EI95%
9BO
General Government37EI100%
Nonfinancial Sector813EI92%

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Exemption/Simplification Thresholds

No threshold was established for survey coverage. For holdings of nonresident, short-term debt securities, data obtained from the SII were combined with estimated contributions from smaller units. (Information on small units that were not sampled was extrapolated from census data obtained one year earlier.)

Exclusion of Direct Investment

Respondents were asked to record separately data for direct investment and portfolio investment. Guidance notes to the CPIS provided clear definitions of direct investment, direct investment relationships, and portfolio investment. Respondents were required, for each instrument, to record separately the particulars of nonresident direct investors, direct investment groups residing abroad, and other nonresident investors. This method of reporting further emphasized the distinctions and relationships among these three categories.

Market Price Valuation

Respondents were asked to report portfolio investments at market prices. Guidance notes to the CPIS described possible methods for determining market value in the absence of market price data. Positions denominated in foreign currencies were converted to Australian dollars at the midpoints of appropriate buy and sell rates applicable on the reference dates.

Treatment of Accrued Interest

Respondents were asked to report market prices that did not include accrued interest. Accrued interest was added to market prices to obtain market value positions.

Country Attribution of Portfolio Investment Assets/Liabilities

Respondents were asked to identify the countries in which immediate nonresident creditors or debtors resided. For assets, these attributions were accurate parallels to countries of issuers. However, it was not possible to determine the countries of ultimate beneficial owners of liabilities.

Views of National Compilers on Accuracy of Reported Data

The overall accuracy of Australia’s CPIS data was considered good. Reported data were subject to a variety of input and output checks. The level of coverage was considered very high. Significant previous effort had been devoted to compiling and maintaining a suitable register of Australian enterprises engaging in investment activity with nonresidents; therefore, undercoverage arising from the possible exclusion of new enterprises was thought to be minor. A source of residual error was the incomplete coverage of enterprise groups. The leading enterprise in each group was asked to report for the group as a whole. Reporters occasionally failed to include all or part of the activities of one (or more) of their resident subsidiaries. Another source of residual error was the data reported by domestic custodians for portfolio liabilities. There were known reporting problems with quarterly data supplied by custodians for nonresident holdings of Australian equity and debt securities. Discrepancies also arose with annual data supplied by the same respondents. These discrepancies were largely due to differences in methods of valuation (particularly valuations of debt instruments), to different treatments of such instruments as options and warrants, and to different treatments of nonresident securities traded in Australia. Suitable corrections to quarterly data could only be made after the relevant annual data were obtained.

Austria

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Austrian Central Bank (Oesterreichische Nationalbank), which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The CPIS was compulsory.

Survey Approach

The central bank collected data on a security-by-security basis from custodians and end-investors. This approach was chosen because Austrian end-investors employed—as custodians for about 90 percent of their securities—domestic banks using ISIN codes for settlements.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authorities1EI100%
Banks660BO100%
Other Financial Institutions21EI100%
General Government1EI100%
Nonfinancial Sector110EIN/A

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Exemption/Simplification Thresholds

A threshold of ATS1 million was established for end-investors only. No threshold was established for custodian banks, which were required to provide full reports.

Exclusion of Direct Investment

A sample of resident direct investors was taken to determine the proportion of equity held (1) in custody with resident banks, (2) in custody with nonresident banks, and (3) in other ways. This information was checked against company security registers whenever the value of an investment held in custody was more than 10 percent of the total outstanding on the custodial company’s register. A similar approach—the sampling of direct investment enterprises—was taken on the liability side. This approach permitted estimates to be made of the direct investment component of all reported equity investments made abroad and of all reported inward equity investments.

Market Price Valuation

Market prices were not reported when there were no official security numbers or when Telekurs and/or the national numbering agency were unable to report market prices. Otherwise, Telekurs or OeKB (acting as the national numbering agency) provided market valuations.

Treatment of Accrued Interest

Telekurs and OeKB were asked to report market values that did not include accrued interest. The calculation of accrued interest and the addition of that interest to market prices were performed by the central bank.

Views of National Compilers on Accuracy of Reported Data

The quality of reported data was regarded as high because there were thorough internal checks and comparisons with external sources. Portfolio investment data may have included some equity shares held by Austrian residents engaging in direct investment relationships with nonresident entities. The coverage of financial and nonfinancial corporations was excellent. Some coverage problems existed for accounts held abroad by private households or by nonprofit institutions.

Belgium

Institutions Responsible for Data Collection and Compilation

CPIS data pertaining to Belgium were collected and compiled by the Belgian-Luxembourg Exchange Institute (BLEI). (Data pertaining to Luxembourg were excluded from the CPIS collection.) The BLEI is also responsible for the collection and compilation of data for the balance of payments of Belgium and Luxembourg.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

Royal decree and BLEI Regulation B6, which provide authority for the BLEI to collect balance of payments data, also provided authority for the CPIS to be conducted on a compulsory basis. In the event of noncooperation, the BLEI was permitted (by BLEI Regulation B7) to go on site and record, at the expense of the nonreporters, the required data.

Survey Approach

The BLEI collected aggregate data from custodians and end-investors. The aggregate approach was chosen because, as the CPIS was the first portfolio investment survey in Belgium, there were no resources available in the BLEI to organize the survey on a security-by-security basis. The questioning of both end-investors and custodians permitted a partial estimate to be made for the household sector.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authorities1BO100%
Banks133BO100%
Other Financial Institutions
Investment Enterprises78BO100%
Collective Investment Undertakings84EI100%
Holding Companies113EI99%
Insurance Enterprises232EI100%
Private Provident Institutions241EI100%
Nonfinancial Sector
Enterprises178EI98.9%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Exemption/Simplification Thresholds

A threshold of BEF 200 million was established for nonfinancial enterprises.

Exclusion of Direct Investment

A separate survey of direct investment was undertaken, and an effort was made to ensure that CPIS data did not include direct investment positions.

Market Price Valuation

Instructions from the Survey Guide were followed in determining market values for portfolio investments.

Treatment of Accrued Interest

No attempt was made to adjust market values for accrued interest.

Views of National Compilers on Accuracy of Reported Data

The quality of reported data was regarded as high because questionnaires were subjected to thorough controls as well as to comparisons with external sources. Possible sources of error included the use of book rather than market values and underestimation of the portfolio investments of the household sector. Although the survey response rate was nearly 100 percent, part of the household sector may not have been covered.

Bermuda

Institutions Responsible for Data Collection and Compilation

The Bermuda Monetary Authority (BMA) was solely responsible for collecting and compiling CPIS data. The BMA is responsible, as well, for compiling balance of payments statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The Bermuda Monetary Act of 1969 provided authority to conduct the CPIS, but participation was voluntary. Because of Bermuda’s status as an offshore financial center, the coercion of potential participants was considered inappropriate.

Survey Approach

The BMA collected aggregate data from custodians and end-investors. An aggregate approach was chosen because Bermuda is a small jurisdiction and the government was concerned that confidentiality issues would arise if the survey were conducted on a security-by-security basis. In addition, it was thought that a security-by-security approach would require a larger allocation of resources.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks3BO100%
Other Financial Institutionsc88EI45%
General Government1EI100%
Nonfinancial Sector172EI51%

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Many of these respondents are administrators of companies, trusts, and other legal entities and responded on behalf of the entities under their administration.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Many of these respondents are administrators of companies, trusts, and other legal entities and responded on behalf of the entities under their administration.

Exemption/Simplification Thresholds

No statistical threshold was established. The government’s decision to conduct the CPIS on a voluntary basis necessitated significant use of extrapolation in compiling survey results. It is difficult to assess the full impact of this decision on the quality of the estimates. Responses from most sectors were treated as samples of the sectoral populations, but the survey cannot be considered random because respondents thought to have significant portfolio investment holdings were targeted.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions. As the survey was conducted on an aggregate basis, it was impossible to determine whether or not, in practice, data on direct investment were excluded.

Market Price Valuation

Nearly all respondents reported, as requested, the market prices of portfolio investments at December 31, 1997. No attempt was made to adjust data that were not reported on a market price basis.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Views of National Compilers on Accuracy of Reported Data

Estimates were thought to be reasonably accurate and suitable for use as benchmarks for future projects. However, the estimate of portfolio investment held abroad by residents was considered lower than the actual amount. As Bermuda is an offshore financial center, a significant number of the companies, trusts, and other units surveyed were owned and controlled by non-Bermudan entities that did not maintain any physical presence on the island. These entities were typically administered by specialized management service firms that provided the data for the CPIS. The investments held, normally outside of Bermuda, by these entities were thought to be substantial. However, service providers located in offshore financial centers often experience difficulty in assessing the total investments of the trusts, companies, or other entities that they administer. Recognition of this difficulty led the BMA to assume that the actual amount of portfolio investment held abroad was greater than the amount indicated by CPIS results. Such potential undercounting was viewed as the primary deficiency in the estimates compiled for Bermuda, although nonresponses occasioned by the voluntary nature of the survey were also a factor.

Canada

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by Statistics Canada, which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The Statistics Act, Chapter S19 of the 1985 Revised Statutes of Canada, provided the authority to conduct the CPIS on a compulsory basis. Nonrespondents received telephone calls, explanation letters, and—if necessary—follow-up letters.

Survey Approach

Statistics Canada collected CPIS data on a security-by-security basis from custodians and end-investors. This approach permitted the reconciliation of data on transactions and positions, the validation—through details on currency and type of security—of these data, and the calculation of investment income.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks50EI100%
Other Financial Institutions160EI82%
95CU

El (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

El (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

With the exception of banks, all sectors surveyed (pension funds, mutual funds, segregated funds, insurance funds) were assumed to have no direct investment relationships. For banks, a separate direct investment survey was undertaken, and results were used to exclude direct investment from the CPIS.

Market Price Valuation

Portfolio investments were reported at market prices by all sectors except banks. Nontraded portfolio investments held by banks were reported at book values.

Treatment of Accrued Interest

No respondents provided data that included accrued interest in the prices of debt instruments. A few respondents provided such data separately. No attempt was made to adjust market values for accrued interest.

Views of National Compilers on Accuracy of Reported Data

Brokerage firms and nonfinancial corporations were not included in the CPIS. The combined amount of portfolio investment for brokerage firms and nonfinancial corporations was estimated to be a minimum of C$20 billion held in foreign stocks, bonds, and short-term securities. Portfolio assets of banks were valued at the lesser of book or market values. As brokerage firms and nonfinancial corporations were not covered and bank assets were reported at book values, the resulting market value coverage amounted to approximately 87 percent.

Chile

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Central Bank of Chile (Banco Central de Chile), which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The CPIS was conducted on a voluntary basis. In the event of nonresponses, alternative sources of data (information from regulatory agencies) were used.

Survey Approach

The central bank collected data on a security-by-security basis from end-investors and on an aggregate basis from custodians. As end-investors possessed the most accurate and complete information about their own transactions and as the relatively small number of institutional investors in Chile held the greatest portion of portfolio investment instruments, it was possible to collect information on a security-by-security basis from these survey participants. Additional information was collected, at an aggregate level, from resident custodians. The lack of strong legal authority for requiring participation in the CPIS was an underlying reason for using the aggregate approach with resident custodians.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks11EI100%
Other Financial Institutions
International Mutual Funds12EI100%
International Investment Funds2EI100%
Pension Funds13EI100%
Fund Managers and Brokers16BO80%
General Government
Nonfinancial Sector

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No statistical threshold was established for survey coverage.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions.

Market Price Valuation

Respondents reported portfolio investments at market prices. This information was readily available as a consequence of prevailing accounting practices and existing regulations.

Treatment of Accrued Interest

Respondents included accrued interest in market values of debt instruments.

Views of National Compilers on Accuracy of Reported Data

That portion of portfolio investment measured by the CPIS was considered fairly accurate, but incomplete coverage was a major source of possible error. Coverage of institutional investors was good, but that for the household sector was incomplete. Coverage of households investing through trader agencies was limited in spite of the initial willingness of such agencies to provide information. Coverage of portfolio investments made by enterprises and households through foreign custodians or on their own accounts was lacking. The amounts invested by enterprises were considered insignificant but amounts for households may have been substantial.

Denmark

Institutions Responsible for Data Collection and Compilation

The CPIS was conducted by the Central Bank of Denmark (Danmarks Nationalbank). Statistics Denmark is responsible for the compilation of balance of payments data.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The CPIS was conducted on a voluntary basis.

Survey Approach

The central bank collected aggregate data from custodians and end-investors. The aggregate approach was chosen because of the reduced reporting burden. It was assumed that smaller companies and households used domestic custodians for portfolio investments made abroad and that larger companies used nonresident custodians for some portfolio investments made abroad.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authorities1EI100%
Banks103BO100%
Other Financial Institutions80EI100%
General Government1EI100%
Nonfinancial Sector
Enterprises32EI100%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Exemption/Simplification Thresholds

An exemption threshold of DKK5 million was established for custodians; custodians with holdings of less than DKK10 million and banks having bank-owned holdings of less than DKK10 million were asked to report totals without geographical details. A threshold of DKK50 million was established for end-investors. No sample survey was conducted to cover portfolio investments made abroad by households and held directly or in custody.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions. Data on all custodial portfolio investments amounting to more than DKK1 million were compared with data in direct investor records.

Market Price Valuation

Respondents reported portfolio investments at market prices.

Treatment of Accrued Interest

Respondents did not include accrued interest in the market values of debt instruments. No attempt was made to adjust market values for accrued interest.

Views of National Compilers on Accuracy of Reported Data

Overall results were considered reliable, although there was some doubt about the ability of respondents to distinguish accurately between direct investment and portfolio investment. Even though portfolio investment assets made by resident households and held in custody abroad were not measured, overall survey coverage was viewed as high.

Finland

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Bank of Finland, which is responsible for the compilation of balance of payments statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The Bank of Finland Act provided authority to conduct the CPIS on a compulsory basis.

Survey Approach

The Bank of Finland collected aggregate data from custodians and end-investors. The aggregate approach had been used as part of the collection system for IIP reporting.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks3EI100%
Other Financial Institutions38EI100%
23CU
General Government
Nonfinancial Sector
Enterprises10EI100%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions.

Market Price Valuation

Respondents reported portfolio investments at market prices.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Views of National Compilers on Accuracy of Reported Data

Overall results were considered reliable and consistent with aggregate balance of payments data. Comparison, by instrument, of aggregate-level CPIS data with aggregate-level balance of payments data indicated a high level of consistency

France

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Central Bank of France (Banque de France), which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The CPIS was compulsory.

Survey Approach

The central bank collected data on a security-by-security basis from custodians (banks and investment firms) having sole legal authority to manage foreign securities on behalf of residents. For this reason, a survey of depository institutions was preferred to a survey of end-investors as the latter would have provided poor coverage of households. In addition, the security-by-security approach was already used for balance of payments compilation.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks215CU100%
Other Financial Institutions11CU100%
General Government
Nonfinancial Sector

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

The central bank maintains a separate database on direct investment relationships. The database contains information on direct investments made abroad by resident banks. These records were used to ensure that direct investments made abroad by banks were excluded from the CPIS. Direct investments made abroad by authorized, nonbank depositories were thought to be minimal.

Market Price Valuation

Respondents reported portfolio investments at market prices—with the exception of those that were recorded on the nontrading books of resident banks and reported at book values. The latter accounted for about 25 percent of total holdings of outward, long-term debt securities. No attempt was made to adjust the book values to market prices. Data on trading portfolios of banks and CPIS data collected on a security-by-security basis were collected through separate systems.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Views of National Compilers on Accuracy of Reported Data

Overall results were considered reliable. The major sources of possible error were (a) the lack of data for portfolio investments made abroad and held directly by nonbanks, (b) unreported direct investment shares, and (c) the book valuation of nontrading portfolios of resident banks. Coverage of CPIS compulsory items was complete.

Iceland

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Central Bank of Iceland, which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The CPIS was compulsory.

Survey Approach

The central bank collected aggregate data from custodians and end-investors. This approach was consistent with the system used to compile balance of payments data.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks4BO100%
Other Financial Institutions99BO100%
General Government
Nonfinancial Sector

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions.

Market Price Valuation

Respondents reported portfolio investments at market prices.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Views of National Compilers on Accuracy of Reported Data

Survey results were thought to be very accurate and complete. The main source of potential error was the inadequate coverage of portfolio investments made abroad and held directly by households and enterprises, but this omission was not considered significant.

Indonesia

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Central Bank of Indonesia, which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The CPIS was voluntary. Custodial nonresponse was followed up by telephone and letter.

Survey Approach

The central bank collected aggregate data from custodians and end-investors. For the collection of data on portfolio investment assets, a survey of end-investors was preferable because many investors used the services of nonresident custodians. For the collection of data on portfolio investment liabilities, a survey of custodians was preferable because domestic portfolio investment securities bought by nonresident investors were usually purchased through domestic brokers.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks108EI100%
25CU
Other Financial Institutions
General Government
Nonfinancial Sector
Enterprises10 groupsEI30%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

No attempt was made to exclude data on direct investment from CPIS responses. In part, the lack of exclusion was due to the inability of custodians to make the distinction between direct investment and portfolio investment.

Market Price Valuation

Respondents reported portfolio investments at market values.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Country Attribution of Portfolio Investment Liabilities

End-investors were identified on the basis of ownership records maintained by custodian banks.

Views of National Compilers on Accuracy of Reported Data

Overall results were considered reliable. There were no obvious sources of error. On the asset side, coverage of the banking sector and of large companies was complete.

Ireland

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Central Statistical Office (CSO), which is also responsible for the collection and compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

Information required for the CPIS was collected as part of arrangements for quarterly and annual collections of balance of payments data. These arrangements were established in 1996 in accordance with a ministerial order issued under the aegis of the Statistics Act of 1993. In the event of nonresponse, sanctions could be pursued through the courts. Nonrespondents could, upon conviction, be fined and subject to subsequent daily fines for continued nonresponse. The CSO attempted to encourage responses without recourse to legal enforcement and offered assistance to respondents. Legal enforcement was used by the CSO only as a last resort.

Survey Approach

The CSO collected aggregate data from end-investors. Because of the integrated nature of the collection system for balance of payments, IIP, and CPIS data, the end-investor approach was preferred. Custodians were unable to provide all of the data requested. Ireland’s data collection system targets full coverage of the financial services sector and a specially selected sample of nonfinancial enterprises relevant to the balance of payments. All of the larger entities, as well as a sample of the smaller units, are covered.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banksc70EI100%
Other Finanial Institutionsd2500EI86%
General Government
Nonfinancial Sectore
Enterprises500EI
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Credit institutions licensed to undertake banking activities

Includes institutional investors that acquire holdings of securities issued by nonresidents on behalf of resident Irish businesses and private clients (households)

Nonfinancial enterprises in Ireland do not appear to engage in portfolio investments to any significant extent. The response rate from the units surveyed was about 40 percent. When grossed, aggregate results for this sector accounted for only 0.1 percent of total portfolio investment abroad. This result was expected as these entities tend to be either direct investors or direct investment enterprises that may also engage in some borrowing from, lending to, or holding of deposits with nonaffiliates.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Credit institutions licensed to undertake banking activities

Includes institutional investors that acquire holdings of securities issued by nonresidents on behalf of resident Irish businesses and private clients (households)

Nonfinancial enterprises in Ireland do not appear to engage in portfolio investments to any significant extent. The response rate from the units surveyed was about 40 percent. When grossed, aggregate results for this sector accounted for only 0.1 percent of total portfolio investment abroad. This result was expected as these entities tend to be either direct investors or direct investment enterprises that may also engage in some borrowing from, lending to, or holding of deposits with nonaffiliates.

Exemption/Simplification Thresholds

Financial service companies are required to report BOP and IIP data on an ongoing basis. Companies with low levels of activity have been permitted, upon submission of proper requests to the CSO, to provide annual rather than quarterly data. The sample selection process for nonfinancial service companies was geared towards those companies engaging, according to information in the CSO’s statistical register, in significant portfolio investment transactions (more than IE£100,000 annually) with nonresidents.

Exclusion of Direct Investment

The CPIS survey was conducted as part of the balance of payments/IIP survey; this combined approach facilitated the separate reporting of direct investment and portfolio investment. CPIS results for Ireland therefore excluded direct investment.

Market Price Valuation

Respondents reported portfolio investments at market prices. Plausibility checks on survey responses confirmed that market price valuations were provided by respondents. Market valuation appears to be used extensively in the day-to-day business of companies.

Treatment of Accrued Interest

Respondents were asked to report market prices that did not include accrued interest as well as market price transactions and valuation changes. In addition, details of outstanding interest (positions) and movements in interest (flows) were required. For the CPIS, the market price positions that excluded accrued interest were combined with the outstanding interest to obtain the full market values of the investments.

Views of National Compilers on Accuracy of Reported Data

Survey results were considered reliable because of the internal and external data plausibility checks that were undertaken. Coverage of financial services sector activity vis-à-vis nonresidents was largely complete. The portfolio investment activity of the nonfinancial services sector appeared insignificant as it accounted for only 0.1 percent of the total portfolio assets reported for all sectors. There were no deviations from the recommended CPIS methodology for data collection, and there is no evidence to suggest that respondents departed, to any significant extent, from specific CPIS requirements.

Israel

Institutions Responsible for Data Collection and Compilation

The Assets and Liabilities Unit of the Bank of Israel’s Foreign Exchange Control Department compiled the CPIS. The Central Bureau of Statistics compiles balance of payments statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The Foreign Exchange Control Law provided the authority to conduct the CPIS on a compulsory basis.

Survey Approach

The Bank of Israel collected data on a security-by-security basis from custodians. This approach facilitated reporting of data by custodians and compilation by the Bank of Israel.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks19BO100%
Other Financial Institutions
Brokers12BO100%
Institutional Investor380EI100%

El (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data

El (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data

Exclusion of Direct Investment

The holdings of each investor were compared with total issues of each security. To complement these comparisons, investor identification data were matched on a security-by-security-basis to identify direct investment relationships.

Market Price Valuation

Respondents reported portfolio investments at market prices.

Treatment of Accrued Interest

No attempt was made to adjust market prices for accrued interest.

Country Attribution of Portfolio Investment Liabilities

This information was supplied by custodians.

Views of National Compilers on Accuracy of Reported Data

Overall results were considered reliable. Coverage of financial intermediaries and institutional investors was complete, and coverage of other sectors was 70–80 percent.

Italy

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Ufficio Italiano dei Cambi (UIC), which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

Ministry of Treasury Decree No. 721963 of October 6,1995 provided the authority to conduct the CPIS on a compulsory basis. Nonrespondents could be fined according to the national statistical law.

Survey Approach

The UIC collected data on a security-by-security basis from custodians and end-investors. Use of the end-investor approach was necessary for compilers to distinguish between direct investment and portfolio investment.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks630BO95.2%
Other Financial Institutions497BO92.3%
General Government
Nonfinancial Sector1676EI89.7%

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

A threshold of LIT 500 million was established for survey coverage.

Exclusion of Direct Investment

The UIC collected separate data for direct investment and portfolio investment. For purposes of estimating outward direct investment, sample data were grossed to form the basis for an estimate of direct investment equity.

Market Price Valuation

Most respondents reported portfolio investments at market prices. Reported prices were checked against market valuations provided by administrators of resident investment funds.

Views of National Compilers on Accuracy of Reported Data

The UIC conducted a very detailed survey. There were no significant deviations from the methodological practices recommended in the Survey Guide. Overall results were considered reliable. The main deficiencies were the omission of data on portfolio investments made abroad and held directly by households and the possible misclassification of some direct investments as portfolio investments. Survey data were broadly in line, at an aggregate level, with previous estimates for Italy’s IIP.

Japan

Institutions Responsible for Data Collection and Compilation

The Japanese Ministry of Finance was responsible for conducting the CPIS. However, in practice, data collection and compilation for the survey were entrusted to the Bank of Japan.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

Authority for conducting the CPIS was provided by the Foreign Exchange and Foreign Trade Law and was compulsory for major banks, securities firms, and institutional investors. In the event of nonresponse, telephone calls and reminder letters were sent.

Survey Approach

The Bank of Japan collected aggregate data from custodians and end-investors. The aggregate approach was chosen because this approach reduced the reporting burden for respondents. Custodians were surveyed because the bulk of securities transactions with nonresidents were handled by resident custodians. Only institutional investors transacted directly with nonresidents.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks362BO100%
Other Financial Institutions231BO100%

EI (end-investor). CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor). CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

A threshold of yen 100 million was established for positions denominated in yen, and a threshold of yen 1 million was established for positions denominated in foreign currencies.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions.

Market Price Valuation

Almost all respondents reported portfolio investments at market prices. No adjustments were made for the few instances when portfolio investments were reported at book values.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Country Attribution of Portfolio Investment Liabilities

A transactor principle was applied to disaggregate inward portfolio investment by region. Data on portfolio investment liabilities pertaining to 30 countries had previously been published by Japan; additional geographical details were expressly compiled for the CPIS.

Views of National Compilers on Accuracy of Reported Data

Overall results were considered reliable. The main deficiency was the incomplete coverage of portfolio investments made abroad and held directly by households and enterprises. However, the amounts involved were considered small. The availability of information on portfolio investment liabilities was limited by the absence of data sources for nonresidents using nonresident custodians. These amounts were also considered small. As major investors were obliged to report data in accordance with terms of the Foreign Exchange and Foreign Trade Law, the level of coverage for the CPIS was high.

Korea

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Bank of Korea (BOK), which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

Article 86 of the Bank of Korea Act of April 1, 1998 provided the authority for the BOK to conduct the CPIS on a compulsory basis. Provisions of Article 86 permit the BOK to compile statistics to be used for formulation of the bank’s monetary and credit policies.

Survey Approach

The BOK collected aggregate data from custodians and end-investors. According to present Korean foreign exchange regulations, portfolio investments must be made by authorized foreign exchange banks or institutional investors such as securities companies, insurance companies, or investment trust companies. Therefore, the preferred approach was to collect data from intermediary investors.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authorities1EI100%
Banks116EI100%
Other Financial Institutions108BO100%

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions. Direct investment abroad was defined as that occurring when a resident investor owns 20 percent or more of the ordinary shares or voting rights of an incorporated enterprise or, for an unincorporated enterprise, the equivalent of 20 percent or more or the ordinary shares or voting rights.

Market Price Valuation

Respondents reported portfolio investments at market prices. The Survey Guide provided information on valuations methods that could be used when market price data were not available.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Views of National Compilers on Accuracy of Reported Data

Overall results were considered reliable. However, outward portfolio investment may have been underestimated as a consequence of reporting problems resulting from closures of some foreign exchange banks in late 1997.

Malaysia

Institutions Responsible for Data Collection and Compilation

The Central Bank of Malaysia (Bank Negara Malaysia) was responsible for data collection and compilation for the CPIS. The Department of Statistics is responsible for data collection and compilation of balance of payments and IIP statistics. Both CPIS and balance of payments data are exchanged by the central bank and the statistics department.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

Section 30 of the Central Bank Act of 1958 provided the authority to conduct the CPIS on a compulsory basis.

Survey Approach

The central bank collected data on a security-by-security basis from custodians and end-investors. This approach was chosen because it resulted in broader, more comprehensive coverage.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Others804BO
Of Which:
Respondents60480%
Respondents Having
Portfolio Investments207
Banks25
Other financial Institutions180
General Government2
Nonfinancial Sector

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions.

Market Price Valuation

Respondents reported portfolio investments at market prices. The Survey Guide provided information on valuation methods that could be used when market price data were not available.

Treatment of Accrued Interest

CPIS respondents were asked to report debt instruments at issue prices.

Country Attribution of Portfolio Investment Liabilities

Requirements for details on the residency of nonresident holders of portfolio investment liabilities were specified in the survey forms. Only a few respondents were unable to identify the countries of attribution.

Views of National Compilers on Accuracy of Reported Data

Overall CPIS results were considered reliable. For inward portfolio investment, data were broadly similar to data from Kuala Lumpur Stock Exchange (KLSE) surveys. Thus, the net portfolio investment position (RM62.4 billion) was comparable to the RM69 billion of KLSE stocks (23 percent of KLSE capitalization) held by nonresident investors.

The Netherlands

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Central Bank of the Netherlands (De Nederlandsche Bank), which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The External Financial Relations Act provided the authority to conduct the CPIS on a compulsory basis.

Survey Approach

The central bank collected aggregate data from custodians. A survey of custodians was preferred as custodians were already reporting data for balance of payments transactions. Custodians were considered to be the only realistic source of information on portfolio investment liabilities. Custodial sources of data for outward portfolio investment were supplemented by direct sources.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks32CU80%
Other Financial Institutions11CU90%

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

Although no formal exemption threshold was established, minor custodian banks were excluded from the survey.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions.

Market Price Valuation

Respondents reported portfolio investments at market prices.

Treatment of Accrued Interest

Respondents did not include accrued interest in the market values of debt instruments. No attempt was made to adjust for accrued interest.

Country Attribution of Portfolio Investment Liabilities

A transactor principle was applied to disaggregate inward portfolio investment by region. When counterparts to resident custodians were nonresident banks or clearing institutes, the residency of end-investors could not be determined. The largest Dutch enterprises listed on foreign stock exchanges were asked to provide geographically disaggregated estimates of the market values of shares that were issued by the enterprises themselves and held directly abroad by nonresidents. The results of these inquiries, which were mainly based on geographical distributions of dividend flows, were added to the results of the custodian survey.

Views of National Compilers on Accuracy of Reported Data

The most important sources of error for portfolio assets were incomplete data for portfolio investments made abroad and held directly by nonfinancial institutions and private households, the inability of some banks to provide the required disaggregations of data by sector, and the lack of geographical disaggregation for a relatively small portion of the reported data (especially that for direct holdings of securities abroad). The most important sources of error for portfolio liabilities were geographical disaggregation by nominee (foreign bank or clearing institute) rather than by beneficiary and the lack of information on Dutch bonds directly held abroad. Data on portfolio assets and liabilities were generally consistent with balance of payments flows and share, bond, or currency indices.

New Zealand

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by Statistics New Zealand, which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The Statistics Act of New Zealand provided the authority to conduct the CPIS on a compulsory basis.

Survey Approach

Statistics New Zealand collected aggregate data from custodians and end-investors. An aggregate survey for balance of payments and IIP data was on-going, and the aggregate approach could be implemented at a lower cost.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks10BO100%
Other Financial Institutions30BO100%
General Government1EI100%
Nonfinancial Sector
Enterprises381EI95%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian).

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian).

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

CPIS data were checked against balance of payments data to identify direct investment relationships. Spot checks of information provided by larger enterprises were made to detect direct investment relationships.

Market Price Valuation

A significant majority of the respondents reported portfolio investments at market prices. Only a few reported book values. The gross value of portfolio investment assets reported at book values was insignificant as a proportion of total portfolio investment assets. Therefore, no adjustments for book values were made.

Treatment of Accrued Interest

Respondents included accrued interest in the market prices of debt instruments.

Views of National Compilers on Accuracy of Reported Data

Overall results were considered reliable. Some of the estimates made by respondents for country and instrument disaggregations reduced the level of accuracy. All known transactors were included in the survey, and overall coverage of 95 percent was achieved. Complete coverage was achieved for all the major respondents.

Norway

Institutions Responsible for Data Collection and Compilation

Statistics Norway was responsible for collection and compilation of the CPIS. Balance of payments statistics are also compiled by Statistics Norway and are partly based on data collected by the Central Bank of Norway. Both CPIS and balance of payments data were exchanged by the two institutions.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The Currency Control Act provided the authority to conduct the CPIS on a compulsory basis. The act included a provision for fining nonrespondents.

Survey Approach

Statistics Norway collected aggregate data from end-investors. This approach was preferred because most outward portfolio investment was concentrated among institutional investors. In addition, the end-investor approach allowed the CPIS to be conducted as a supplement to an existing financial survey on nonresident positions by type. A sample survey was made of the household sector; information reported to Norwegian tax collectors was used to gather data on holdings of nonresident securities. A sample of data from supplementary tax records that showed the composition and country allocation of nonresident securities was grossed to provide a total for all households. The sample was intended to cover the largest holders; actual coverage amounted to nearly 10 percent of the total sector. When sample data were grossed, corrections were made for the market values of the securities. As this was the first portfolio investment survey in Norway and a reporting system had not been established, it was assumed that more experience would be necessary before custodians were able to report all data satisfactorily. A security-by-security survey was considered too costly.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks25EI100%
Other Financial Institutions243EI98%
General Government49EI100%
Nonfinancial Sector
Enterprises3022cEI97%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Of these, 700 entities reported portfolio holdings. The holdings were highly concentrated. Any entities not included in the sample were probably insignificant.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Of these, 700 entities reported portfolio holdings. The holdings were highly concentrated. Any entities not included in the sample were probably insignificant.

Exemption/Simplification Thresholds

No threshold was established. CPIS coverage of all but the household sector encompassed the universe of security holders.

Exclusion of Direct Investment

A survey of outward direct investment was regularly undertaken by the central bank. However, this survey was not fully comparable with the CPIS because different definitions of direct investment were used for the surveys. The central bank survey of direct investment did provide an indication of which entities should be further investigated for possible inaccurate reporting of direct investment as portfolio investment. This approach was hindered by the absence of balance sheet data for nonfinancial enterprises; such data could have confirmed misclassifications.

Market Price Valuation

Most respondents reported portfolio investments at market prices. No adjustments were made to data reported at face or book values.

Treatment of Accrued Interest

Respondents did not include accrued interest in the market prices of debt instruments. No attempt was made to adjust for accrued interest.

Views of National Compilers on Accuracy of Reported Data

The initial reconciliation between flows and stocks indicated that CPIS results for total portfolio assets were fairly accurate because flow data were reported and compiled independently of CPIS data. In addition, comparisons with data from balance sheets, the financial survey, and the direct investment survey confirmed the reliability of CPIS data. The most important source of error was probably the lack of data on portfolio investment made abroad and held directly by the household sector.

Portugal

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Central Bank of Portugal (Banco de Portugal), which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The law establishing the Banco de Portugal and providing the authority to collect data for balance of payments statistics also provided the authority to conduct the CPIS on a compulsory basis. Although this legislation established a sanction procedure that results primarily in fines, sanctions were not necessary as almost all follow-up contacts with nonreporters were successful.

Survey Approach

The central bank collected data on a security-by-security basis from custodians and end-investors. The security-by-security approach was considered the most effective way to ensure that data of adequate quality were reported. It was assumed that the bulk of outward portfolio investment was engaged in through resident custodians.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks40BO85%
Other Financial Institutions14BO93%
General Government
Nonfinancial Sector
Enterprises22EI59%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions. To avoid double counting, CPIS data were cross-checked with the results of a parallel survey on direct investment.

Market Price Valuation

Most respondents reported portfolio investments at market prices. However, a lack of information on the market prices of some securities, the practice of reporting the unit values of securities (the market value of a security divided by its price) rather than the total market values of securities, and the reporting of unit values of securities as percentages of nominal values led to further discussion with respondents. Publications of the Lisbon Stock Exchange were consulted as a means of checking the unit values reported by CPIS respondents for some securities and as a means of checking the consistency of total values reported for the CPIS against the total value of each investment fund. Publications of individual investment funds provided information on portfolio composition (quantity, market price, and currency) as of relevant dates. Respondents were asked to use the alternative market valuation methods suggested in the Survey Guide if they were unable to report market prices.

Treatment of Accrued Interest

Respondents were asked to include accrued interest in market price valuations. When custodians and investors were unable to do so, no adjustments were made. When accrued interest could be determined by checking publications of the Lisbon Stock Exchange, such interest was added to the total stock values of investment funds.

Country Attribution of Portfolio Investment Liabilities

Resident custodians are required to notify the Banco de Portugal of nonresident transactions and positions in domestic securities. A notification must provide, among other things, information on the country of the nonresident investor. In practice, residency was determined on the basis of the country of the first known counterpart, which could have been the end-investor or the intermediary.

Views of National Compilers on Accuracy of Reported Data

Overall results were considered reliable, although there was some miscoding of short- and long-term instruments. Satisfactory coverage was achieved, but the quality of data for the nonfinancial sector should be improved.

Singapore

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Singapore Department of Statistics, which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The Statistics Act of 1990 provided the authority to conduct the CPIS on a compulsory basis. All statistics collected, compiled, and published by the Singapore Department of Statistics are governed by this act.

Survey Approach

The Singapore Department of Statistics collected aggregate data from custodians and end-investors. The survey of end-investors was supplemented with data reported by custodians. Custodian reporting extended survey coverage to the household sector and provided a means of making cross checks with data from end-investors. Collection of data on a security-by-security basis was not feasible as ISIN codes are not widely used in Singapore.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks172EI95%
Other Financial Institutions129EI
119CU96%
37BO
General Government
Nonfinancial Sector
Enterprises5,262EI89%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions. Custodians were assumed not to hold direct investments.

Market Price Valuation

Respondents reported portfolio investments at market prices.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Views of National Compilers on Accuracy of Reported Data

Investors with significant holdings provided more comprehensive data as such investors usually had more complete records for their investments. Some investors with smaller holdings were unable to provide accurate market value data. CPIS coverage of custodians was complete. Coverage of end-investors was reasonably complete as the register of the Singapore Department of Statistics was considered reliable and current.

Spain

Institutions Responsible for Data Collection and Compilation

The Spanish Ministry of Economy and Finance was responsible for the collection and compilation of the CPIS. The Bank of Spain is responsible for the collection and compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The Spanish Foreign Investment Law provided the authority to conduct the CPIS on a compulsory basis. This law, which is different from the regulation governing the collection of balance of payments data, permits the Spanish Ministry of Economy and Finance to collect information on cross-border holdings of securities. Both decrees include penalties for nonresponse.

Survey Approach

The Spanish Ministry of Economy and Finance collected data on a security-by-security basis from custodians and end-investors. These approach was preferred as it was already in use for balance of payments compilation. Reporting by custodians made the process simpler and more efficient. Reporting by end-investors was required only when investments were held directly abroad.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks104BO100%
Other Financial Institutions80EI100%
10BO
General Government
Nonfinancial Sector
Enterprises170EI100%
Households350EI100%

El (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

El (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Direct equity investments made abroad were identified from administrative records and excluded from the CPIS. Direct investments in debt instruments were assumed to be minimal.

Market Price Valuation

Respondents reported portfolio investments at market prices. When securities were not actively traded, custodians used market prices of similar instruments to provide data that approximated market values. Securities denominated in foreign currencies were converted into pesetas at end-of-period exchange rates.

Treatment of Accrued Interest

Respondents included accrued and unpaid interest in reported market values.

Country Attribution of Portfolio Investment Liabilities

Domestic custodians were asked to report the countries of nonresident holders of Spanish securities.

Views of National Compilers on Accuracy of Reported Data

Survey results were considered fairly accurate. There was a gap in the coverage of portfolio investments made abroad and held directly by households and enterprises. A possible source of error was the inclusion of data on sell and buy-back transactions. Survey coverage was otherwise considered complete as all institutions were obliged to report.

Sweden

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Central Bank of Sweden (Sveriges Riksbank), which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

Article 50 of the Sveriges Riksbank Act provided the authority to conduct the CPIS on a compulsory basis. This legislation requires that Sveriges Riksbank shall make public statistical data concerning foreign exchange and credit conditions. Article 22 gives the central bank authority to collect balance of payments data. The legal authority for collecting balance of payments statistics is the same as that for conducting the CPIS.

Survey Approach

The central bank collected aggregate data from custodians and end-investors. This approach, which was used for existing balance of payments and IIP surveys, made it possible to compare CPIS results with those of previous surveys.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks31BO87%
Other Financial Institutions332BO85%
General Government16EI100%
Nonfinancial Sector
Enterprises1,077EI86%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions.

Market Price Valuation

Respondents reported portfolio investments at market prices.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Views of National Compilers on Accuracy of Reported Data

Overall results were considered reliable. All known end-investors were included in the survey.

Thailand

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Bank of Thailand, which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The CPIS was conducted on a voluntary basis. The Bank of Thailand Act (B.E. 2485) and the Royal Decree Regulating the Affairs of the Bank of Thailand (B.E. 2485) authorize the Bank of Thailand (BOT) to collect data from financial institutions. Foreign exchange regulations require resident households and entities in the corporate sector to obtain approval from the BOT to invest abroad.

Survey Approach

The BOT collected aggregate data from end-investors. The aggregate approach, which was used for existing balance of payments surveys, was expected to provide reliable data.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks15EI100%
Other Financial Institutions54EI100%
General Governmentc
Nonfinancial Sectord

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

The public sector had no holdings of nonresident equity or debt securities.

Portfolio investment holdings of the nonfinancial sector were estimated from the accumulation of net balance of payments flows.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

The public sector had no holdings of nonresident equity or debt securities.

Portfolio investment holdings of the nonfinancial sector were estimated from the accumulation of net balance of payments flows.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions. With the exception of those made by nonfinancial corporations, direct investments in equity securities were excluded from the CPIS in accordance with the 10 percent criterion. Holdings, by nonfinancial corporations, of nonresident equity securities of less than 25 percent of ordinary shares were treated as portfolio investment.

Market Price Valuation

Most respondents were unable to report portfolio investments at market prices. Values reported were mostly book values. Securities denominated in foreign currencies were converted at appropriate end-of-period exchange rates into domestic currency.

Treatment of Accrued Interest

Respondents did not include accrued interest in the market values of debt instruments. No attempt was made to adjust for accrued interest.

Views of National Compilers on Accuracy of Reprted Data

Overall results for banks and financial companies were considered relatively accurate, although problems arose from the use of book values. Coverage of banks and financial institutions was complete; coverage of other sectors was much less reliable, but foreign exchange records provided useful data.

United Kingdom

Institutions Responsible for Data Collection and Compilation

The Bank of England, which collects balance of payments and IIP data from the banking sector, was also responsible for collecting bank sector data for the CPIS. The U. K. Office for National Statistics, which collects and compiles balance of payments and IIP data from the nonbank sector, was also responsible for collection and compilation of nonbank sector data for the CPIS. There is close coordination between the Bank of England and the U. K. Office for National Statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

Nonbanks participated in the CPIS on a statutory basis in accordance with the 1947 Statistics of Trade Act, which provides for legal enforcement. However, as the CPIS was not intended to be a regular survey, legal enforcement was not applied. Nonresponse was a problem; nonresponding companies were reminded by telephone and letter. Although the Bank of England has statutory authority to collect statistics from the banking sector, in practice, the participation of banks was voluntary. Follow-up action was not necessary as the response rate of banks was high.

Survey Approach

The Bank of England and the U. K. National Office of Statistics collected aggregate data from end-investors. This approach was consistent with that used to compile balance of payments data. Nonbanks were surveyed through the attachment of an additional form to the annual balance of payments inquiry sheets used to collect data for the international investment position. Data on short-term debt securities were collected only from banks; the amount of short-term debt securities held by nonbanks was thought to be minimal.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banksc330EI100%
Other Financial Institutionsd680EI61%
General Government
Nonfinancial Sector
Enterprises22EI60%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

As of end-December 1997, the Banking Department of the Bank of England was classified within the banking sector.

Response rates for nonbanks were 75 percent for long-term insurance firms, 58 percent for general insurance firms, 63 percent for pension funds, 51 percent for unit trusts, 43 percent for investment trusts, 71 percent for securities dealers, and 59 percent for nonfinancial companies.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in CPIS data.

As of end-December 1997, the Banking Department of the Bank of England was classified within the banking sector.

Response rates for nonbanks were 75 percent for long-term insurance firms, 58 percent for general insurance firms, 63 percent for pension funds, 51 percent for unit trusts, 43 percent for investment trusts, 71 percent for securities dealers, and 59 percent for nonfinancial companies.

Exemption/Simplification Thresholds

No threshold was established for nonbank financial institutions. Only those nonfinancial companies reporting holdings of foreign equity or debt in the quarterly financial assets and liabilities survey were approached for the CPIS. Banks with thresholds of £100mn of total external business (mainly deposits and/or lending) in all currencies or thresholds of £20mn of external sterling business were asked to complete existing bank survey forms for the CPIS.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions.

Market Price Valuation

Respondents reported portfolio investments at market prices. When nonbanks could only provide data for the financial year end rather than the calendar year-end reference period, the reported data were accepted without adjustment.

Treatment of Accrued Interest

Respondents did not include accrued interest in the market values of debt instruments. No attempt was made to adjust for accrued interest.

Views of National Compilers on Accuracy of Reported Data

Data from U. K. banks were regarded as accurate. As the survey of nonbanks covered only a sample of these institutions, there was sampling error. In addition, there was nonresponse error because nonbank data were compiled from the responses of approximately 60 percent of the nonbanks in the sample. In terms of the total value of assets held by the nonbanks in the sample, the response rate was somewhat higher. Possible errors in reported data were minimized through cross checks with other sources. For regular reporting of balance sheet data (global data only), the target response rate, which varied among institutional sectors, was usually at least 75 percent. Because compulsory reporting was not enforced and because compilers did not wish to jeopardize responses to the global inquiries, they exercised judgment in pursuing nonresponses to the CPIS part of the inquiry. Consequently, the response rate was lower. Imputation procedures that normally would have been used for individual nonrespondents were not appropriate as there were no historical data upon which to base the calculations. There was no evidence to suggest that CPIS results were biased in any particular direction.

United States

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the U. S. Treasury Department. Balance of payments and IIP data are collected and compiled by the U. S. Commerce Department. There is close coordination between the two departments.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The International Investment and Trade in Services Survey Act provided the authority to conduct the CPIS on a compulsory basis. Nonresponse was punishable by a fine of up to $25,000 and a prison term not to exceed one year. Until satisfactory responses were received, firms were reminded of the responsibility to report.

Survey Approach

The U. S. Treasury Department collected data on a security-by-security basis from custodians and end-investors. This approach produced the best combination of comprehensive coverage and minimum reporting burden.

Exemption/Simplification Thresholds

A threshold of US$20 million was established. Sampling was used for custodians and end-investors unlikely to respond, but the exemption level remained the same. The sampling procedure consisted of grouping types of investors or custodians (for example, pension funds with assets between X and Y dollars) and surveying approximately 100 such firms in each group.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authorities1BO100%
Banks427BO100%
Other Financial Institutions677BO100%
General Government
Nonfinancial Sector
Enterprises98EI100%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Exclusion of Direct Investment

To determine the existence of any direct investments, data reported for positions in equities were compared with available data on the number of shares outstanding. If any respondent reported a position in excess of 10 percent of total shares outstanding, the respondent was contacted to verify that no direct investment was included.

Market Price Valuation

Information on market prices for large, frequently traded debt instruments (which constituted the preponderance of debt security issues and the vast majority of equity security issues) was readily available from custodians and end-investors. When respondents were unable to provide market prices, information from commercial sources and other reporters was used. When all such sources were exhausted, market prices for portfolio investments were estimated.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Views of National Compilers on Accuracy of Reported Data

CPIS data on total portfolio investment, types of securities, and investors’ countries of residence were consistent with the expectations of compilers. These results were considered reasonably accurate as reported data were closely examined and tested. Whenever possible, reported positions were crosschecked with other information sources to ensure accuracy. Particular attention was paid to correcting positions reported in local currency units rather than the U. S. dollars specified in the survey instructions. However, misreported positions that eluded detection could have resulted in overstated debt security holdings. The collection of data on a security-by-security basis permitted a detailed analysis to be made of survey results. During the analysis, some of the reported data were excluded because these data pertained to short-term securities. Survey coverage was good: all major custodians and institutional investors were included; reporting was mandatory; and the collection of some confidential data was permitted. The tendency of U. S. investors to entrust their foreign holdings to a relatively small group of resident custodians facilitated conduct of the survey, and all of these custodians responded to the survey. CPIS coverage did not include portfolio investments made by households (or by smaller institutional investors) and held by foreign custodians. However, the omission was not thought to be significant.

Venezuela

Institutions Responsible for Data Collection and Compilation

CPIS data were collected and compiled by the Central Bank of Venezuela, which is also responsible for the compilation of balance of payments and IIP statistics.

Legal Authority to Collect CPIS Data (Compulsory/Voluntary Reporting)

The Central Bank Law, the Capital Market Law, and the Insurance and Reinsurance Law provided the authority to conduct the CPIS on a voluntary basis. For institutions not regulated by these laws, provision of information was compulsory. In implementing the survey, the central bank worked with various regulatory agencies (e.g., the Venezuelan National Security Commission, bank supervisors, insurance supervisors).

Survey Approach

The central bank collected data on a security-by-security basis from end-investors. An end-investor survey was preferred because a substantial portion of outward portfolio investment was engaged in through nonresident custodians. The security-by-security approach was preferred because this approach was already being used for the collection of balance of payments data.

Coverage of Institutional Units by Sector
SectorsNo. of

Units

Surveyed
ApproachaResponse

Rate
Monetary Authoritiesb
Banks81EI100%
Other Financial Institutions185EI62%
General Government1EI100%
Nonfinancial Sector
Enterprises79EI28%
Households

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in the CPIS data.

EI (end-investor), CU (custodian), or BO (both end-investor and custodian)

Securities held as reserve assets were not included in the CPIS data.

Exemption/Simplification Thresholds

No threshold was established for survey coverage.

Exclusion of Direct Investment

Guidance notes to the CPIS provided clear definitions of direct investment and portfolio investment and clear instructions for excluding direct investment positions.

Market Price Valuation

Nearly all respondents reported portfolio investments at market prices. A few market prices were obtained from commercial databases, and adjustments were made to the exchange rates supplied by respondents.

Treatment of Accrued Interest

Respondents included accrued interest in the market values of debt instruments.

Views of National Compilers on Accuracy of Reported Data

Coverage of institutions required to respond to the CPIS was good, but the quantity of information actually obtained (especially from the nonfinancial sector) was less than desirable because the regulations governing compulsory response to the CPIS were limited in scope. For this reason, portfolio investments made abroad and held directly by households and enterprises were poorly covered.

    Other Resources Citing This Publication