DEVELOPMENTS IN INTERNATIONAL FINANCIAL MARKETS

Unprecedented growth in the volume, complexity, and globalization of international financial transactions occurred during the 1980s and 1990s. Changes in international capital markets created additional avenues for raising capital, investing, and hedging risk. A broad range of new participants entered the markets. As a result, there was commensurate growth in the portfolio investment component of total international financial flows (see Chart 1).

During these two decades, countries experienced difficulty in adjusting their data collection systems to these developments—partly because the factors (such as the removal of exchange controls) that caused the rapid expansion of financial flows also adversely affected primary data sources. The globalization of financial markets required compilers to develop new data sources (such as information reported by custodians) and new methods of compilation (such as the establishment and use of databases containing information on securities). Moreover, as new instruments (such as financial derivatives) were introduced onto the markets, new methods of tracking and recording these instruments had to be found.

DISCREPANCIES IN THE WORLD FINANCIAL ACCOUNT

In the early 1980s, world Current Account balances showed large negative discrepancies. The existence of such discrepancies indicated that either the deficits of some countries had been overstated or the surpluses of other countries had been understated.⁴ During the same period, the world Financial Account showed that financings of Current Account balances had also been imperfectly reported; large, positive global discrepancies indicated that financial outflows had been under-reported by some countries or that financial inflows had been over-reported by others.⁵

Both the Godeaux Report and the Esteva Report called attention to some key implications of these discrepancies for macroeconomic policy.⁶ The authors of these reports pointed out that substantial portions of income had not been accounted for in countries’ national accounts and, therefore, the severity of recessions may have been overstated. The authors also observed that the understated Current Account surpluses and overstated Current Account deficits were likely to prompt the adoption of inappropriate corrective measures at national levels. In addition, it was noted in these reports that the under-reported financial outflows and over-reported financial inflows provided misleading signals about the financing of resource transfers and, more generally, about saving behavior in a global setting.

To provide a basis for assessing the adequacy of data collection systems and any inconsistencies in the classification of international financial flows, the authors of the Godeaux Report looked closely at the underlying causes of the discrepancies in the global Financial Account and, in particular, at the apparent excess of liability flows existing at the world level from 1986 through 1989. This excess was consistent with the overall excess of debits reported in the global Current Account and identified by the authors of the Esteva Report. (More than one-third of the excess debits were attributable to investment income.)

FINDINGS OF THE WORKING PARTY ON THE MEASUREMENT OF INTERNATIONAL CAPITAL FLOWS (GODEAUXREPORT)

According to the Godeaux Report, an attempt was made to determine whether global discrepancies in financial flows were caused by the under-reporting of financial assets, the overstatement of liabilities, or both. It was observed that the global excess of liabilities was distributed through four major components: direct investment, portfolio investment, other investment, and reserve assets (see Table 1). More than US$66 billion was concentrated in other investment; opposite imbalances of US$16 billion and US$15 billion were present, respectively, in direct investment and reserve assets. Average discrepancies in portfolio investment flows were small but highly volatile; larger ones existed in subcomponents (bonds and equities)of portfolio investment. Factors that contributed to these discrepancies were

• the inability of some countries to record direct investment liabilities in the form of reinvested earnings

• the frequent classification of issues and redemptions of international bonds as long-term loans rather than as portfolio investments

• the absence of data, especially on portfolio assets, from offshore financial centers

• the non-uniform treatment of repurchase agreements (repos), which were treated, by some countries, as short-term collateralized loans and, by others, as outright sales/purchases of securities

• the fact that reserve assets held by monetary authorities were not matched by the measured liabilities of other countries

• the inadequate reporting of assets and liabilities of international organizations.

On the basis of information collected through questionnaires and many other sources, the authors of the Godeaux Report made a partially successful attempt to adjust for these inconsistencies.⁷ As a result, they were able to narrow global imbalances in financial asset and liability flows from US$40 billion to US$17 billion (see Table 1). The conclusion presented in the Godeaux Report was that the residual discrepancy was essentially attributable to portfolio investment. The “adjusted” excess of portfolio liabilities had increased, mainly because of growth in long-term bonds, to US$13 billion. To analyze this problem, the authors of the Godeaux Report examined end-1988 position data (that is, stock rather than flow data) on cross-border bonds.⁸ They drew attention to the large discrepancy (amounting to 25.3 percent of all liabilities identified) between assets (including monetary authorities’ holdings of nonresident securities) of US$1.2 trillion and liabilities of US$1.6 trillion. These findings were consistent with the undercoverage of portfolio assets of offshore financial centers and the widespread underreporting of portfolio assets. The latter was presumably caused by holders who changed to new intermediaries (some of which were located abroad). However, the evidence was not conclusive because stock data were partially derived from cumulated flows, and reported liabilities may have been overstated.⁹ To explore these issues further, the authors of the Godeaux Report also examined balance of payments data disaggregated by country. However, bilateral comparisons were inconclusive because financial flows had been allocated geographically on the basis of two different criteria—the debtor-creditor principle and the transactor principle.¹⁰ (For balance of payments statistics to be comparable from country to country, such statistics must be compiled on the basis of a single, consistent principle.)

The authors of the Godeaux Report were only partially successful in reconciling global discrepancies (especially within portfolio investment) in financial flows. Substantial discrepancies in net data were not reconciled, and it is likely that more existed in the underlying gross numbers. The Godeaux Report therefore presented the following recommendations:

• Countries should endeavor to implement improved coverage and to apply international balance of payments standards in the compilation of Financial Account statistics.

• Appropriate resources and legal powers should be allocated to statistical activities.

• Stock data should be collected on a regular basis.

• The major countries should undertake coordinated benchmark surveys of international portfolio assets and liabilities disaggregated by partner countries.

TREATMENT OF PORTFOLIO INVESTMENT IN THE FIFTH EDITION

While an analysis of discrepancies in the global Financial Account was being prepared for the Godeaux Report, the classification of Financial Account transactions was being reviewed. Thereafter, the coverage (described in the fourth edition of the Balance of Payments Manual) of financial flows and stocks was expanded, restructured, and described anew in the fifth edition. The new Financial Account presentation was accompanied by a broader definition, which included short-term instruments and financial derivatives, of portfolio investment and an integrated presentation of financial flows and stocks. (See boxed material on pages 16-17.) The integrated presentation required the classification of stocks in the IIP to be identical to the classification of transactions in the Financial Account. To link the two, reporting of additional details on price and exchange rate changes and other adjustments was also required. While useful in and of themselves, these additional details were considered necessary for ensuring the coherence of flow and stock data.

Some national compilers are still in the process of implementing the definitions and standards of the BPM5. One consequence of implementation is that global discrepancies in the Financial Account can be analyzed, for both flows and stocks, and assessments of inconsistencies and deficiencies in the underlying data can be made. This benefit was fully recognized in the Godeaux Report, which stated that a first step towards identifying and addressing deficiencies in flow data at national levels would be to establish benchmarks for stocks. The establishment of benchmarks was particularly important at the time the Godeaux Report was being prepared because few countries were compiling IIP statements that were fully reconcilable with the Financial Account. Other changes affecting portfolio investment were made to the BPM5 in 2000. These changes, which are discussed in a subsequent section, pertain to the treatment of financial derivatives in the Financial Account and have implications for future coordinated portfolio investment surveys.