Addressing global data availability challenges
Panel remarks by Benoît Cœuré, Member of the Executive Board of the ECB,at the Eurofi Financial Forum 2015,Luxembourg, 10 September 2015
It is a great pleasure for me to participate in this panel alongside such distinguished fellow speakers. I will share with you my views on remaining data gaps. The closing of those gaps is essential for the ECB’s conduct of monetary policy and the effective implementation of the ECB’s micro-prudential, macro-prudential and financial market oversight functions. I will also briefly touch on the important subjects of data sharing and data transparency.
However, let me first point out that there are a number of ongoing international initiatives aimed at improving the information frameworks on which key policy decisions are based. These have largely been driven by the Data Gaps Initiative (DGI) launched in April 2009, which supports several G20 projects, such as bilateral and multilateral surveillance, and initiatives focusing on global systemically important financial institutions. 
As a member of the Inter-Agency Group on Economic and Financial Statistics, the ECB has been deeply involved in the development and implementation of these recommendations and strongly supports the second phase of this initiative (DGI-2).
In this context, I would like to mention three important achievements. First, the ECB has introduced new statistics on securities holdings. These provide very detailed insight into who holds which securities in the euro area.  Second, earlier this year, the ECB, together with the BIS and the IMF, published a “Handbook on Securities Statistics”. This enables the collection of better securities data by clarifying concepts and providing guidance on the harmonised presentation of securities data. Third, the recent ECB regulation on supervisory financial information, which will gradually extend reporting requirements to entities that have not yet been reporting on the basis of FINREP, is one example of where we have closed a data gap and increased the consistency of data at the same time.
Despite this encouraging progress, serious data gaps, inconsistencies and inefficiencies still exist in several major policy and supervisory fields. Allow me to briefly mention a few key examples.
In the field of credit, significant data gaps still exist. The availability of complete and granular credit register information and corresponding risk parameters is critical in order to assess the concentration of exposure throughout each banking group across relevant dimensions. The absence of this information makes it difficult for supervisors to perform stress tests without collecting additional ad hoc data (as in the case of the European Banking Authority’s stress test). Such additional data requests place a large burden on banks, but also on the supervisors tasked with handling and ensuring the quality of these ad hoc data. Furthermore, those ad hoc requests have long lead times and limit supervisors’ ability to react quickly to changes in the financial environment. Thus, credit register information improves overall monitoring and the ability to react in a changing environment.
There are also large remaining gaps in the field of derivatives, where the ability to monitor the size of derivatives markets and the interconnections between financial market infrastructure (FMI) participants is key if we are to successfully assess the risks across them, especially considering the global nature of these markets and the fragmentation of the FMI landscape worldwide. Data are widely available, but they are fragmented and reported in a non-harmonised way.
In the area of bank funding, we are still missing a timely and comprehensive overview of a bank’s access to different funding sources in the money markets and the conditions under which a bank is able to finance its activity.
We definitely need to gain more insight into the shadow banking sector. This sector is continuing to expand and increase its share of financial intermediation, including as regards lending to the real economy. More comprehensive reporting and reliable data on shadow banking entities and activities will be required in order to extend the macro-prudential framework to cover the shadow banking sector. The forthcoming EU regulation on securities financing transactions will be very helpful in this respect.
From a policy perspective, there is also a need to have sufficiently long time series for macro variables, such as credit growth, but also for micro data (including bank-specific information), in order to be able to identify the build-up of financial imbalances.
In addition to critical data gaps in terms of scope, there are also plenty of data inefficiencies and inconsistencies caused by a lack of harmonisation and standardisation. Where harmonised identifiers do exist, they have not been globally implemented, as in the case of the Legal Entity Identifier (LEI). This contributes to the current inability to properly integrate and aggregate existing data in a comprehensive manner and the lack of a consistent overview of inter linkages and concentration risks in the banking sector and other parts of the financial system. To this end, standardisation and harmonisation efforts need to be stepped up, as demonstrated by the ongoing work of the Committee on Payments and Market Infrastructures (CPMI) and the International Organization of Securities Commissions (IOSCO).  When developing globally harmonised identifiers, such as Unique Transaction Identifiers (UTIs) and Unique Product Identifiers (UPIs), we should rely, wherever possible, on data standards that already exist (e.g. ISO standards). These efforts are critical to ensure national, regional and global aggregation where needed (for example, in the case of globally systemic banks and data on derivatives). In addition, the compulsory use of these standards for reporting and wide access to a minimum set of standardised identifiers for all stakeholders is key
The issues that I have mentioned clearly show that there are serious data gaps in some areas. Meanwhile, in other areas, data are available, but we have the issue of data sharing and data transparency. A wide variety of authorities are involved in collecting data, and efforts are needed at the level of authorities requesting data. Stronger coordination will, over time, facilitate the integration of data, which are often requested on the basis of a stovepipe approach and governed by separate legal frameworks. As a result, underlying concepts, definitions and standards can all vary from dataset to dataset, making the combining of data very labour-intensive and subject to all kinds of misinterpretation and overlaps. This endeavour requires political commitment from all stakeholders. It also requires that authorities collecting data share relevant information with other authorities.
I fully acknowledge the existing constraints and barriers that may currently hamper data sharing between authorities, such as data confidentiality and the different mandates of market supervisors, regulators and central banks. However, I think that sharing data among policy-makers and supervisory authorities is essential. Understanding interconnectedness in the global financial system and identifying sources of systemic risk were key motivations for establishing the Financial Stability Forum in 1999.  Sixteen year after the Tietmeyer Report, and one crisis later, we’re still not there. We need to establish a “circle of trust” bringing together certain authorities among which data should be shared more easily.  Authorities have a duty to overcome the obstacles that are currently preventing the sharing of data and endeavour to use harmonised data wherever possible.
I am sure the panel will explore these issues in more detail during our discussions.
As early as April 2009, the finance ministers and central bank governors of the G20 called on the International Monetary Fund (IMF) and the Financial Stability Board (FSB) to identify major financial and economic information gaps that needed to be filled. As a result, a few months later, in September 2009, the IMF and the FSB presented the report that would launch the Data Gaps Initiative. G20 leaders have pledged to close data gaps to support financial stability, with those gaps including: data measuring the build-up of risk in the financial sector (e.g. financial soundness indicators and coverage of risk transfer instruments, including data on the credit default swap markets); data measuring financial network connections (particularly for systemically important global financial institutions) and identifying the activities of non-bank financial institutions. It is important to note that these gaps do not relate solely to banks and insurance companies. Data gaps are also prevalent in the non-regulated sector.
The ESCB’s Centralised Securities Database contains reference data on millions of individual securities. These reference data are linked to a granular database on securities holdings, which allow policy-makers to assess “who holds what” when it comes to securities. The main euro area sectors and the 26 largest banking groups holding individual securities are covered. A legislative process aimed at extending the collection to cover all groups under the ECB’s direct supervision is under way.
See Financial Stability Board, „Feasibility study on approaches to aggregate OTC derivatives data“, September 2014, and the consultative report by the CPMI and the Board of IOSCO on „Harmonisation of key OTC derivatives data elements (other than UTI and UPI) – first batch“, September 2015.
„While the national authorities, the IFIs, the international regulatory bodies and the central bank groupings each gather and assess data on segments or the whole of the international financial system, and regularly monitor aspects of financial system stability, none has the breadth of information or the capacity to formulate a complete assessment of evolving risks. In view of the many sources from which systemic risk can arise, better procedures are needed for systematically pooling and efficiently using the information available to these bodies in their assessment of domestic financial vulnerabilities and global conditions.“ in Tietmeyer, H. “Report on international cooperation and coordination in the area of financial market supervision and surveillance”, Bank for International Settlements, 11 February 1999.
The recommendations for the second phase of the G20 Data Gaps Initiative include Recommendation II.20 on the promotion of data sharing by G20 economies, which includes an increase in “the sharing and accessibility of granular data, if needed by revisiting existing confidentiality constraints”.