Remarks on the collection of credit data (AnaCredit)

Speech by Sabine Lautenschläger, Member of the Executive Board of the ECB, at the Committee on Economic and Monetary Affairs (ECON) at the European Parliament, Brussels, 25 January 2016

Mr Chair,

Honourable members of the Economic and Monetary Affairs Committee,

Ladies and gentlemen,

Thank you very much for this opportunity to discuss “AnaCredit” with you. Such occasions are highly valued both by the European Central Bank (ECB) as an institution and by me personally, as they enable open discussions where we can explain our policies and, in this case, one of our major projects.

There have already been several exchanges between us in writing; Mr Bier, the ECB’s Deputy Director General Statistics, attended a hearing in December 2015; and there was also an inquiry by the Ombudsman, which is now closed. So a considerable amount of information on AnaCredit is available. Nevertheless, I understand that several questions and possible misunderstandings remain, some of which I would like to address in my introductory remarks by:

  1. recalling why AnaCredit is pivotal in supporting the ECB in the performance of its statutory tasks;
  2. defining AnaCredit’s requirements; and
  3. setting out the costs linked to AnaCredit and the stakeholders’ involvement.

I would be happy to answer other questions later.

Rationale for AnaCredit

The ECB’s Governing Council decided to launch the AnaCredit project in 2011. It will establish a Eurosystem database comprising loan-by-loan data on credit granted by euro area financial institutions. The background for this decision was that, currently, the information provided on granular credit datasets across the euro area is very heterogeneous. AnaCredit will standardise the collection of credit data and significantly improve the Eurosystem’s ability to conduct in-depth analysis and meaningful comparisons of the provision of credit within the euro area.

Loan-by-loan information provided by AnaCredit will enable the consistent analysis of information on the demand and supply of credit for monetary policy purposes. Slow credit growth may be linked to supply hindrances, such as banks’ low asset quality, or demand obstacles, such as the high indebtedness of certain industries being reflected in higher borrowing costs. Being able to match the size of the loan to its interest rate and the collateral used will also enable us to gain a better understanding of how small and medium-sized enterprises (SMEs) are affected by monetary policy measures. Clearly, having an accurate breakdown of how monetary policy measures affect the economy in general, and price levels in particular, is an essential ingredient for their preparation and for ensuring their effectiveness. Data provided by AnaCredit will also support our financial stability analysis, as it will enable us to consistently evaluate macro-financial risks originating from exposure to sectoral risks, and to suggest fine-tuned macroprudential policies.

AnaCredit will also be critical in supporting the ECB’s monetary policy operations and risk management activities, both of which are key for a central bank.

Banks will also benefit from AnaCredit. Credit institutions are allowed to take part in monetary policy operations, provided they are eligible counterparties and provide adequate collateral for the liquidity provided by the ECB. Loans to non-financial institutions or households are eligible for use in these operations, subject to certain conditions. When assessing the eligibility of this collateral, AnaCredit will be fundamental.

AnaCredit, therefore, is a crucial development that will greatly aid the Eurosystem in conducting its core tasks, ranging from monetary policy and monetary policy operations, through financial stability analysis and macroprudential policy, to collateral and risk management, as well as economic research and the production of statistics.

It is very clear that, while some of the data could also be used for supervisory purposes, AnaCredit in its current format is not designed for banking supervision. The AnaCredit project was launched in 2011 – long before European banking supervision was even considered. The draft regulation on AnaCredit, which was published on 4 December 2015, does not include any specific supervisory requirements.

Nevertheless, based on the experience gained with AnaCredit, the ECB may decide to follow a staggered approach of progressively covering different market segments in different stages at significant intervals. AnaCredit could also be extended to cater for supervisory requirements in the future. In line with our legal obligations, any such plans would have to undergo public consultation before the Governing Council could take a decision on the matter.

Defining the requirements

The usefulness of AnaCredit results from the requirement to provide 94 data attributes relating to features of the debtor, the loan, the interest rate and the collateral securing the loan. However, the number of actual variables is fewer than 94. For instance, the borrower’s address alone contributes seven data attributes. However, given the multiple purposes for which the data are intended, and in order to reduce the reporting burden for banks, which might otherwise have to provide similar information several times, these 94 data attributes are necessary.

More importantly, AnaCredit requires information that is already available to every bank and that they need in order to grant credit, while borrowers will hardly have to provide more information than they currently do.

The setup costs of AnaCredit

Nevertheless, AnaCredit does not come without costs, and I fully acknowledge that the reporting burden for banks is of particular concern.

The initial investment by banks to improve their IT systems and reporting practices will, over time, be offset by the gains arising from lower running costs. Compared with collecting aggregated statistical information, there will be fewer ad hoc data requests from central banks. Such on-demand requests are typically burdensome and costly for banks. Thanks to the flexibility offered by granular data, the ECB will be able not only to compile the aggregate statistical information but also to combine and reconfigure the existing data to continually construct new statistics and relevant indicators.

However, the setup costs will vary greatly across banks, and costs for small and medium-sized banks in particular have been a focus of attention. It is important to note that many small and medium-sized banks are part of an association with a common IT service provider. Combined, they are, in many credit market segments, bigger than some large banks and just as cost efficient. In fact, due to shared IT solutions that automate data management, credit data collection and processing are the same irrespective of whether a credit is provided by a small or large credit institution within the same association.

Moreover, the draft regulation on AnaCredit explicitly gives national central banks the option to take into account the specific situations of credit institutions by exempting them, in part or in full, from reporting. For some countries (those with many small banks), as many as 700 to 1,000 small and medium-sized banks could be granted such exemptions. I am confident that the national central banks will find the right balance to ensure that their small and medium-sized banks are not overburdened.

The stakeholders’ involvement

In 2014 the ECB conducted the “merits and costs” procedure as referred to in Council Regulation (EC) No 2533/98 concerning the collection of statistical information by the ECB. This quality assurance procedure rationalised the requirements of AnaCredit with a view to balancing the benefits and costs, in particular for banks. To ensure that the ECB’s cost assessment would benefit from the views and inputs of stakeholders, the ECB held over 25 meetings on AnaCredit with banking representatives at the European level. In addition, at the national level, national central banks have held countless meetings with their banking representatives.

The Governing Council also decided to publish the draft regulation on AnaCredit and provided the public with the opportunity to submit observations by 29 January 2016. Before the regulation is adopted, the Governing Council will carefully consider any comments received.

Conclusions

Credible and high quality statistics lie at the heart of the ECB’s policy-making. AnaCredit reflects a key statistical feature, namely the ability to adapt flexibly to new economic circumstances. We should also ask ourselves the following question: what would be the costs of not implementing AnaCredit?

The crisis reminded us how hard it is to predict the policy questions that will need to be addressed in a few years from now. As a result of the collection of granular data through AnaCredit, Eurosystem policy considerations will be far less constrained by statistics not being available in a timely manner.

Speaking engagements

Media contacts