1.1 The purpose of the CPIS is to improve statistics of holdings of portfolio investment assets in the form of equity, long-term debt, and short-term debt. Specifically, the objectives are:

I. Purpose of the CPIS

1.1 The purpose of the CPIS is to improve statistics of holdings of portfolio investment assets in the form of equity, long-term debt, and short-term debt. Specifically, the objectives are:

  • to collect comprehensive information, with geographical detail on the country1 of residence of the issuer, on the stock of cross-border equities, long-term bonds and notes, and short-term debt instruments for use in the compilation or improvement of international investment position (IIP) statistics on portfolio investment capital. The IIP statistics, in turn, can provide information to check the coverage of recorded estimates of portfolio investment financial flows and associated investment income transactions recorded in the balance of payments.

  • to exchange the bilateral data. When the results of the CPIS become available, the data may be exchanged among the participating countries as well as with other countries. By exchanging comparable data (subject to confidentiality constraints), participating countries should be able to improve their statistics of nonresident holdings of their portfolio investment liabilities as well as associated financial flows and investment income data. (Information on the geographic distribution of nonresident creditors also becomes known.)

II. Background

1.2 The CPIS is being undertaken in response to the recommendations contained in the Report on the Measurement of International Capital Flows (the Godeaux Report), which was published by the IMF in 1992. The report highlighted the increasing importance of portfolio investment across international borders, reflecting the liberalization of financial markets, financial innovation, and the changing behavior of investors. The increased liberalization of international flows, however, has brought measurement difficulties. These difficulties have been reflected in the imbalances at the worldwide level between recorded financial assets and liabilities, with higher flows usually being recorded for liabilities than for assets. Imbalances have generally increased since the publication of the Godeaux Report: between 1990 and 1998, recorded flows in portfolio investment liabilities at the global level exceeded those in portfolio investment assets by an accumulated $950 billion.

1.3 Concerned by these global imbalances and the associated discrepancies in income flows at the global level, the IMF Committee on Balance of Payments Statistics (the IMF Committee) decided in 1993 to promote the idea of an internationally coordinated benchmark survey of long-term portfolio investment holdings to facilitate cross-country comparisons, permit data exchanges, and encourage standardization and best practice. By undertaking a benchmark survey of holdings, it was hoped that many countries, for the first time, would be in a position to obtain a reasonable estimate of the outstanding balances, at market price, of the level of portfolio investment their residents held, rather than merely summing the balance of payments flows. In this manner, a more accurate estimate might be obtained that would reduce to some extent the imbalance at the global level.

1.4 Developing such data was also seen as a basis for future data exchange and possible substitution of asset for liability data, particularly on a bilateral basis.

  • One of the problems in recording portfolio investment liabilities on a bilateral basis is that, for reasons of practicality, they are often not allocated to the country of residence of the acquirer of the claim. That is, the country of initial sale is recorded, but subsequent transactions with residents of other countries are frequently not recorded by the compilers in the country of residence of the issuer.

  • Moreover, the transaction is frequently recorded with the country of residence of the broker rather than with the country of residence of the beneficial owner, which may be different from that of the broker. Consequently, if a security of country A is sold to a resident of country B, who in turn sells it to a resident of country C, balance of payments compilers in country A may not know that the second transaction has occurred. In addition, if the same security is subsequently sold by the resident in country C to a resident of country A, that, too, may not be recorded, and country A’s external liabilities would be overstated.

1.5 As a result of these types of problems, the IMF organized the first internationally coordinated portfolio investment survey, using a common set of definitions and a common reference date. It was conducted as of the end of December 1997 (the 1997 CPIS). All major investing countries had been invited to participate in the 1997 CPIS. In the event, 29 countries2 took part. The results were published by the IMF in December 1999.3 The results are summarized in Table 1.1.

Table 1.1:

Global Portfolio Investment Assets and Liabilities (as of the end of December 1997)

(Trillions of U.S. dollars)

article image
(A) Shows global holdings of portfolio investment assets and liabilities compiled by summing across countries’ international investment position (IIP) data and, in the absence of such sources for particular countries, by summing balance of payments flows, or a combination of the most recent IIP data plus accumulated flow data to the end of 1997.(B) Shows the changes to the data in row A that result from using data from the 1997 CPIS, the international banking statistics system of the Bank for International Settlements (BIS), the results of the IMF Survey of the Geographical Distribution of Securities Held as Foreign Exchange Reserves (SEFER) that was conducted in conjunction with the 1997 CPIS, and data provided by major international organizations on their portfolio investment holdings.

1.6 Following the publication of the results of the 1997 CPIS, the IMF published an analysis4 that explores these results in depth.

III. The 1997 CPIS

1.7 Following the completion of the 1997 CPIS, the national compilers of the participating countries established the Coordinated Portfolio Investment Survey Task Force (the Task Force) to review the results, to examine to what extent the survey had met its goals, and, in the light of that assessment, to determine whether it should be repeated. The composition of the Task Force is shown at the front of this Survey Guide. The Task Force was convened in June 1999 and reported to the IMF Committee at its meetings in Santiago, Chile, in October 1999, and in Washington, D.C., in October 2000.

1.8 At the 1999 meeting, the Task Force reported that the 1997 CPIS had produced several benefits, including:

  • demonstrating that an organized effort toward standardizing the scope, coverage, timing, definitions, and concepts used in the compilation of data could be successfully coordinated among a large number of countries;

  • proving an effective vehicle for establishing and disseminating, on a worldwide basis, appropriately high methodological standards;

  • giving higher visibility to the coordinated, cross-country approach and resulting in improved budgetary support for the reporting of data on portfolio investment;

  • facilitating a greater understanding of country practices with respect to survey design and alternative approaches to data collection and encouraging exchanges of experience in these matters;

  • increasing the confidence of countries in portfolio investment data;

  • facilitating data exchange;

  • encouraging awareness of the IMF’s Balance of Payments Manual, fifth edition (Washington, 1993; BPM5), and promoting implementation of the methodology and standards presented therein; and

  • facilitating the attribution of portfolio investment assets by country, although there were some difficulties relating to issuers resident in small economies with international financial centers.

1.9 The Task Force was of the view that the lessons of international financial crises of 1997 and 1998 reinforced the importance of the 1997 CPIS.

1.10 Notwithstanding the positive aspects of the 1997 CPIS, the size of the resulting global discrepancy (about $1.7 trillion, see Table 1.1) between measured outstanding portfolio investment assets and liabilities indicated that there were significant gaps in coverage, the possibility of overestimation of global portfolio investment liabilities, and other measurement problems. The main gaps in the coverage of the 1997 CPIS for portfolio investment assets were believed to comprise:

  • the nonparticipation of some important investing countries for which alternative data sources proved inadequate;

  • the nonparticipation of a number of countries with offshore financial centers (OFCs; in this Survey Guide, the preferred term for these jurisdictions is small economy with an international financial center, or SEIFiCs) for which there were no alternative data sources;5 and

  • difficulties faced by many participating countries in capturing cross-border portfolio investment by households (and in some cases, enterprises) that do not use the services of resident custodians.

1.11 Given the underlying reasons for undertaking the CPIS in the first place—global imbalances in financial flows and stocks—the Task Force felt that the results of the 1997 CPIS were an encouraging first step. In light of the positive aspects of the 1997 CPIS and the potential room for improvements, the Task Force recommended to the IMF Committee that the CPIS be repeated, with the reference date of the end of December 2001 (2001 CPIS), and that it be undertaken on a regular basis thereafter. The IMF Committee agreed to the recommendation to conduct the 2001 CPIS and was inclined to support conducting the survey on an annual basis thereafter.

IV. The 2001 CPIS

1.12 It was decided to include cross-border portfolio investment holdings of each of the following items in the 2001 CPIS, on a “mandatory” basis:

  • equity securities;

  • long-term debt securities; and

  • short-term debt securities;

  • by country of residence of issuer.6

1.13 In addition, it was decided to include the following “encouraged” categories for all countries in a position to provide the data:

  • portfolio investment liabilities—equity securities, long-term debt securities, and short-term debt securities (each broken down by country of residence of issuer).7 The designation of these liabilities as “encouraged” reflected the predominant view that existing collection systems are unlikely to produce a reliable breakdown of portfolio liabilities by country of foreign holder. It was, however, judged useful to compare with partner countries’ holdings to determine the size of the discrepancies;

  • institutional sector of the holder, as a further disaggregation of the instrument by country of residence of the issuer; and

  • currency breakdown of all securities for assets (equities, long-term debt, and short-term debt) in aggregate (i.e., not by counterpart country). Even though it is not directly relevant for the CPIS in constructing a portfolio of assets by counterparty of issuer, the currency breakdown would provide important additional information for the analysis of a country’s balance of payments flows and reconciliation with its IIP. The Task Force considered that to include a full currency breakdown within the CPIS framework (i.e., by instrument by country of issuer8) would add materially to the collection costs and would make the resulting matrix of data too big to manage or understand.

1.14 To ensure an improved coverage in the 2001 CPIS, the IMF Committee recommended that an effort be made to encourage new participants among major investing countries and SEIFiCs. In addition, a broader participation by countries on a regional basis should be encouraged to ensure a wider application of best practice.

1.15 The IMF Committee recommended that participating countries also supply a geographical breakdown of the value of securities holdings that are included in reserve assets. Accordingly, the Survey of the Geographical Distribution of Securities Held as Foreign Exchange Reserves (SEFER) is being held in parallel with the 2001 CPIS.

1.16 The IMF Committee also recommended that a survey be conducted of large international organizations, the Survey of Geographical Distribution of Securities Held by International Organizations (SSIO), in order to obtain the value of their holdings of securities.

1.17 As with the 1997 SEFER, in order to maintain confidentiality of individual returns, special procedures will be used by the IMF to collect the data from these two surveys. Similarly, data from these two surveys will be published only in aggregated form so that no country’s or organization’s data will be identifiable. The SEFER and SSIO are discussed further in Chapter 2.

V. Organization of the Survey Guide

1.18 The Survey Guide is divided as follows. After this introduction, Chapter 2 outlines the scope and modalities of the CPIS; Chapter 3 sets out the concepts and principles; Chapter 4 describes various collection strategies; Chapter 5 describes other practical matters; and Chapter 6 reviews country experiences in conducting the 1997 CPIS. There are several appendices that provide model report forms, definitions and descriptions of instruments, and the International Securities Identification Number (ISIN) code system.


The term “country,” as used in this publication, does not in all cases refer to a territorial entity that is a state as understood by international law and practice; the term also covers some nonsovereign territorial entities.


The countries were Argentina, Australia, Austria, Belgium, Bermuda, Canada, Chile, Denmark, Finland, France, Iceland, Indonesia, Ireland, Israel, Italy, Japan, Korea, Malaysia, the Netherlands, New Zealand, Norway, Portugal, Singapore, Spain, Sweden, Thailand, the United Kingdom, the United States, and Venezuela.


See IMF, “Results of the 1997 Coordinated Portfolio Investment Survey,” (Washington, 2000).


See IMF, “Analysis of the 1997 CPIS: Results and Plans for the 2001 CPIS,” (Washington, 2000).


From a statistical perspective, a SEIFiC may be characterized in terms of the size of its transactions and positions on both sides of the balance sheet with the rest of the world in relation to the size of the domestic economy. From a statistical point of view, SEIFiCs can be characterized as jurisdictions that have financial institutions engaged primarily in business with nonresidents and having financial systems (i) with external assets and liabilities out of proportion to domestic financial intermediation designed to finance domestic economies and (ii) where enterprises owned or controlled by nonresidents play a significant role in providing financial services to nonresidents.


Only equity and long-term debt securities were “mandatory” items in the 1997 CPIS. Short-term debt was a “nonmandatory” item.


These were “nonmandatory” items in the 1997 CPIS.


And by sector of holder if that encouraged item were also compiled.