Financial Stability Analysis: What are the Data Needs?

Contributor Notes

Author’s E-Mail Address:;

The growing incidences of financial crises and their damage to the economy has led policy makers to sharpen the focus on financial stability analysis (FSA), crisis prevention and management over the past 10–15 years. The statistical world has reacted with a number of initiatives, but does more need to be done? Taking a holistic view, based on a review of experiences of policy makers and analysts, this paper identifies common international threads in the data needed for FSA and suggests ways to address these. While there has been an encouragingly constructive response by statisticians, not least through the G-20 Data Gaps Initiative, more work is needed, including with regard to shadow banking, capital flows, corporate borrowing, and granular data. Further, to support FSA, the paper identifies potential enhancements to the conceptual advice in statistical manuals including with regard to foreign currency and remaining maturity.


The growing incidences of financial crises and their damage to the economy has led policy makers to sharpen the focus on financial stability analysis (FSA), crisis prevention and management over the past 10–15 years. The statistical world has reacted with a number of initiatives, but does more need to be done? Taking a holistic view, based on a review of experiences of policy makers and analysts, this paper identifies common international threads in the data needed for FSA and suggests ways to address these. While there has been an encouragingly constructive response by statisticians, not least through the G-20 Data Gaps Initiative, more work is needed, including with regard to shadow banking, capital flows, corporate borrowing, and granular data. Further, to support FSA, the paper identifies potential enhancements to the conceptual advice in statistical manuals including with regard to foreign currency and remaining maturity.

I. Introduction

The first chapter of Charles Kindleberger’s 1978 seminal work on financial crises “Manias, Panics and Crashes” is entitled “Financial Crisis: A Hardy Perennial.” But the chapter starts by pointing out the relative lack of such crises, particularly in advanced economies, during the several decades after World War II, the period when the core economic and financial statistical manuals of national accounts and balance of payments were developed to help support macroeconomic policy making.

Recent decades have witnessed the return of significant financial crises, notably, but not only, the global financial crisis (GFC) of 2007/8 that have resulted in significant losses to the real economy2 including years of under-performance in economic growth. The growing incidences of financial crises and their damage to the economy and the well-being of the population, together with the increased scale and interconnectedness of financial transactions, and their complexity, has led policy makers to give a greater focus to financial stability analysis (FSA),3 financial system resilience, crisis prevention, and management over the past 10–15 years.

Consequential to the focus on financial stability, the desired composition of the policy makers’ tool boxes and the nature of the data needed to support policy has changed. Indeed, the greater policy focus on financial stability has resulted in a global regulatory reform agenda endorsed by the Group of 20 (G-20) leaders and a significant demand for financial and economic data to support the monitoring of the risks and vulnerabilities in the system. Even though not the cause of the crisis, a lack of data hampered such monitoring in advance of the GFC. While the statistical world has subsequently reacted with a number of initiatives, including the G-20 Data Gaps Initiative (DGI) and others,4 the inevitable question arises as to what more needs to be done.

Against this background, the paper has two main aims. First, based on the experience of policy makers and analysts, the paper takes a holistic view of data actually used for FSA, drawing out the common international threads in analysis. To our knowledge this is the first paper to take such an approach at the international level. Second, the paper sets out the data gaps identified by policy makers that remain to be filled particularly with regard to shadow banking, capital flows, corporate borrowing, as well as a demand for more granular data, and more broadly how official statisticians can adapt their conceptual advice to better meet the needs of FSA including with regard to remaining maturity and foreign currency data.

Inevitably, before embarking on any data enhancements statisticians would need to address the challenges, costs, and trade-offs of implementation, as well as discuss priorities with users. This paper does not specifically address these issues as it is mainly focused on setting out user needs and exploring how they could be addressed. Further, the paper recognizes that users’ data needs and the priorities attached to these needs will differ based on the country circumstances.

The paper starts with a discussion of what is financial stability and how it is analyzed, identifies the datasets typically used in FSA, addresses the data gaps that have emerged, and sets out proposals for how economic and financial (macroeconomic) statistical manuals can better meet financial stability data needs without undermining their conceptual framework.5

II. Financial Stability Policy and Analysis

The August 2016 Bank for International Settlements (BIS)/Financial Stability Board (FSB)/International Monetary Fund (IMF) report to the G-20 (G-20 report) stated that macro-prudential policy is aimed at avoiding “the risk of widespread disruption to the provision of financial services that is caused by an impairment of all or parts of the financial system, and which can cause serious negative consequences for the real economy.”6

While the above definition and observation might be considered as applying to financial stability policy more broadly, the latter appears to have a wider remit than macro-prudential policy. As the then Head of Financial Stability at the Reserve Bank of Australia observed in 2013, macro-prudential policy is subsumed in the broader financial stability policy framework—prudential supervision, market conduct regulation, consumer protection, land supply, tax system, and exchange rate regime.7 To this list could be added the functioning of market infrastructure such as clearing houses, and corporate governance and investor protection.8

Indeed, beyond avoiding financial crises, it would appear that financial stability policy is concerned not only with the risk of widespread disruption to the provision of financial services, but with the efficiency of those services on an on-going basis, helping to identify where policy actions might improve efficiency and reduce systemic risk. For instance, the Bank of Korea considers that financial stability is “a condition in which the financial system works smoothly with all of its key components satisfactorily performing their roles: financial institutions carrying out their financial intermediary functions, market participants maintaining a high level of confidence in their financial market, and the financial infrastructure being well developed.”9 Indeed, while financial stability does not have a universally accepted definition, there seems to be a broad consensus that financial stability refers to the smooth functioning of the key elements that make up the financial system.10 11

What has developed, particularly since the GFC, has been a greater focus on strengthening financial stability policy and analysis, with consequential data demands. Further, new governance arrangements have been established bringing together macro-prudential analysis, micro prudential analysis, and other aspects of FSA, in a holistic approach to FSA.

Relationship between micro-prudential, macro-prudential and macro-economic policy and analysis

Micro-prudential supervision12 and market conduct regulation, etc., existed well before the GFC and as indeed, in some economies, did macro-prudential policy. But the GFC demonstrated that micro-prudential policy is necessary but not sufficient to ensure financial stability; hence the emergence of macro-prudential policy to complement micro-prudential policy with a more systemic perspective. As a former First Deputy Managing Director of the IMF observed, “macro-prudential analysis looks at the intersection of the real economy and the financial sector, providing a bird’s eye view of the entire system instead of focusing on individual instruments and individual institutions.”13

Macroeconomic developments and policy are directly relevant for financial stability policy as economic developments have an impact on the financial stability risks facing an economy, and vice versa. For instance, a rapidly growing economy might encourage excessive credit growth, while an economy with a weak economic position might have increasing levels of nonperforming loans. Therefore, traditional macroeconomic indicators are relevant for FSA. Nonetheless, as the August 2016 BIS/FB/IMF G-20 report observes, the difference between macroeconomic and macro-prudential policy is that “rather than managing the level and composition of aggregate demand or the business cycle, macro-prudential policy aims to strengthen the financial system’s defenses in the face of economic and financial shocks.”

The holistic approach of FSA helps “straddle the gap” between micro and macro analysis, as it became increasingly obvious that micro-prudential analysts need macro data and macro-prudential analysts need granular micro data, and they potentially benefit from each other’s insights in order to identify emerging systemic risks and vulnerabilities. As the IMF Financial Surveillance Strategy published in 201214 emphasized: there is a need to understand “the interactions between macro-prudential, macroeconomic, and micro-prudential policies, as well as potential costs and side effects.” In a similar vein, there have been calls for a breaking of the silos between macro-economists and financial sector specialists.15

Governance arrangements that have emerged for financial stability assessment

There has been a significant growth of governance arrangements around financial stability policy in recent years.

First, there have been enhanced institutional arrangements at both the domestic and international levels: the allocation of financial stability responsibility within domestic economies, often to the central bank; the creation of the FSB and the convening of G-20 leaders annually in support of economic and financial cooperation at the international level;16 and enhanced regulation, notably of banks, both nationally and internationally.

Second, central banks, in addition to their traditional focus on monetary policy and price stability, have increased their focus on financial stability. In some countries, financial stability committees have been established, perhaps involving multiple agencies including those with fiscal and regulatory policy responsibilities, to keep an ever-watchful eye on these risks. The allocation of financial stability responsibilities has facilitated the publication of financial stability reports on a regular, usually semi-annual, frequency to inform the public on the risks to the financial system and economy more broadly.17 At the international level, the IMF produces a semi-annual Global Financial Stability Report (GFSR) as a contribution to global financial stability and sustained economic growth of member countries.18 Clearly, meaningful data are an essential feature of financial stability reports at both national and international levels.

Further, at the national level, financial stability departments have been created and strengthened to support this enhanced analysis. At the international level, the FSB coordinates the work of national financial authorities and international standard setting bodies in order to develop and promote the implementation of effective regulatory, supervisory and other financial sector policies;19 the IMF mandates financial system stability assessments under the Financial Sector Assessment Program (FSAP) every five years for economies with globally systemically important financial sectors (See Box 1);20 while the BIS’s Financial Stability Institute assists supervisors around the world in improving and strengthening their financial systems.21

Financial Sector Assessment Program (FSAP)

The most comprehensive international approach to assessing the financial sector in individual economies is the IMF’s FSAP. The goal of FSAP assessments is twofold: to gauge the stability and soundness of the financial sector, and to assess its potential contribution to growth and development.22

This assessment examines three key components:23

  • the soundness of banks and other financial institutions, including stress tests;

  • the quality of financial market oversight in banking and, if appropriate, insurance and securities; and

  • the ability of supervisors, policy makers, and financial safety nets to respond effectively in case of a crisis.

The data requirements are focused on the first bullet—as the second and third bullets are clearly of a qualitative and judgmental nature, and include Financial Soundness Indicators (FSIs).24

As mentioned above, the IMF has identified economies that from a global perspective have systemically important financial sectors. Such economies are required to have mandatory FSAPs every five years with the intent of better safeguarding global financial stability. This identification is based on an assessment of the size and interconnectedness of an economy’s financial sector, using Gross Domestic Product (GDP) data on a purchasing power parity basis, and data from the BIS’s Locational International Banking Statistics (IBS) and the IMF’s Coordinated Portfolio Investment Survey (CPIS).25

The FSAP reports for individual economies are available at

Stress testing

Stress tests have increasingly become integral to FSA as a method of testing the resilience of the financial sector. As noted in the foreword to the IMF book A Guide to IMF Stress Testing: Methods and Models,26 “the GFC has placed a spotlight on the stress testing of financial systems.” These tests typically take extreme but plausible stress scenarios and test the extent to which different elements of the financial system would be able to cope and continue to provide financial services. Many central banks and/or regulatory agencies run stress tests, and some publish the results.

The balance sheet approach27 is a common approach to stress testing, drawing on balance sheet data of deposit-takers and other financial institutions.28 A second approach is the market price-based approach that uses market data and statistical techniques to capture interlinkages between institutions, markets, or sources of risk.29 Stress tests can be top-down, conducted by the national authorities or IMF staff (typically in FSAPs) using bank-by-bank data and applying a consistent methodology and assumptions, or bottom-up, conducted by individual financial institutions using their own internal data and models based on a common scenario.30 As stress tests are data intensive, with granular data often needed, it is important to have well maintained and consistent databases with appropriate access available for those conducting the stress tests.31

There remains room for improvement to use stress testing as a tool for macro-prudential risk assessment going beyond its traditional use for micro-prudential supervision. Demekas points out that very few stress testing models focus on, and measure correctly, the resilience of the financial system as a whole and its ability to continue providing financial intermediation services under stress in a way that makes the results readily actionable for individual banks and their supervisors.32

Financial stability policy33

The policy applications of FSA, and particularly the tools used to meet financial stability needs, are still developing.34 While there is long experience of the use of prudential regulation and micro-prudential policy tools, the same is not true for macro-prudential policy tools. Nonetheless based on international experience, the August 2016 BIS/FSB/IMF G-20 report discussed the various tools in use and their application. These include:

  • broad-based capital tools (e.g., dynamic provisioning, countercyclical capital buffers, and time-varying leverage ratio caps);

  • sectoral capital and asset-side tools (e.g., foreign currency loans to corporates, caps on loan-to-value (LTV), debt-service-to-income (DSTI), or loan-to-income (LTI) ratios); and

  • liquidity-related tools (e.g., liquidity coverage ratio (potentially calibrated by currency)), as well as tools to contain maturity mismatch (such as core funding ratios), price-based tools (such as a levy on volatile funding), and caps such as on the loan-to-deposit ratio.

Among other policy tools have been capital surcharges for global and domestic systemically important banks (G-SIBs and D-SIBs) and, scheduled from 2022, for global systemically important insurers (G-SIIs), and increases in risk-weights and large exposure/concentration limits. Further, interbank exposure limits, and foreign and domestic currency reserve requirements are being used as policy tools to lower macro-prudential risks.35

Policies have also been developed to address potential financial stability risks arising from non-bank activities, such as central clearing of over-the-counter derivatives, and in market infrastructure, such as ensuring the resilience of central counter-parties (e.g., margining requirements and liquidity resources).36

It is also important to note that beyond micro, and macro-prudential policies, other types of policies can affect financial stability, such as the tax system with incentives for debt finance,37 land and housing policies, affecting the supply and demand for real estate, and consumer protection affecting lending standards. Further there is an on-going debate as to whether interest rates could be used to meet financial stability policy needs,38 given that monetary policy and financial stability objectives are interrelated,39 and regarding the relationship between price and financial stability.40 Also, movements in the exchange rate can have domestic financial stability implications.41 Indeed, the objectives of capital flow measures—designed to limit capital flows by influencing their size or composition, can overlap with macro-prudential policies, if the latter are designed to limit systemic risks by limiting capital flows.42

To promote information sharing, the IMF, in consultation with the FSB and the BIS, is compiling a publicly available macro-prudential policy database.43

III. Data used In Financial Stability Analysis

In drafting this paper, the authors examined a cross-section of financial stability reports and IMFs FSAP reports to identify the datasets used for FSA.44 This section sets out the main “story” lines that emerge from this research. It is important to realize that FSA is constrained to available data and this has led policy makers to make a number of requests to official statisticians to expand available information. These requests are discussed in the next section. A more detailed discussion of the data used in FSA is provided in Appendix 1.

The complexity of modern economies is such that the potential risks and vulnerabilities are many and varied. They can also differ according to the nature of the economy, its financial system, and over time. Consequently, FSA has a very large demand for, and access to, meaningful data.45 Having said this, it is important to recognize that not all aspects of FSA involve data as issues such as the strength of the regulatory framework and of the “safety net” also arise.

At the core, the datasets used for FSA appear to be those that have the purpose of:

  • monitoring the soundness and efficiency of the financial system (institutions and markets), and the growth of credit to and indebtedness of non-financial sectors;

  • identifying pockets of vulnerability emerging within the financial system;

  • assessing the sustainability and vulnerability of the non-financial sectors debt positions; the potential impact on FSA of the growth in asset prices; and the financial links within and across economies that might cause shocks to permeate within the domestic economy; and

  • testing for potential vulnerabilities in the system through stress tests.

Against this background, the research reveals both a common frame of analysis to address the first three bullets above and cross-cutting issues regarding the use of time series/cross sectional data and residence-based/cross-border consolidated data. The rest of this section explores these topics.

Also from the research undertaken, the degree of sophistication and depth of markets, the range and number of financial institutions, and the extent of interconnectedness both domestically and cross-border, impacts the scope of data monitored for FSA by countries of different economic development. But the main impression arising from the research was of the similarities of analysis and commonalities of data monitored (e.g., the structure of the financial sector, the relevance of credit and debt statistics, the need to monitor asset prices, etc.). Some datasets that are particularly relevant for developing economies are highlighted below.

Framework of analysis

Macro-economic analysis is focused on economic behavior among resident entities and between resident entities and nonresident entities, within well-defined frameworks of analysis, such as the national accounts framework, and with well-established indicators of economic performance, such as growth, inflation, employment, etc. On the other hand, FSA is focused on potential risks and vulnerabilities to the system without a firmly established theoretical framework. Nonetheless, a common frame of analysis emerges from the research the authors have undertaken broadly consistent with the three interlocking objectives set out in the August 2016 BIS/FSB/IMF G-20 report.46 These objectives were:

  • (1) increasing the resilience of the financial system to aggregate shocks by building and releasing buffers that help maintain the ability of the financial system to function effectively, even under adverse conditions;

  • (2) containing the build-up of systemic vulnerabilities over time by reducing pro-cyclical feedback between asset prices and credit and containing unsustainable increases in leverage, debt stocks, and volatile funding; and

  • (3) controlling structural vulnerabilities within the financial system that arise through interlinkages, common exposures, and the critical role of individual intermediaries in key markets that can render individual institutions “too-big-to-fail.”

Figure 1 provides a schematic overview of the key data needs that emerged from the authors’ research.

Figure 1.
Figure 1.

Key Data used for FSA

Citation: IMF Working Papers 2017, 153; 10.5089/9781484306215.001.A001

Resilience of the financial system: Data are used to undertake a holistic review of the financial system that is common to financial stability reports. Such reviews encompass not just deposit-takers, but also other financial institutions, and the relationships between them; the structure of the system and concentration measures—not least for assessing the potential impact on competition; the markets in which these institutions, and other debtors and creditors, operate; the infrastructure of the financial system, such as clearing houses; and, particularly for developing countries, financial inclusion.

For deposit-takers, data collected and compiled to support prudential supervision of individual banking institutions remain essential.

Systemic vulnerabilities arising from credit and debt, and asset prices (including leverage, currency, and liquidity): Data on credit and debt are generally considered central to FSA, as research suggests that fast growth in credit can be an early warning indicator of financial crisis,47 while liquidity and solvency problems can arise with high levels of debt relative to income and wealth. As customers of the financial sector, data on non-financial corporations (NFC) and households (HH) are used to identify potential problems in these sectors that might cause problems for the financial sector, and vice versa. Due to the inherent risks, data on connected, concentrated and/or government directed lending are monitored, while the growth of credit through the FinTech industry is beginning to be assessed where relevant.

Among asset prices, real estate prices, both for residential and commercial property, equity and bond prices, as well as for land are closely monitored because fluctuations in prices affect their use as collateral, directly impact financial wealth and, indirectly impact the economy through the effect on consumer and corporate confidence.48 There is also growing interest in volatility measures so as to understand better the uncertainties/risk in financial markets.

Structural vulnerabilities within the financial system arising from financial interconnections and spillovers, both domestic and cross-border: There is an increasing use of data that supports an understanding of financial interconnections and spillovers among individual financial institutions,49 the financial and domestic non-financial sectors, and, between each sector and the rest of the world. This is often the most complex area of FSA in that financial connections between different sectors are complicated by second or third round inter-linkages—who lends to the entity funding my position, and by common exposures—I have no relationship to you except the fact that we are both exposed to the same kind of risks. Indeed, vulnerabilities can arise from the complexity of increased interconnectedness as well as from the use of complex, and often opaque, financial instruments.

Also, policies of major economies can potentially have spillover implications for the domestic economy, perhaps through unexpected channels.50 Given this, national FSA typically monitors data that helps assess developments in the international environment and the potential impact of capital flows.

Cross-sectional and time-series data

The literature suggests that it is important to distinguish between the cross-sectional and time dimension aspects of FSA. The August 2016 BIS/FSB/IMF G-20 report picks up on this distinction in noting that “systemic risk is generally recognized as having two dimensions: vulnerabilities related to the build-up of risks over time (“time dimension”), and vulnerabilities from interconnectedness and the associated distribution of risk within the financial system at any given point in time (“cross-sectional” or “structural” dimension).” This has an important implication for statistical work in that traditionally economic and financial statistics have been focused on the time dimension rather than cross-sectoral dimension, although the increasing analytical focus and use of position data is beginning to give more emphasis to the latter.51 52

Cross-border consolidated- and residence-based data

The primary interest of the authorities when analyzing financial stability is on the impact on residents53 and the domestic economy, as the ultimate goal of domestic policy makers is to protect the domestic economy. Therefore, the majority of the datasets used by domestic policy makers for FSA are residence-based. This is primarily true for data on credit and debt, financial markets, interest rates, financial market infrastructure and inclusion, and domestic and cross-border inter-connectedness.

Nonetheless, risks to financial stability may come from the activities of domestically-owned individual institutions in foreign markets – the involvement of European banks in the sub-prime market prior to the GFC being a prime example. This implies that cross-border consolidated statistics of domestically-owned individual institutions (incorporating foreign branches and subsidiaries) located in an economy are also relevant for FSA.

Indeed, data for deposit-takers is typically analyzed both for micro- and macro prudential purposes on a cross-border consolidated basis.54 For instance, the scope of application under the Basel standards for banking supervision provides that to the greatest extent possible, all banking and other relevant financial activities (both regulated and unregulated) conducted within a group containing an internationally active bank will be captured through consolidation.55 Similarly data for FSIs for deposit-takers are typically compiled and analyzed using one of a range of consolidation approaches including those based on the Basel standards.

In this context, there has been the longstanding use of BIS IBS data on a cross-border consolidated basis that captures the nationality of international banking activities, including where the ultimate risk lies. This is because, as noted by Tissot, “the IBS consolidated data yield a comprehensive picture of the national lenders’ risk exposures, in particular to country risks,”56 and so can help identify potential risks and vulnerabilities to the domestic economy arising from the foreign activities of domestically headquartered international banks.

Also, the activities of domestic subsidiaries and branches of foreign deposit-takers can be significant in the host market but relatively small within the context of the consolidated foreign banking group. In such circumstances, the behavior of these foreign affiliates can be affected as much, if not more, by activity, and decisions made, outside as inside the host market—for instance a funding shock to the parent bank or economy. In addition, vulnerabilities of subsidiaries in foreign markets may not be apparent in the home country’s residence-based data. These insights were one reason why the recent enhancements to the locational BIS IBS datasets included more granular information by nationality of the reporting bank.57 As McGuire and von Peter noted, “in any particular host country, a long or short net cross-border position in a particular currency booked by the offices of foreign banks there may be offset or hedged elsewhere on those banks’ global balance sheet.”58

Nonetheless, deposit-takers residence-based data are used for FSA, not least in terms of analyzing domestic interconnectedness and the relationship between the domestic lending and funding sides of the balance sheet. Indeed, for foreign-owned deposit-takers, the extent to which domestic lending is matched by domestic retail deposits, provides insights into the stability of their lending activity within the economy.59

Data for non-bank financial institutions (NBFI) might be analyzed on a cross-border consolidated basis, if the relevant data are available. However, residence-based data are often the only data available. For instance, the FSB annual global shadow banking monitoring report draws heavily on national financial accounts data although it also includes estimates of shadow banking that excludes NBFI that are part of a regulated banking group.

Further, while residence-based data are the basis of analyzing debt and credit, FSA is also increasingly interested in data on borrowing by subsidiaries of resident entities located abroad.60 As was seen in the GFC, many countries, particularly emerging market economies (EME), found that borrowing by foreign subsidiaries of domestic NFCs came onto the domestic balance sheet in the crisis. Even outside of a crisis, significant recent U.S dollar borrowing by foreign subsidiaries of emerging market NFCs, often through issuance of debt securities in foreign markets (offshore borrowing), has raised questions of the extent to which they are facing foreign currency risks that might in turn affect the domestic parent (see also the next section under “corporate borrowing”).

IV. What Data Gaps Have Emerged?

The previous section discussed the data used by national and international authorities in their FSA. While acknowledging the progress that statisticians have made in closing data gaps (see Appendix 2), policy makers at the national and international level have continued to draw attention to specific gaps that they consider need to be addressed. Drawing on these calls, this section sets out the most significant of these needs and suggests a way forward for each.

Before addressing the specific gaps to be filled, some more general observations about the data needed for FSA can be made.

First, from a review of the data used (see Appendix 1) it is evident that many of the needed data are already available to the financial institutions and authorities, although coverage varies across countries. Since the GFC, statisticians have taken a number of initiatives to expand the availability of data for FSA as circumstances have demanded. There have been increased efforts in several fora and significant progress has been made in closing the gaps identified, notably at the international level through the G-20 DGI61 and the IMF’s Special Data Dissemination Standard Plus (SDDS Plus).62 While these international initiatives are not relevant for all economies, implementation by countries for which they are relevant would support FSA in a significant way. In an interconnected global economy, the benefits of implementing such initiatives not only accrue for the implementing economy but also for the broader international community.

In particular, the G-20 DGI has promoted work to close gaps and strengthen datasets that support both the monitoring of risk in the financial sector (for example, FSIs, securities issuance, and credit default swaps data), and the monitoring of domestic and cross border risks and vulnerabilities (such as through sectoral balance sheets and the major cross-border surveys of the BIS and IMF) (see Appendix 2 for more details).

Despite these improvements there is often a need for further steps (i.e., enhancements of several datasets in terms of coverage, scope, quality, consistency) including some new data collection initiatives (e.g., collection of new data such as the granular dataset on globally systemically important financial institutions).

In addition, there does remain a question as to whether all the available official data are being fully used to meet user needs. Official statisticians may need to do a better job in communicating to policy makers the possibilities of available data. The user may not be aware that the data they need are available either directly, or indirectly through manipulating available data, or that available data have an informational content that is of relevance to FSA.63 One attempt to address this “publicity” issue has been through the Principal Global Indicators (PGI) website set up by the Inter-Agency Group on Economic and Financial Statistics (IAG).64

Finally, financial stability policy makers and analysts increasingly use market/private sector data as well as official statistics in their work. This is particularly true for market-related data and high-frequency data. Private sector data can be more timely if less comprehensive. But policy makers often want early indications of emerging risks and vulnerabilities. In this regard, there is also growing interest in big data as they can provide timely data at a high speed. In other words, official statistics do not, and do not need to try to, meet all the FSA data needs.

Specific Data Gaps

Shadow banking

While the banking sector has traditionally been at the heart of the financial sector, the GFC demonstrated the key role shadow banking financial institutions and markets play in credit and maturity transformation, performing bank-like activities.65 However, unlike deposit takers, these institutions are usually not strictly regulated and supervised, and have no access to deposit insurance, to the rediscount operations, or to the last resort credit lines of central banks.66

As has been emphasized by the FSB, the IMF, and other international and national authorities, there is a need for data that identifies and estimates the scale of shadow banking activity, provides a better understanding of both the entities involved and the risks they are facing, and can indicate potential vulnerabilities to the financial system arising from their activities. In doing so, the relationship with the banking industry can be assessed along with the risks to financial stability arising from shadow banking activities.

Experience during the GFC has shown that risks to financial stability may emerge in these institutions and markets from high leverage, maturity mismatches, and/or illiquidity, materialization of which could spread through the whole financial system. An example was the experience of money market funds (MMFs).67 While not typically leveraged institutions, the GFC illustrated how rapidly the risks and vulnerabilities of MMFs can be transmitted to the rest of the financial system when investors start withdrawing their funds on a significant scale: MMFs liquidated financial assets so helping to depress market prices and scaled back their wholesale funding of deposit-takers, particularly to European banks.68

This has led policy makers to adopt stricter regulatory oversight on shadow banking institutions and markets, including, greater disclosure on asset valuations and collateral haircuts, reforms of governance and ownership, as well as stricter oversight, regulation and limitations on collateral lending.69 The FSB has led the work at the international level, producing an annual monitoring report using available data (as well as addressing the regulatory aspects of shadow banking). 70

Unlike the detailed information available to the supervisors and central banks for the monitoring of the banking sector, data on shadow banking has generally been lacking due to the heterogeneous nature of the institutions, lack of regulatory oversight, a previous lack of recognition of the systemic importance of shadow banking, and a lack of a consistent definition of shadow banking.

In its May 2016 Financial Stability Review (FSR),71 the ECB pointed out the limited availability of disaggregated data needed for FSA on assets, liabilities, capital, and profitability of financial institutions other than deposit-takers and insurance companies. Eichner, Kohn and Palumbo72 pointed out that the growth of maturity transformation outside the traditional banking sector contributed to the severity of the financial crisis but was not conveyed in aggregate financial statistics for the U.S. economy.73

In addition, there is a lack of data with regard to securities financing activities for FSA considering the reliance of shadow banking institutions on wholesale funding (such as through repo and securities lending markets). For instance, the importance of closing the data gaps in securities financing markets was pointed out by the U.S. Financial Stability Oversight Committee (FSOC)74 emphasizing that data are needed to assist policy makers’ understanding of (1) how the repo market operates; (2) the interdependencies of institutions and participants; and (3) changes in risk characteristics, such as collateral and haircuts.

At present, existing balance sheet and other relevant data are collected, in most cases, under jurisdictions’ existing statistical (and regulatory) reporting requirements, with the level of granularity and frequency of reporting varying across entity types within and across jurisdictions.75 Data gaps are particularly prominent for non-regulated entities for whom the national authorities’ data collection powers often do not extend.

While the national accounts-based sectoral balance sheet and flow of funds data provide a good initial basis for assessment of the shadow banking risks, there are a number of limitations that require addressing either through methodological developments (see next section) or new data collections. In particular, data by economic function, with more granular information on maturity and liquidity transformation, and foreign currency exposures, is needed to support the risk metrics used for assessing the extent of shadow banking risks. To have a full picture of the risks and vulnerabilities associated with NBFIs, including a thorough analysis of their cross-border linkages, cross-border consolidated data on a nationality basis are needed to complement the currently available residency-based data.

The FSB sees the need for granular data on shadow banking entities on an economic functions basis, inter alia covering leverage, liquidity, and maturity transformation activities, currency mismatches, and credit intermediation activities.76 In addition, the FSB is working to develop a regular flow of data on securities financing markets at the national and global levels by end-2018, that will shed light into the size, composition, pricing, and risk profile of these markets.77

Suggested way forward

To contribute to the global efforts to improve the availability of data on the shadow banking sector, it is important for statisticians, regulators, and other users to share experiences in compiling and analyzing shadow banking data, including on ways to ensure comprehensive coverage and avoid duplication of effort. In addition, frequent dissemination of data would facilitate the timely assessment of the shadow banking system and its linkages with the rest of the financial system, and hence provide a better assessment of the systemic risks associated with shadow banking institutions and markets.

The FSB’s efforts to improve the availability of information are key78 and to this end the FSB has set up a Shadow Banking Experts Group that shares national and regional experiences in compiling and analyzing shadow banking data in the context of the FSB’s annual global shadow banking monitoring report. Further, the FSB-led work on securities financing markets will provide important information on markets in which shadow banking institutions operate. Also, the IAG Working Group on Institutional Sector Accounts is currently working on better capturing shadow banking activity using macro-economic based data, by exploring the possibility of capturing more granular sub-sectoral breakdowns and instruments (see next section) for the non-bank financial sector. All this work is endorsed by the second phase of the DGI (DGI-2) Recommendation 5 on shadow banking.

Assessment of capital flows

During the past years, there has been an increased policy interest in the financial stability policy implications of large swings in international capital flows. While the freer flow of capital is considered to have significant benefits for domestic economies including by enhancing efficiency, promoting financial sector competitiveness, and facilitating productive investment and consumption smoothing, the potential risks associated with the swings in capital flows need to be closely assessed as financial interconnectedness associated with greater capital flows can exacerbate the transmission and spillover of shocks between economies.79

In 2016, the BIS and IMF reported to the G-20 their assessments of the effects of capital flow volatility with a particular emphasis on data needs.80 Both the IMF and the BIS recognized the Balance of Payments (BoP) as a key source of information on cross-border capital flows but identified data gaps that need to be addressed to order to obtain a detailed picture of capital flows. These included:

  • More timely BoP data (shorter reporting lag) with a higher frequency of indicators to assess capital flows.

  • Identification of the direction of flows between individual countries or groups of countries, e.g., capital flows to advanced economies both from other advanced economies and emerging market and developing economies (and similarly for capital flows to emerging market and developing economies).

  • Separation of the flows associated with non-financial corporate activity from those of the financial sector in the BoP. In DGI-2 the possibility of separate identification of NFCs is being investigated.

  • Need for an increase in the number of countries disseminating the breakdown of direct investment data by geographical location, sector and currency. In DGI-2, inward and outward investment by country is promoted through the IMF Coordinated Direct Investment Survey (CDIS).

  • Need for an increase in the number of countries disseminating the breakdown portfolio investment asset and liability data by the geographical location of debtors/creditors and by currency. Under DGI-2 sector breakdowns within the CPIS are being promoted, with a move to quarterly reporting by 2019.

  • External balance sheet data on currency composition, remaining maturity of debt, and off-balance sheet items such as contingent assets and liabilities, guarantees and lines of credit, and hedging using financial derivatives.81

Further, the G-20 International Financial Architecture (IFA) Working Group in their 2016 Final Report underlined the importance of enhancing capital flows and stocks data collection to better identify currency and maturity mismatches, while also explicitly supporting the recommendations in the G-20 DGI that support capital flow analysis.82 Also, in late 2016 the IMF published a paper on “Capital Flows—Review of Experience with the Institutional View” that considered improving capital flows data a priority with a focus on the timeliness, scope and granularity of balance of payments data.83 Also highlighted was the importance of more detailed balance sheet (by sector and foreign currency exposure) and off-balance sheet data (such as contingent liabilities and derivative transactions).84

Suggested way forward

There is considerable data available on cross-border positions and flows. A holistic review of cross-border exposures data could be undertaken by the IMF Committee on Balance of Payments Statistics (BOPCOM) to see how these data could be leveraged to best meet policy makers’ needs.

In addition, data from the G-SIBs common data template that covers these institutions exposures to national markets and sectors (see Appendix 2) could be aggregated to provide an early indication of cross-border capital flows from the largest global banks. Further the template could be used by a broader range of national authorities to collect granular information on national banking systems exposures and funding dependencies. Such data would shed light on flows to and among EMEs.

Corporate borrowing

Since the GFC borrowing by NFCs, particularly in EMEs, has increased significantly, as highlighted by BIS research that has drawn on a BIS database of total credit to NFCs.85 86 These data show that NFC debt in the major EMEs increased from less than 60 percent of GDP in 2006 to 110 percent at end-2015. Further, the BIS research points out that any analysis of the vulnerability of EME debtors to foreign currency exposures must take account of leverage, debt maturity, and the external/domestic distinction of debt. Against this background, specific data gaps for NFCs include (i) foreign currency borrowing, particularly through off-shore affiliates; and (ii) information on corporates’ risk exposures, such as maturity mismatches and foreign currency exposures (including hedging activities).

Regarding NFCs’ foreign currency borrowing, the BIS international debt securities (IDS) database provides comprehensive information on total issuance of international debt securities, with currency and maturity breakdowns. But there is a lack of data on NFCs offshore foreign currency borrowing from deposit-takers as noted in the August 2016 BIS note to the G-20 IFA Working Group.87

The BIS IDS database highlights the scale of off-shore borrowing in debt securities. As of September 2015, offshore borrowing accounted for a significant amount of total (including offshore) borrowing through international debt securities by Chinese (93 percent), Brazilian (53 percent), and Russian (45 percent) nationality NFCs. As noted by the Bank of England (BoE)88 this offshore borrowing by NFCs with a global presence cuts across traditional residence-based data either, as BIS explains, not showing up in residence-based external debt statistics (when proceeds are not repatriated) or classified as foreign direct investment (FDI) flows (when repatriated). In either case, residence-based measures could paint an overly benign picture of vulnerabilities89 and does not capture all the potential financial stability risks facing a country.

In addition to cross-border foreign currency borrowing by NFCs in international debt securities, domestic foreign currency borrowing, e.g., from domestic deposit-takers, also needs to be assessed as this form of borrowing also exposes NFCs, and through the NFCs, the domestic deposit-takers, to foreign exchange risks.

Consistent information on off-balance sheet activities, such as contingent assets and liabilities, guarantees and lines of credit, and hedging using financial derivatives also remain data gaps. While countries, at the national level, generate some information based on different data sources, through surveys or through information from derivatives exchanges, lack of consistency in the coverage and definitions used across jurisdictions does not allow for meaningful aggregation at an international level.90

Also, the August 2016 BIS note to the G-20 IFA Working Group points out that there is no international database on NFCs financial assets including currency and maturity composition as well as on the country and sector of their counterparty debtors.

According to the BIS, the lack of information contributes to the uncertainty about NFCs volume of foreign currency exposures, the links with the banking system, and the degree to which hedging reduces systemic risk.

Suggested way forward

Recommendation 14 of DGI-2 asks the IAG to improve the consistency and dissemination of data on NFCs’ cross-border exposures, including those through foreign affiliates and intra-group funding, to better analyze the risks and vulnerabilities arising from such exposures including foreign currency mismatches.

The BIS note to the G-20 IFA Working Group suggested that in the short term combining the residence-based BoP data with the BIS IBS and IDS could shed more light on NFCs cross-border exposures and their evolution. The IAG document produced for the DGI also sets out some ideas for further work with regard to capturing NFCs cross-border exposures.91

As also emphasized by the BIS, enhanced disclosures of financial hedges and derivatives positions (including detailed currency and maturity information on financial hedges and their underlying positions) on a timely basis through improved accounting standards could also contribute to the availability of consistent information on the risk exposures of NFC.

Granular and micro-data

With the nature of financial stability risks changing over time, FSA needs to be sufficiently flexible to address shifting vulnerabilities. As Jenkinson and Leonova92 emphasized, given the increasing focus of financial stability on the risks to the financial system as a whole, new approaches to financial data based on the uniform representation and standardization of its key elements is becoming more important to allow for flexible data aggregation to support multiple policy objectives.

To this end, detailed and granular information is increasingly being requested to contribute to the flexibility of FSA tools.93 Several statistical initiatives mentioned in this paper aim to increase the granularity of available information (e.g., the common data template on G-SIBs’, data on repo and securities financing transaction, enhanced BIS IBS data, sectoral balance sheets, and the enhanced IMF CPIS). There is particular emphasis on the sector, country, and currency dimensions of both creditor and debtor positions, all of which are important to FSA.

If shared, granular data would allow statistical compilers to identify, and resolve inconsistencies between data compiled in different institutions and in different countries, while possibly reducing the burden for the data reporters. In addition, as the policy makers’ data needs change, through the availability of granular data, statistical agencies could compile aggregates in ways that meet these changing needs without sending data requests to data reporters.94

To meet the need for increased availability of granular data not only could the collection of more granular data be considered but more use could be made of existing micro data (data that are collected for supervisory or micro-prudential purposes). In this context, the development of principles for effective risk data aggregation and risk reporting by the Basel Committee on Banking Supervision (BCBS),95 the creation of a common data template for G-SIBs to include bi-lateral exposures and exposures to countries/sectors/instruments, and the development of a legal entity identifier system (LEI),96 to identify unique parties to financial transactions are all relevant.

Other initiatives to strengthen financial institutions’ risk reporting practices include data reporting requirements arising from the implementation of Basel III97 and the Solvency II rules; the development of recovery and resolution plans by national banking groups; and the efforts to enhance international financial reporting standards.98 In addition to contributing to financial institutions’ own risk managements, the improvements in regulatory reporting can contribute to the quality of the more aggregate macro-prudential data for the assessment of system-wide financial stability risks at the national, regional, and international levels.

However, the use of micro data for macro financial assessment has its challenges, the most important being the strict confidentiality requirements associated with the use of micro data. Such requirements typically limit data sharing among statistical and supervisory agencies, and with users. But also granular information brings data quality and consistency issues that need to be dealt with to be able to draw appropriate conclusions for macro-prudential analysis. Tissot points out the importance of being able to aggregate micro information so it can be analyzed, and communicated to policy makers while on the other hand the “macro” picture on its own can be misleading, as it may mask micro fragilities that have system-wide implications.99

Macro-stress testing is a key tool to assess the resilience of financial institutions and sectors to shocks and would benefit from more detailed information particularly for the top-down stress tests.

Suggested way forward

Work is ongoing as part of Recommendation 20 of DGI-2 to promote the sharing of data within jurisdictions and with other national authorities.100 However, given the differences in legal, statistical structures and cultural backgrounds across jurisdictions, enhancing data sharing is a challenging task and cannot be accomplished overnight. Going forward, international organizations (IOs) should continue their facilitator role by creating platforms to exchange experiences and to help the building of trust.

Real Estate Markets

Considering the potential direct and indirect effects on the stability of the financial system, as demonstrated during the GFC, national and regional authorities are placing increasing emphasis on the monitoring of real estate markets.

Significant improvements have been made by national authorities since the GFC in both the scope and coverage of data on residential and commercial real estate markets. This improvement has been promoted in particular through the support of the DGI, the BIS public property price statistics database, and the inclusion of residential property price index (RPPI) among the core FSIs. Residential real estate prices are also an item in the SDDS Plus.

Despite the progress in the number of economies disseminating real estate price indices, the datasets on residential and commercial property prices vary in terms of quality and coverage. For the residential property prices, given the availability of conceptual guidance,101 the situation is relatively well covered by the BIS database, with wide country coverage, some consistency of data, and, for several advanced and emerging economies, with long-time series data. But the geographical and type of property coverage varies significantly among countries. As regards commercial property prices, their coverage in the BIS public property price statistics database has been expanding significantly since 2016 in the context of the DGI-2, although data are currently available from only a few number of countries with differing frequencies and scopes (e.g., in terms of type of property, area covered, compilation method).

Against this background, the need to improve the quality and availability of data on real estate markets has been emphasized in the Financial Stability Reviews of many economies.102 In November 2016, the European Systemic Risk Board (ESRB) published a recommendation103 on closing data gaps for residential and commercial real estate markets, underscoring the significance of developments in the real estate sector for financial stability and the considerable data gaps that continue to exist in this area. The aim is to establish a more harmonized framework for monitoring developments in real estate markets in the European Union. The recommendation sets out a common set of indicators that national macro-prudential authorities are recommended to monitor along with working definitions of these indicators.

Finally, in addition to the price indices, there is an FSA need for additional housing-related indicators to complement the price indices.

Suggested way forward

At the international level, guidance has been provided on the compilation of RPPI, while for Commercial Property Prices Indices (CPPI), conceptual guidance is in early stages of development.104 Going forward, national efforts are key to ensuring the availability of consistent data on property prices, and other indicators of the property market.

Under Recommendation 17 of DGI-2, the Inter-Secretariat Working Group on Price Statistics (ISWGPS), led by the OECD, and in collaboration with the IAG, is developing a list of other housing-related indicators, such as price-to-rent and price to income ratios.

Insurance companies

As emphasized in the IMF April 2016 GFSR,105 before the GFC insurance companies were not thought to pose significant systemic risks having longer-term liabilities, greater diversification of assets, and less extensive interconnections with the rest of the financial system than deposit-takers. However, the near-collapse of the AIG in 2008 revealed the potential systemic risks that could be associated with large insurance companies. As a consequence, the International Association of Insurance Supervisors (IAIS) has identified| G-SIIs whose distress or disorderly failure would cause significant disruption to the global financial system and for whom additional capital surcharges are scheduled to be applied starting in 2022.106

While there is more comprehensive data on insurance companies available from micro and supervisory data sources compared to other non-bank financial institutions, data gaps (such as information on liability structures) still remain, addressing of which would allow for more complete risk assessments.107 In this context, the April 2016 GFSR emphasizes that while progress is being made on the micro side, there needs to be a greater macro-prudential focus.

Enhancements to insurance sector data would include better data on common exposures, on interconnections with other financial institutions including cross-border, on the duration gap between assets and liabilities, and on the structure of liabilities including for life insurance companies the relative size of minimum guaranteed products108 and variable annuities within total liabilities.

Suggested way forward

Under Recommendation 4 of DGI-2 the FSB, in close consultation with the IMF and IAIS, is to consider the possibility of a common data template for G-SIIs. As with the G-SIBs common data template, developing such a template and the subsequent collection of systematic granular information could be challenging, although the work would benefit from the experiences with G-SIBs. Depending upon the outcome of this initiative, in the long-term collection of granular information using the template could be considered including more widely by the regulators for domestic and non-systemic insurers.


Another area where better data are needed to assess financial stability risks is related to the monitoring of the household sector.109 Such data include comprehensive information on the composition of assets and liabilities, and household income and debt service payments.110 Further, the growing interest of policy makers in the inequality gap (i.e., of consumption, saving, income and wealth) has led to a demand for distributional information.

Suggested way forward

Countries could share their experiences in the compilation of household data including as part of their sectoral accounts statistics. While household surveys are key data sources to provide structured information, they are costly to conduct therefore could be complemented with administrative data to the extent that the confidentiality restrictions allow.

As part of Recommendation 9 of DGI-2, the OECD, in cooperation with Eurostat and the ECB, is working with G-20 economies to encourage the production and dissemination of distributional information on income, consumption, saving, and wealth, for the household sector based on the sectoral accounts framework.

Figure 2.
Figure 2.
Figure 2.

Key Data Gaps and the Suggested Way to Close Them

Citation: IMF Working Papers 2017, 153; 10.5089/9781484306215.001.A001

V. How can FSA Data Needs be Addressed in Economic and Financial Statistical Manuals?

While the FSA data needs identified cover a wide range of data series, and the previous section discussed the data gaps that are requested be filled, the question arises as to whether there are common themes in the data needed for FSA that could be met through adaptions of the System of National Accounts (SNA), BoP and related manuals (macroeconomic statistical manuals).111 The authors believe that such common themes do exist and so advocates a discussion on how the national accounts framework might be best developed to help meet the needs of FSA in the upcoming review of the manuals, likely to start later this decade.

The paper makes this call for three main reasons:

  • Since the last update round in the 2000s there has been a much-increased policy focus on financial stability, and it is the purpose of each update round to take account of economic and financial developments, and the consequential needs of policy makers, that have inevitably occurred since the last round;

  • the macroeconomic statistical manuals have a central role in the production of economic and financial statistics at national and international statistical offices, with the SNA covering the whole economy; and

  • to support an integrated approach to the use of datasets for different policy purposes avoiding duplication of data collection.112

As background, the underlying conceptual framework is grounded in sound theoretical economic concepts with the consequence that it has remained largely unchanged over many decades. The periodic updates of the core manuals have thus focused on enhancements that: (1) address new economic and financial developments, and new and emerging policy needs; (2) provide a fuller exposition of existing conceptual advice; and (3) further integrate conceptual advice across the various macroeconomic statistical manuals.

To contribute to the discussion this section sets out some suggestions for enhancements to the macroeconomic statistical manuals with regard to credit quality, financial derivatives, remaining maturity and foreign currency, the sub-sector breakdown of NBFI, and distinctions by size. In choosing these items, two considerations were taken into account:

(1) That the proposed enhancements are consistent with, but avoid as far as possible overwhelming, the SNA framework. This is an important consideration because the conceptual framework of these macroeconomic statistical manuals is designed primarily for national authorities to collect, compile, and disseminate data to help support macroeconomic policy making; and

(2) The items are aligned with financial stability policy needs and tools based on identified user needs that have been commonly expressed and referenced earlier in the paper.

When referring ahead to including the enhancements in the "central framework,” this means a proposal to include the enhancements in the core statistical accounts, not as supplementary or as memorandum items.

A. Nonperforming loan (NPL) and provisions

While the macroeconomic statistical manuals provide guidance to compile credit data, and the System of National Accounts (SNA) recommends supplementary items on contingent items such as loan commitments, letters of credit, guarantees, etc. (2008 SNA paragraph 11.24), there is a lack of information on credit quality for non-traded instruments in the central framework.113 Yet credit quality information is important to FSA as it is an indicator of problems borrowers are having, with implications for creditors.

The SNA does recommend memorandum items on NPLs, at nominal and market value, for the government and financial corporations sectors and if significant, as supplementary items for the other sectors, including the rest of the world (2008 SNA, paragraph 13.67). Provisions are in the central framework of the Monetary and Financial Statistics Manual and Compilation Guide (MFSMCG), as they are taken into account when determining the capital of deposit-takers by being included in other accounts payable [Monetary Statistics[MS]] (see MFSMCG paragraph 2.32 and Figure 2.2).

Two possibilities exist for bringing some measure of creditworthiness into the central framework of the macroeconomic statistical manuals. First, NPLs at nominal value could be introduced into the central framework for all sectors with data from creditors providing information on the counterpart borrower sector.114 Flows for NPLs would be recorded as other changes in volume of assets (OCVA).

Second, provisions for losses on assets that are valued at nominal value could be brought into the central framework as provisions affect economic activity, both through the impact they have on the profitability of credit extension and, for deposit-takers, on capital through regulatory provisioning practices. Further, as credit quality worsens and provisions increase deposit-takers typically become more cautious in their lending activity. The flows would be recorded as OCVA given provisions are not an exchange between parties, allowing the outstanding value of loans to be calculated more closely reflecting their market value.115

This paper considers including provisions in the central framework rather than NPLs to be the more robust approach for the reasons given below, while there is already compilation experience through monetary data.

In the national accounts, if credit quality deteriorates, for traded instruments, the market price changes, resulting in lower net worth of the creditor or a decrease in the market price of the equity liability. However, for instruments valued at nominal value, such as loans, a deterioration in credit quality is not reflected in the value of the instrument but because it might well feed through to a lower market price of equity liabilities of creditors, is likely to be reflected in an increase in net worth as measured in the national accounts system. The latter arises because net worth is the balancing item of the national accounts balance sheet. So, the present approach reduces the analytical value of the accounts because it does not reflect economic developments in, or attribute them to, the relevant instruments, and disguises signs of worsening creditworthiness among debtors.

Unlike debt securities which the debtor can buy back in the market, for non-traded instruments there is not that opportunity, so from the debtor, and indeed creditor perspective, the value of the debt obligation remains the full contractual amount. So, it can be argued that the value of non-traded instruments valued at nominal value should remain the amount owed without adjustment for provisions, as indeed is the approach in the MFSMCG and FSI Compilation Guide, with provisions added as a separate line item in the accounts. This approach would have the advantage of not only ensuring that the amount owed continues to be recorded but that provisions and write-offs would be separate line items in OCVA, because write-offs, unlike provisions, reduce the amount owed and hence the outstanding value of the instrument.

Including provisions in the central framework of the macroeconomic statistical manuals would affect the timing of the transfer of value within the system as value would transfer when the provisions are made rather than when write-offs occur.116 But as indicated above such timing more accurately reflects the profitability and net worth of deposit-takers, and avoids disguising a deterioration in the creditworthiness of debtors.

B. Notional value of derivatives

Financial derivatives were introduced into the central framework in the 1990s as these markets begun to flourish. The data are compiled at market value consistent with the principles of the SNA. However, financial derivatives are not debt instruments through which economic agents finance imbalances in consumption and production but rather instruments through which risk is transferred around the system. Recording only market value misses the extent of risk exposures and transfers, and it is these risk exposures and transfers around the system that interest FSA.

So, to gain a fuller picture, not least to measure foreign currency exposures and leverage more broadly, data on notional value (in addition to market value) are needed. Indeed, the Balance of Payments and International Investment Position Manual, sixth edition (BPM6) includes the notional value of foreign currency derivatives in its memorandum table on foreign currency, while the BIS publishes notional (and market) values in its six-monthly survey of over-the-counter derivatives (on a cross-border consolidated basis). But despite these data sets, important as they are, there lacks a residence-based economy-wide picture of financial derivative positions by risk category by sector and sub-sector.

So, while recognizing that notional value does not fit the conceptual framework of the macroeconomic statistical manuals, but to provide a more comprehensive view of the risks underlying the economic and financial system, and how they change over time, the full range of financial derivative positions held, by type of risk category, by counterpart sectors, at notional value could be added as memorandum items.

C. Remaining maturity

Original maturity of debt assets and liabilities is the standard approach to maturity in the macroeconomic statistical manuals, with a distinction between short-term (up to one-year) and long-term (over one-year). While data on an original maturity basis is of interest to FSA in that it provides information on borrower’s access to the short and long markets, there is greater FSA interest in remaining maturity as it informs on debt falling due in the near-term. Remaining maturity data helps indicate the amount of debt that needs to be refinanced, the liquidity of debtors and creditors, and the extent of maturity mismatches between assets and liabilities.

A number of manuals including the BPM6, MFSMCG, and Public Debt Statistics and External Debt Statistics Guides, have already introduced remaining maturity as a memorandum or supplementary item to position data: long-term original maturity data is broken down into up-to-one year due and over-one year due; and by adding the up-to-one year due data to short-term original maturity data, short-term maturity on a remaining maturity basis can be calculated without undermining the concept of original maturity. Bringing this distinction into the position data in the central framework of the macroeconomic statistical manuals would help meet the needs of FSA.117

D. Foreign currency

Policy makers have clearly indicated through the G-20 a need for more information on foreign currency exposures. The MFSMCG includes a domestic and foreign currency breakdown through its sectoral balance sheet by instrument and counterpart; BPM6 includes a memorandum table with a foreign currency breakdown of positions by sector, major currency, that also takes account of financial derivatives; and the Public Debt Statistics and External Debt Statistics Guides have domestic/foreign currency splits in their presentational tables.

Introducing a foreign currency/domestic currency breakdown into the central frameworks of both the SNA and the BPM, combined with the introduction of a remaining maturity breakdown in the position data would immensely improve understanding of the foreign currency risks facing the economy. Supplementary or memorandum items breaking down foreign currency data by major currency could also be considered.

E. Sub-sector breakdown of NBFI

The 2008 SNA introduced a new breakdown of NBFIs118 with seven sub-sectors.119 The composition of the seven sub-sectors is logical and well-considered but depending on countries experience with compiling and analyzing data for the seven sub-sectors, the subsections of the NBFI could be reviewed to determine if the sub-sectoring of NBFIs should be modified to meet the analytical needs of FSA.

The FSB has developed an analysis of shadow banking using five economic functions.120 While such a characterization might not be appropriate for the SNA, given the interest of policy makers in shadow banking activity and entities, the work of the FSB could inform any SNA review of NBFI sub-sectoring.121

F. Distinctions based on size

The SNA and BPM6 frameworks make no allowance for size of entity in advising on the compilation of sector and sub-sector aggregate data. Nonetheless, there is considerable interest among financial stability analysts in activity by households by income level, by non-bank financial corporate by sales, assets, and/or employment, and deposit-takers by assets.

While such data are not typically disseminated by national statisticians, the raw data they receive from reporters often allows for such data to be compiled. Therefore, the possibility of compiling and disseminating data by size, probably as memorandum items, could be investigated. The data for households could be informed by the work under the DGI on household distributional information.

VI. Concluding Remarks

The past 10–15 years have seen a major change in policy makers’ attitude to analyzing financial stability. Particularly since the GFC there has been an emergence of FSA governance arrangements alongside an increased focus on macro-prudential analysis. With FSA firmly established this paper has undertaken a holistic stock-take of the types of data series used. Our understanding is that this is the first review of its kind at the international level. The paper has found that while the specific datasets used can differ across country and over time, common patterns of data use emerge.

Where does this leave statisticians? Overall there has been an encouragingly constructive response to this increased policy focus on FSA, not least through the G-20 DGI. But more work is required to meet FSA data needs, not least in implementing the initiatives underway. This includes data relating to shadow banking, capital flows, and corporate borrowing, as well as the increased demand for granular data. Further, with the start of the update round of the SNA and BPM expected later this decade, this paper has identified enhancements such as adding provisions data and including remaining maturity and foreign currency breakdowns in the central framework to support FSA without undermining the conceptual framework of the manuals.

Financial Stability Analysis: What are the Data Needs?
Author: Mr. Robert M Heath and Evrim Bese Goksu