Keywords: Balance Sheet Approach, global flow of funds, financial crisis, IMF surveillance, International Monetary Fund, Balance of Payments Division.

40 KB – 54 Pages

PAGE – 1 ============
Ninth IFC Conference on fiAre post -crisis statistical initiatives completed?fl Basel , 30-31 August 2018 The IMF balance sheet approach: towards from -whom -to-whom information on cross -border portfolio securities 1 Artak Harutyunyan and Carlos Sánchez Muñoz, International Monetary Fund 1 This paper was prepared for the meeting. The views expressed are those of the author s and do not necessarily reflect the views of the BIS , the IFC or the central banks and other institutions represented at the meeting .

PAGE – 2 ============
Balance Sheet Statistics: towards from -whom -to -whom information on cross -border portfolio se curities 1 The IMF Balance Sheet Approach : towards from -whom -to -whom information on cross -border portfolio securities Artak Harutyunyan and Carlos Sánchez Muñoz 1 Abstract This paper proposes an international exchang e of information on issuers of securities and the ir institutional sector . This exchange would contribute substantially to break ing down portfolio positions by country and sector of both holders and issuers and thus to uncovering the financial interconnections between sectors and countries at the global level. The paper first discuss es how the Balance Sheet Approach is constructed and how it contribute s to major activities carried out at the IMF . It then explains areas for future development towards the longer -term goal of compiling a Global Flow of Funds matrix . Such a matrix would require breaking down all financial positions by counterpart country and sector. To compile from -whom -to -whom information for portfolio securities , the paper describes an ongoing project of the IMF Statistics Department in coordination with the IMF Committee on Balance of Payments Statistics . The project consist s of setting up a centralized database to stor e information abou t securities issuers and their institutional sector. By putting this information at the disposal of the IMF Coordinated Portfolio Investment Survey reporters (more than 80 economies), the database would permit breaking down cross -border portfolio securities by country and sector of holders and issuers with a view to compiling from -whom -to -whom portfolio information. Keywords: Balance Sheet Approach , global flow of funds, financial crisis, IMF surveillance , from -whom -to -whom, portfolio investment, securities, Coordinated Portfolio Investment Survey JEL classification: -Term Capital Movements 1 International Monetary Fund, Balance of Payments Division. We are grateful to Thomas Alexander, Marco Espinosa and Giovanni Ugazio for useful comments on the paper. The views expressed in this paper are entirely our own and do not necessarily reflect those of the International Monetary Fund, its Executive Board, o r its management.

PAGE – 3 ============
2 The IMF Balance Sheet Approach : towards from -whom -to -whom information on cross -border portfolio securities Contents Introduc tion .. 3 1. The Balance Sheet Approach 3 1.1 What is the Balance Sheet Approach? Why is it important? . 3 1.2 How to compile and analyze the Balance Sheet Approach 4 1.3 Using the Balance Sheet Approach at the IMF 7 1.4 Areas for future development 9 2. A Balance Sheet Approach extension: breaking down portfolio investment positions by geography and sector . 9 2.1 Portfolio liabilities and the Coordinated Portfolio Investment Survey 9 2.2 Still missing: which foreign sectors are domestic investors financing? 11 2.3 Exchanging information internationally to close the gap .. 11 Conclusions . 13 Annex 1: Pilot project between the European Central Bank and the US Fed . 15 Characteristics of the exchange based on the pilot conclusions 15 References 17

PAGE – 4 ============
Balance Sheet Statistics: towards from -whom -to -whom information on cross -border portfolio se curities 3 Introduction Ever -increasing global financial inter linkages among different sectors and economies require that , besides monitor ing transactions , analysts and policy makers pay increasing attention to positions . In a globalized world with complex financial inter -linkages, draw ing a global financial network between different economies and sectors require s sufficiently detailed sectoral financial balance sheets. Sectoral b alance sh eets (financial assets and liabilities) broken down by counterpart country and sector may facilitate early detection of potential vulnerabilities and risk s of spillovers from individual sector s and countr ies to the rest of the global economy . Developing a global flow of funds Š expanding the traditional single -economy flow of funds Š has been identified by the IMF as a longer -term goal that can fundamentally change the IMF bilateral and multilateral surveillance work . Given the complexity of bui lding such a global flow of funds for both positions and transactions covering all sectors and instruments, a step -by -step approach is need ed. In the last years, t wo data collection frameworks managed by the IMF Š the Coordinated Direct Investment Survey ( CDIS) and the Coordinated Portfolio Investment Survey (CPIS) Š have helped compilers address bilateral asymmetries . This paper describes how an international exchange of information could substantially help CPIS reporters compile from -whom -to-whom information on cross -border portfolio securities . Section 1 describes the importance of the Balance Sheet Approach (BSA) ; how to compile it; how the IMF use s it for surveillance; and how it can be further develop ed. Section 2 focuses on how exchanging inf ormation internationally could help break down portfolio investment positions by counterpart country and sector , a step towards the longer -term goal of a Global Flow of Funds matrix . 1. The Balance Sheet Approach 1.1 What is the Balance Sheet Approach ? Why is it important ? The BSA pull s together the assets and liabilities of each sector within a country , i.e. its aggregate ( including cross -border ) position s/balance sheet. Such sectoral balance sheets convey important information on risks and vulnerabilities to policies and shocks , as well as on the inter linkages and exposures between the different sectors of an economy and vis -à-vis the rest of the world . Renewed interest on balance sheets arose from the 2007 -2008 Global Financial Crisis .2 An important trigger of the crisis was the in ability of some sectors in certain countries to service their excessive debt , which generated global spillover effects. Increasingly interconnected financial markets require close examin ation of an econom y™s sectoral balance sheets for complement ing the traditional flow -based analysis. 3 This analysis allows exploring the buildup of balance sheet interlinkages and how they make a sector or an economy vulnerable to shocks. Policy makers need to detect balance s heet risks and vulnerabilities early enough to be able to apply timely policy responses . Balance sheet interlinkages and networks also inform the analysis of 2 This is reflected in the IMF™s 2014 Triennial Surveillance Review, which advocates the development of a global flow of funds fito build on the national balance sheet approach to better analyze cross -border linkages.fl 3 Allen et al. (20 02)

PAGE – 5 ============
4 The IMF Balance Sheet Approach : towards from -whom -to -whom information on cross -border portfolio securities potential shock propagation s from one sector to another, permitting policy makers to take timely preventive measures. Analyzing balance sheet s allows exploring key macroprudential questions . Examples of such questions are: h ow healthy is the aggregate balance sheet of each sector and the external (international investment) pos ition of the economy as a whole; what the potential risks associated to balance sheet vulnerabilities over time are; how such vulnerabilities amplify and propagate the impact of external shocks ; and whether there are potential channels of shock propagation among different sectors of the economy . 1.2 How to compile and analy ze the Balance Sheet Approach Financial balance sheets can be compiled at two -dimensional and three -dimensional levels . 4 The two -dimensional financial balance sheets present, for each sector, financial assets and liabilities without identifying the counterpart sectors. 5 The three -dimensional balance sheets, in addition to the two -dimensional approach contain from -whom -to -whom information, where financial assets and liabilities are broken down by sector and by counterpart sector (Figure 1) , thus allowing to draw the financial network between different sectors of the economy and the rest of the world. The three -dimensional app roach is most suitable for the type of analysis discussed in Subsection 1.1, with interconnectedness being at the core of the analysis . The data requirements for the three -dimensional presentation are much more demanding though . Figure 2 below shows an int egrated system of financial balance sheets of key sectors and identifies possible data sources for each cell. Figure 1: Concept of three -dimensional balance sheets Source: MFSMCG Figure 8.2. Mirror data from counterpart sectors can be used to populate those cells of the matrix for which obtaining source data is difficult . For example, asset side loan data from the banking sector can be used to populate loan liabilities of the household sector instead of obtaining these data from households, resulting in cost efficiency and better quality of data. 4 See the IMF™s Monetary and Financial Statistics Manual and Compilation Guide (MFSMCG) 5 As in the balance sheets of the System of National Accounts 2008 (2008 SNA )

PAGE – 6 ============
Balance Sheet Statistics: towards from -whom -to -whom information on cross -border portfolio se curities 5 The analysis of from -whom -to-whom financial balance sheet s focus es on three intra -sector and inter -sectoral macroeconomic vulnerabilities : maturity, currency, and capital structure mismatches . To analyze such mi smatches, t hree key indicators are compiled and analyzed : net short -term position (short -term assets less short -term liabilities ); n et foreign currency position (foreign currency assets less foreign currency liabilities ); and n et financial position (financ ial assets less liabilities ). Maturity mismatches arise when assets are longer -term, mainly illiquid, while liabilities are short -term. Maturity mismatches can arise in both domestic and foreign currency, creating rollover risk, interest rate risk for the debtor, and reinvestment risk for the creditor. Currency mismatch es arise when assets and liabilities are denominated in different currencies . This mismatch creates exchange rate risk. For example, if assets are held in domestic currency but liabilities are denominated in foreign currency, substantial losses may result if the domestic currency depreciates sharply. Capital structure mismatch es result from excessive reliance on debt financing instead of equity. The absence of an equity buffer can lead to so lvency issues when a sector encounters a shock. Solvency or credit risk emerges when a sector™s financial assets no longer cover its financial liabilities. Solvency risk is closely linked to maturity mismatch risk, currency mismatch risk, and capital struc ture mismatch risk.

PAGE – 8 ============
Balance Sheet Statistics: towards from -whom -to -whom information on cross -border portfolio se curities 7 by counterpart se ctors . An even longer -term objective would be to extend the analysis to identify flows for regular monitoring of bilateral cross -border financial flows. 6 1.3 Using the Balance Sheet Approach at the IMF The Fund BSA is the analysis of financial balance sheets on a from -whom -to-whom basis . While Fund staff developed t he framework in the early 2000s , only recently has it been mainstream ed into Fund surveillance. The 2014 Triennial Surveillance Review (TSR) called for incorporat ing macro -financial analysis and the BSA into the regular Fund surveillance, with the Statistics Department of the IMF (STA) stepping up its role .7 STA support s Fund surveillance by preparing the relevant matrices and supporting analytical work . Following two rounds of Fund -wide pilots on mainstreaming macro -financial surveillance, where BSA was one of the themes, STA has developed a new automated BSA tool using in-house IMF data . The tool guides Fund u sers in the construction of the BSA matrix for about 120 countries. Certain analytical features are also embedded in the tool; for instance, following an exchange rate depreciation/appreciation shock of a designated magnitude the tool automatic ally calculat es balance sheet positions. As an example, the Selected Issues Paper for the 2016 Article IV staff report for Indonesia features the analytical potential of the BSA . Figure 3 presents the balance sheet interconnectedness network for Indonesia in 2007 and 2014 . The evolution of the Indonesia n BSA between the se two years suggests two areas of vulnerability : first , the increasing reliance of non -financial corporations (NFCs) on cross -border funding ; and second, the banking sector increasing exposure to NFCs . Such spillover risks demonstrat e the analy tical power of the BSA, especially when monitored for multiple periods. Figure 3. Indonesia : BSA matrix in network map form Source: Indonesia: Selected Issues Paper, 2016, Figure 1 on page 29, IMF. Note: The thickness of the arrow indicates the size of gross exposure, while the color of the nodes distinguishes net creditors (green) from net debtors (red). 6 Errico et al. (2014), for example, present an approach to understanding the shad ow banking system in the United States using a similar approach anchored on the analysis of global flow of funds. 7 The 2015 paper on Balance Sheet Analysis in Fund Surveillance takes detailed stock of the developments until 2015.

PAGE – 9 ============
8 The IMF Balance Sheet Approach : towards from -whom -to -whom information on cross -border portfolio securities Abbreviations: NBFI Œ Non -bank Financial Intermediaries; NFCs – Non-financial corporations ; HHs Œ Hou seholds; ROW Œ Rest of the World BSA -type analysis is also gaining traction for the Fund™s work on financial sector assessments . Several recent Financial Sector Assessment Programs (FSAP) 8 and Financial Sector Stability Reviews (FSSR) 9 have used BSA -type analysis in their assessments as a starting point for detecting major macroeconomic vulnerabilities that may directly or indirectly affect the financial system. The recently concluded Romania FSAP , for example, used the BSA to analyze macro -financial interlinkages, sectoral dependencies, and potential balance sheet vulnerabilities for all resident sectors (Figure 4). Figure 4. Romania: Network of balance sheet exposures (2016) Source: Romania: FSAP Technical Note on Balance Sheet Analysis , 2018, IMF. Note: Red nodes represent net borrowers and green nodes net lenders. The diameter of nodes and thickness of arrows show the relative size of imbalances and exposures, respectively. 8 The FSAP is a comprehensive and in -depth analysis of a country™s financial sector. FSAP assessments are the joint responsibility of the IMF and World Bank in developing economies and emerging markets and of the IMF alone in advanced economies. 9 FSSRs are a new IMF tec hnical assistance instrument providing a diagnostic upon which financial sector reform programs can be built and implemented. FSSRs provide baseline diagnostic assessments, highlight key weaknesses in financial systems and institutional capacities, and set out prioritized medium -term action plans for well -sequenced financial sector reforms, to be supported by follow -up TA from the IMF and other sources.

PAGE – 10 ============
Balance Sheet Statistics: towards from -whom -to -whom information on cross -border portfolio se curities 9 Abbreviations: OFIs Œ Other Financial Intermediaries; MMFs Œ Money Ma rkey Funds; ROW: Rest of the World 1.4 Areas for future development Mainstreaming BSA -type analysis has become possible because of the increased availability of underlying balance sheet data thanks to the IMF ™s and member countries ™ efforts . Yet , important data gaps remain. STA continues to work on the future improvements of balance sheet statistics th rough the Data Gaps initiative for G -20 countries 10. STA is preparing a medium – to long -term strategy to improve the avai lability of balance sheets for the financial, external, fiscal, and real sectors, with a view to better support ing BSA and its use for surveillance. Capacity development in the field of BSA is also supported by one of the statistical modules of the Financial Sector Stability Fund 11. As mentioned previously, o ne of the major BSA future developments is breaking down global cross -border positions by counterpart countries and sector s. The lack of information on portfolio investment assets by issuer sectors is an impediment to continue advancing this work . This is further discussed in the next section .12 2. A Balance Sheet Approach extension: b reaking down portfolio investment positions by geography and sector 2.1 Portfolio liabilities and t he Coordinated Portfolio Investment Survey Compiling accurate statistics on portfolio investment liabilities broken down by country is a well -known statistical challenge . Domestic issuers of securities may be aware of who first acquire s such securities in primary markets , but they most often cannot trace subsequent purchases and sales , nor consequently determine the residence of the final holder . When investors operate via foreign financial intermediaries , it is extremely hard for statistical compilers to find information sources that can determine who ultimately holds domestic securities .13 This could only be possib le by enquiring /survey ing final investors . However , reporting requirements typically only address resident reporters , so foreign investor s cannot be legally obliged to provide such statistical information . In the late nineties, the IMF started to conduct the CPIS 14 with a view to mitigat ing the increasing size of global portfolio investment asymmetries . Such sizeable asymmetries had 10 See latest Progress Report of the Second Phase of the G -20 Data Gaps Initiative (DGI -2) under . Under recommendation II.8 , the DGI -2 encourag es advanced economies to disseminate data on a from -whom -to-whom basis. 11 The Financial Sector Stability Fund is an IMF trust Fund that provides technical assistance and training to low – and lower -middle -income countries in support of financial sector stability, inclusion, and deepening. See ish/pdf/FSSF.pdf 12 Other aspects for BSA future development include reconciliation of asymmetries between different underlying datasets ; sectoral financial accounts with from -whom -to-whom counterpart detail ; maturity breakdowns for all financial instrumen ts; further breaking down the nonbank financial sector ; and compiling matrices of financial flows corresponding to balance sheet positions . 13 This is particularly true because most investors transact and keep securities in custody with financial intermedi aries , often through a long chain of custodians and sub -custodians. The frequently long chains of custodians/sub -custodians/final investors usually entail cross -border relationships involving several countries. 14

PAGE – 11 ============
10 The IMF Balance Sheet Approach : towards from -whom -to -whom information on cross -border portfolio securities result ed from the increased liberalization of cross -border transactions and the above -described measurement issues . The CPIS is a global survey of portfolio investment stocks collect ing semi -annual (end -June and end -December) information on cross border holdings of debt and equity securities. The CPIS provides each economy with mirror statistics on its portfolio investment liabilities , coming from what every other econom y report s to be holding of the securities issued by residents in that country . Compiling detailed information about cross -border portfolio investment assets (domestic holdings of foreign securities) is substantially more straightforward than compiling similar information about liabilities (domestic securities held abroad). Therefore, with the mirror information provided by the CPIS , each economy can compile the geographic distribution of its nonresident creditors . In this way, t he information provided by the CPIS permits that cross -border position statistics portray complex financial interlinkages between different economies as in figure 5: Figure 5: CPIS networking analysis Source: End -June 2017 CPIS data Size of circles is prop ortional to total cross -border issues/ holdings and size and direction of arrow s, to the scale and direction of the financing More than 80 economies participate in the CPIS, including all major industrialized economies as well as most offshore financial centers and emerging markets (see figure 6). 15 Figure 6: CPIS participating economies 15 The list of CPIS reporters can be consulted here: -4E58 -467E-9B90 -9DE0C3367363&sId=1481580274211

40 KB – 54 Pages