Abstract

Recent improvements in statistical methodologies and data availability are enhancing the potential for detecting and monitoring macroeconomic balance sheet vulnerabilities. In particular, some of the datasets introduced in recent years permit a much more frequent, detailed, and up-to-date analysis.

Recent improvements in statistical methodologies and data availability are enhancing the potential for detecting and monitoring macroeconomic balance sheet vulnerabilities. In particular, some of the datasets introduced in recent years permit a much more frequent, detailed, and up-to-date analysis.

These databases are compiled according to particular statistical methodologies, which themselves are evolving partly due to the requirements of greater stock-based analysis. This chapter discusses these methodologies and datasets, and illustrates their usefulness in terms of meeting the data requirements for the BSA.

Relationship Between the BSA and 1993 SNA Methodologies and Datasets

The 1993 SNA is the internationally agreed-upon integrated set of production, income, accumulation, and financial accounts, balance sheets, and supporting tables that describe all economic flows and stocks of assets and liabilities in an economy, with full reconciliation of flows and stocks. As such, the BSA framework is a component of the 1993 SNA and is grounded in its methodology for defining transactions, institutions, sectors of the economy, classifications of assets and liabilities, and accounting rules. In addition, the 1993 SNA provides the framework and methodology for the main sectors of an economy. Specific methodologies for these main sectors and their databases have drawn on the 1993 SNA, but have adopted definitions of sectors and classifications of assets and liabilities that may differ in some respects (Box 2). The BSA can draw on the many sectoral methodologies based on the 1993 SNA.

The 1993 SNA sequence of accounts applies in principle to any institution or sector. If the BSA is narrowed to examine the vulnerabilities of a particular sector known to be problematic—for example, the financial sector and its potential to trigger a macroeconomic crisis—then the balance sheet for that sector provides the framework for the BSA. Even in the case of applying the BSA to one sector, balance sheets for other sectors can be useful for cross-checking or filling in data missing in the balance sheet of the sector under examination.

Relevant Data Methodologies

  • Monetary and Financial Statistics Manual (MFSM)

  • Standardized report forms (SRFs)

  • Compilation Guide on Financial Soundness Indicators

  • Fifth Edition of the Balance of Payments Manual (BPM5)

  • External Debt Statistics: Guide for Compilers and Users

  • International Investment Position: A Guide to Data Sources (IIP)

  • Coordinated Portfolio Investment Survey, Second Edition (CPIS)

  • International Reserves and Foreign Currency Liquidity: Guidelines for a Data Template

  • Government Finance Statistics Manual 2001 (GFSM 2001).

Potential Databases for the BSA

Databases based on methodologies relevant for the BSA are potential sources of data for its application. The BSA can be applied to an individual country or for cross-country analysis of vulnerability using information from statistical databases for the 1993 SNA and its major component systems. These include monetary and financial statistics, in particular, and the SRF data, balance of payments, IIP, QEDS, CPIS, and government finance statistics. Nearly all entries in the 7 x 7 intersectoral framework for the BSA can be filled using data from the SRFs, IIP, QEDS, and CPIS (Table 4).

Financial Sector

The MFSM provides the guidelines on statistical methodology presenting monetary and financial statistics. The methodology set out in MFSM is harmonized with the 1993 SNA, but does not prescribe the detail on currency and maturity required for the BSA.

The introduction in 2005 of the SRFs for monetary and financial sector data fills an important gap in data coverage for the BSA. The SRFs are based on sectoral balance sheets for the central bank (report form 1SR), other depository corporations (report form 2SR), and other financial corporations (report form 4SR), as defined in the MFSM. They provide the required breakdown by domestic and foreign currency as well as information on the maturity structure, sometimes indirectly, 15 for both domestic and external assets and liabilities, as well as the required decomposition by domestic sectors. For countries submitting SRFs, the BSA template can be populated with a high level of detail to provide an up-to-date analysis comparable across countries.

Table 4.

Potential Data Sources for Estimating Intersectoral Asset and Liability Matrix

article image
Notes: SRF = standardized report forms; IIP = international investment position; QEDS = quarterly external debt statistics; and JEDH = joint external debt hub.

This data gap can in the future be filled with data from the public debt data template (which also covers assets), which currently is being piloted in some countries.

Coordinated portfolio investment survey (CPIS) data are not available with the sufficient sectorization.

The new SRF data can provide the information needed to fill in a majority of entries in the 7 x 7 intersectoral framework for the BSA (Table 5). For entries where the assets and liabilities overlap for the central bank, other depository corporations, and other financial corporations, the assets reported by sector should match the corresponding liabilities reported by the other. This is not always the case and the analyst has to decide which information is more accurate (Gulde and others, 2003). Generally, data reported by the central bank are taken to be more reliable than that reported by other depository corporations, and by these two sectors more reliable than by other financial corporations.

Table 5.

Uses of Standardized Report Form (SRF) Data to Estimate Intersectoral Asset and Liability Positions

article image

Given that the SRF data are standardized across countries, the method of estimating intersectoral relationships based on SRF data can be replicated for other countries. The mapping of SRF variables into the BSA framework can be followed for all countries.16 For remaining intersectoral relationships, other data sources, such as QEDS and CPIS, can be used.

The SRF submission for South Africa illustrates the usefulness of these new data for conducting up-to-date monthly analysis of balance sheet vulnerabilities. The monthly SRF data have been combined with data from the QEDS and CPIS in the BSA framework to estimate detailed intersectoral positions, by financial instrument and currency (Table 2). The framework also allows for a breakdown by claims and liabilities, which can be very useful when analyzing net financial positions.

Clearly one of the greatest advantages of this approach is that detailed monthly intersectoral positions can be estimated by financial instrument and by currency, permitting a detailed analysis of changes in macroeconomic vulnerability in an integrated framework over time (Figure 4). The sectoral position can also be investigated vis-à-vis a particular sector (Figure 5). Once a particular vulnerability is identified, any change can be analyzed in detail, including by currency, type of claim, and financial instrument (Figure 6).

The compilation of financial sector indicators supports the BSA. Based on the Compilation Guide on Financial Soundness Indicators (IMF, 2004c), 62 countries are making a concentrated and coordinated effort to compile financial sector indicators and publish results by the of end 2006. The financial sector indicator data, particularly data for the key nonfinancial sectors covered, will usefully support and complement BSA applications. In particular, the cross-border consolidated data underlying the financial sector indicators cover complex banking systems with significant foreign subsidiary and branch networks that may not be adequately covered in the BSA framework.

External Sector

The balance of payments accounts and IIP and QEDS data are closely linked to the 1993 SNA. This linkage is reinforced by the fact that, in almost all countries, balance of payments, external debt, and IIP data are first compiled and subsequently incorporated into national accounts. Although the Fifth Edition of the Balance of Payments Manual (known as BPM5) (IMF, 1993b) does not explicitly call for a currency breakdown, this is not necessarily a serious problem for assets, as for nearly all countries the vast majority of external assets are denominated in foreign currency. IIP and QEDS data present a short- and long-term maturity breakdown on an original maturity basis consistent with the 1993 SNA.

The introduction in 2004 of the online QEDS dataset, based on the External Debt Statistics: Guide for Compilers and Users (IMF, 2003), provides information on external liabilities with breakdowns by currency and maturity that can be used in the BSA framework. It is maintained by the World Bank and updated within one month after the end of each quarter. Breakdowns include short- and long-term maturity of debt based on original maturity, and financial instruments (currency, deposits, money market instruments, bonds and notes, loans, trade credits, other debt liabilities). QEDS also includes information on a remaining maturity basis. The online dataset brings together in a central location detailed quarterly external debt data from 55 of the 62 countries currently subscribing to the SDDS.17 It facilitates both time-series analysis and cross-country data comparisons.

The joint external debt hub brings together external debt data for about 175 countries that are available from the BIS, IMF, OECD, and World Bank, including national external debt data for most SDDS subscribers. Data on selected external debt components (long- and short-term maturities), including bank loans, official bilateral loans, debt securities issued abroad and nonbank trade credits, are disseminated on a quarterly basis. The database complements external debt statistics based on national sources, filling important coverage gaps, particularly in the area of private sector external liabilities.

The IMF’s Data Template on International Reserves and Foreign Currency Liquidity provides a consistent framework for assessing a country’s official foreign currency liquidity position on a comprehensive and timely basis. It facilitates the disclosure of information on international reserve assets together with information on potential short-term foreign currency obligations (and claims) that affect the analysis of international reserve assets, including off-balance-sheet activities (such as those arising from forwards, futures, and other financial derivatives operations). The institutional coverage applies to monetary authorities and the central government, and foreign currency flows are related to both residents and nonresidents.

Figure 4.
Figure 4.

South Africa: Sectoral Net Financial Positions, by Currency

(In percent of GDP, December 2003–November 2005)

Sources: Standardized report forms; joint external debt hub; coordinated portfolio investment survey; and quarterly external debt statistics.Note: Sectoral net financial positions represented in figure are total financial assets minus total financial liabilities.
Figure 5.
Figure 5.

South Africa: Sectoral Net Financial Positions vis-à-vis Nonresidents, by Currency

(In percent of GDP, December 2003–November 2005)

Sources: Standardized report forms; joint external debt hub; coordinated portfolio investment survey; and quarterly external debt statistics.Note: Sectoral net financial positions represented in figure are total financial assets minus total financial liabilities.
Figure 6.
Figure 6.

South Africa: Other Depository Corporations’ Detailed Positions vis-à-vis Nonresidents by Currency and Instrument

(In percent of GDP, December 2003–November 2005)

Source: Standardized report forms.1Holdings of foreign currency are difficult to discern in the figure because they did not exceed 0.05 percent in any month during the period covered.

The IIP has been a useful data source for the BSA. The IIP presents data on a country’s external financial position, with the primary focus on the stock of financial assets and liabilities. Data items include financial claims on and liabilities to nonresidents, equity assets and liabilities, financial derivative instruments, monetary gold, and special drawing rights (SDRs). The liability component of the IIP data is closely related to QEDS.18

The CPIS can complement the datasets above by providing survey data on cross-border holdings of securities (equities, long- and short-term debt) by counterpart jurisdiction of issuer. The CPIS is an annual survey of portfolio investment assets for 71 countries based on a methodology drawn from the BPM5. The CPIS has been undertaken on an annual basis since 2001, but data are also available for the 1997 CPIS. The CPIS collects comprehensive information on the stock of cross-border holdings of equities and short- and long-term debt securities valued at market prices and broken down by the economy of residence of the issuer. This global database includes data on reported cross-border holdings of securities and derived portfolio investment liabilities with the capacity for showing bilateral and partner economy data from the creditor or debtor perspective. The CPIS is a useful data source for estimating intersectoral asset and liability positions with nonresidents both directly and through derived counterparty country information. It contains some information on the sector of holder and currency of issue, but lacks the necessary breakdown on sectoral liabilities to nonresidents. The data are available with a lag of one year or more.

Public Sector

Introduction of the Government Finance Statistics Manual (known as GFSM 2001) (IMF, 2001b) represented a significant step toward the presentation of general government statistics in a manner consistent with the BSA. An innovation of the GFSM 2001 was the integration of a balance sheet in the framework for public sector statistics. As prescribed by the 1993 SNA, this balance sheet integrates transactions and other economic flows with stocks of assets and liabilities. It is similar to balance sheets for other sectors, thereby facilitating intersectoral comparisons.

Data Availability

Data availability for a high frequency and up-to-date country balance sheet approach is improving. Currently, more than 40 countries, including most emerging market countries, have the required data coverage for the detailed BSA framework presented in this paper (Table 6). Clearly, the main improvement is the recently introduced SRFs for monetary and financial sector data, which provide the vast majority of the required inter-sectoral balance. Moreover, the key advantage of these datasets—which so far encompass 58 countries—is that they are compiled monthly and with a high level of detail standardized across countries. The remaining gaps on government and nonfinancial corporations’ liabilities to nonresidents can be closed by the online QEDS introduced in 2004, which is available for 55 countries. IIP data—currently available for more than 100 countries—can be used to fill the remaining gaps on sectoral positions vis-à-vis nonresidents. In cases where IIP data are not available, JEDH data can fill in some of the gaps, particularly for nonfinancial domestic sector liabilities to nonresidents, and the CPIS provides information on domestic sector claims on nonresidents, albeit with a substantial lag and on an annual frequency. However, government liabilities to the nonfinancial domestic sectors are generally not readily available, nor are government claims on the nonfinancial domestic sectors or nonfinancial domestic sector holdings of claims on government, although the latter two gaps generally are considered to be minor.19

Data Reliability

Balance sheet analysis should ideally be based on comprehensive and consistent financial statistics appropriately delineated by sector and financial instruments. However, two types of data deficiencies typically prevent a complete sectoral analysis: lack of appropriate data and multiple (or overlapping) data for a particular financial instrument, either intrasectoral or intersectoral. To minimize discrepancies and determine the extent to which any remaining data deficiencies might undermine the results of the analysis, data reliability can be assessed by sector and financial instrument.20

Table 6.

Available Datasets for Balance Sheet Vulnerability Analysis

(As of November 2006)

article image
article image
Notes: *Standardized report form test data are being reviewed by IMF staff for quality. **Approval of standardized report form test data was expected to be completed by November 2006. EAP = Euro Area Presentation of the standardized report forms. An x for the international investment position simply indicates that those data are provided to the IMF Statistics Department. No distinction is made for completeness of reporting.

Data reliability can vary significantly by sector (Table 7). In general, central bank data are most reliable, followed by data from commercial banks and other financial corporations, international investment position data, and government debt data. Secondary trading in government debt can substantially affect the ability to determine sectoral holdings of government securities. Data on households and nonfinancial corporations are typically very scarce in emerging markets and in many cases are nonexistent. In these circumstances, two basic techniques—counterpart data collection and residual data collection—can be used to obtain data. As CPIS data are allocated by type of security and country of issuer, they represent a useful source for deriving counterpart data on all sectors, particularly households and nonfinancial corporations.

Sectoral data reliability can also vary by methodology. In general, the most reliable data are those that follow the MFSM (financial corporations), BPM5 (balance of payments data), International Investment Position: A Guide to Data Sources (IMF, 2002), and External Debt Statistics: Guide for Compilers and Users (such as QEDS) (IMF, 2003). Data on nonfinancial corporations’ positions vis-à-vis household and nonprofit organizations are generally less reliable. The uncertainly of these data is exacerbated if derived on a residual basis.

Data reliability also varies by financial instrument (Table 8). In general, the most reliable data are currency and deposits, loans, and securities (which together comprise the majority of SRF data). Also very reliable are external debt data on specific financial instruments, which can be obtained from both national sources (such as QEDS) and market and creditor sources (such as JEDH external debt and BIS international banking statistics). Estimates of trade credits and many types of government financial assets are often judged to be less reliable, but source data are still available on a sample basis or with a frequency that is less than quarterly or annually. The least reliable estimates are usually for miscellaneous assets and liabilities, which are commonly derived on a residual basis.

Aggregating sectoral data sets to undertake a balance sheet analysis of intersectoral relationships poses special challenges. As noted above, sometimes estimates for a particular subsector (e.g., households) or a group of financial instruments (e.g., miscellaneous assets/liabilities) have been derived using a residual calculation (this subsector or category of financial instrument is often referred to as a balancing item). These estimates therefore might include substantial discrepancies resulting from imprecise (or missing) data, which, when aggregated, could be magnified.

Table 7.

Data Reliability by Sector

article image
High Middle Low
Table 8.

Data Reliability by Financial Instrument

article image
High Middle LowSource: IMF, Compilation Guide for Monetary and Financial Statistics (2006, Chapter 8).Note: The darker areas indicate where the compiler placed a relatively “high” degree of reliability. The moderately shaded areas indicate series where estimates are judged to be less reliable, but where source data are still available on a sample basis or on a basis where the frequency is less than quarterly or annually. The lightly shaded areas are for series where there is virtually no source data; estimates for series in the nonshaded areas are based largely on residual calculation.

Caution is therefore required when handling economy-wide datasets, as there is a significant risk that unreliable estimates might undermine the results of the balance sheet analysis. The sectoral discrepancies hidden in the balancing item contain potentially valuable information on the size of the statistical error. The balance sheet analysis should therefore acknowledge these weaknesses and, if judged to be substantial, focus on sectoral relationships that are less affected by imprecise data or stress the caveats for using the data.

15

The maturity structure can be derived by defining financial assets that are not included in broad money as long term. However, this classification might be inappropriate in a particular country; in those cases the SRF data should be complemented by, for example, information on the maturity structure of government securities.

16

This mapping assigns the SRF variable codes standardized across countries to their appropriate cells in the 7 x 7 intersectoral framework for the BSA.

17

The availability of QEDS data is expected to expand in the near future. The number of potential countries covered by the database increases with the number of SDDS subscribers. Also, coverage is expected to improve, as SDDS countries are increasing the number of tables of the QEDS for which data are provided, with an emphasis on currency and maturity breakdown. Finally, some non-SDDS countries are expected in the near future to be able to prepare, at least, the SDDS-prescribed external debt data category.

18

The IIP includes some nonfinancial assets whose ownership is construed by convention as ownership of financial assets, owing to its definition as a financial claim of a nonresident on a resident entity that is considered the owner of the asset, as for example in the case of ownership of immovable assets such as land (IMF, 1993, paragraph 316).

19

These gaps are expected to be closed by the public debt template, which covers detailed sectoral claims of and liabilities to government.

20

The data quality assessment framework (DQAF) for external debt statistics issued by the IMF’s Statistics Department in June 2005 provides a useful tool to assess the quality of external debt statistics. The DQAF follows a comprehensive view of quality, which examines quality-related features of governance of statistical institutions, core statistical processes, and statistical outputs, and is intended to be applicable to any country.

Framework and Data Sources and Availability
  • View in gallery

    South Africa: Sectoral Net Financial Positions, by Currency

    (In percent of GDP, December 2003–November 2005)

  • View in gallery

    South Africa: Sectoral Net Financial Positions vis-à-vis Nonresidents, by Currency

    (In percent of GDP, December 2003–November 2005)

  • View in gallery

    South Africa: Other Depository Corporations’ Detailed Positions vis-à-vis Nonresidents by Currency and Instrument

    (In percent of GDP, December 2003–November 2005)

  • Allen, Mark, Christoph Rosenberg, Christian Keller, Brad Setser, and Nouriel Roubini, 2002, “A Balance Sheet Approach to Financial Crisis,” IMF Working Paper 02/210 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Billmeier, Andreas, and Johan Mathisen, 2006, “Balance Sheet Risks in a Dollarized Economy—The Case of Georgia,” IMF Working Paper 06/173 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Bussière, Matthieu, and Christian Mulder, 1999, “External Vulnerability in Emerging Market Economies: How High Liquidity Can Offset Weak Fundamentals and the Effects of Contagion,” IMF Working Paper 99/88 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Calvo, Guillermo, and Carmen M. Reinhart, 2002, “Fear of Floating,” Quarterly Journal of Economics, Vol. 117, No. 2 (May), pp. 379408.

    • Search Google Scholar
    • Export Citation
  • Gapen, Michel, Dale Gray, Cheng-Hoon Lim, and Yingbin Xiao, 2004, “A Contingent Claims Approach to Corporate Vulnerability Analysis: Estimating Default Risk and Economy Wide Risk Transfer,” IMF Working Paper 04/155 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Goldstein, Morris, and Philip Turner, 2004, Controlling Currency Mismatches in Emerging Markets (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation
  • Gulde, Anne-Marie, David Hoelscher, Alain Ize, Alfredo Leone, and David Marston, 2003, “Dealing with Banking Crises in Dollarized Economies,” in Managing Financial Crises: Recent Experience and Lessons from Latin America, ed. by Charles Collyns and Russell Kincaid, IMF Occasional Paper No. 217 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF), 1993a, System of National Accounts 1993 (Washington: International Monetary Fund).

  • International Monetary Fund (IMF), 1993b, Fifth Edition of the Balance of Payments Manual (Washington: International Monetary Fund).

  • International Monetary Fund (IMF), 2000, Monetary and Financial Statistics Manual (Washington: International Monetary Fund).

  • International Monetary Fund (IMF), 2001a, International Financial Statistics (Washington: International Monetary Fund).

  • International Monetary Fund (IMF), 2001b, Government Finance Statistics Manual (Washington: International Monetary Fund). Available via the Internet: http://www.imf.org/external/pubs/ft/gfs/manual/index.htm.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF), 2002, International Investment Position: A Guide to Data Sources (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF), 2003, External Debt Statistics: Guide for Compilers and Users (Washington: International Monetary Fund). Available via the Internet: http://www.imf.org/external/pubs/ft/eds/Eng/Guide/index.htm.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF), 2004a, “Biennial Review of the Implementation of the Fund’s Surveillance and of the 1977 Surveillance Decision—Overview, Modalities of Surveillance, Content of Surveillance, and Public Information Notice on the Executive Board Discussion” (August 24). Available via the Internet: www.imf.org/external/np/pdr/surv/2004/082404.htm.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF), 2004b, “Liquidity Management” (unpublished; Washington: International Monetary Fund).

  • International Monetary Fund (IMF), 2004c, Compilation Guide on Financial Soundness Indicators (Washington: International Monetary Fund). Available via the Internet: https://www.imf.org/external/np/sta/fsi/eng/2004/guide/index.htm.

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF), 2005, Global Financial Stability Report, April 2005 (Washington: International Monetary Fund).

  • International Monetary Fund (IMF), 2006, “Monetary and Financial Statistics: Compilation Guide—Pre-publication Version” (Washington: International Monetary Fund). Available via the Internet: http://www.imf.org/external/pubs/ft/cgmfs/eng/index.htm.

    • Search Google Scholar
    • Export Citation
  • Johnston, R. Barry, Jingquing Chai, and Liliana Schumacher, 2000, “Assessing Financial System Vulnerabilities,” IMF Working Paper 00/76 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Lima, Juan Manuel, Enrique Montes, Carlos Varela, and Johannes Wiegand, 2006, “Sectoral Balance Sheet Mismatches and Macroeconomic Vulnerabilities in Colombia, 1996–2003,” IMF Working Paper 06/5 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Mathisen, Johan, and Mariana Torres, 2005, “Balance Sheet Currency Mismatch and Liquidity Analysis,” Belize—Selected Issues, IMF Country Report 05/353 (Washington: International Monetary Fund), pp. 1626.

    • Search Google Scholar
    • Export Citation
  • Reinhart, Carmen M., Kenneth S. Rogoff, and Miguel A. Savastano, 2003a, “Debt Intolerance,” in Brookings Papers on Economic Activity, No. 1, ed. by William Brainard and George Perry (Washington: Brookings Institution), pp. 174.

    • Search Google Scholar
    • Export Citation
  • Reinhart, Carmen M., Kenneth S. Rogoff, and Miguel A. Savastano, 2003b, “Addicted to Dollars,” NBER Working Paper No. 10015 (Cambridge, Massachusetts: National Bureau of Economic Research).

    • Search Google Scholar
    • Export Citation
  • Rosenberg, Christoph, Ioannis Halikas, Brett House, Christian Keller, Jens Nystedt, Alexander Pitt, and Brad Setser, 2005, Debt-Related Vulnerabilities and Financial Crises: An Application of the Balance Sheet Approach to Emerging Market Countries, IMF Occasional Paper No. 240 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Roubini, Nouriel, and Brad Setser, 2004, Bailouts and Bailins? Responding to Financial Crisis in Emerging Economies (Washington: Institute for International Economics).

    • Search Google Scholar
    • Export Citation