Explaining variation and change in supervisory confidentiality in the European Union

Abstract

Some European Union member states’ financial regulators choose to make some of the data they routinely collect on individual banks publicly available. Others treat this data as confidential. What explains this difference? This paper considers the possible effects of crises, path-dependent legal institutions, and the design of deposit insurance schemes. At the national level, the paper focuses on contrasting German and Dutch cases. After the recent economic crisis, the Dutch released more data while the German authorities maintained strict confidentiality rules. The design of deposit insurance schemes provides a key reason why the level of secrecy varies, with the Dutch move from an ex post to an ex ante scheme where the government served as the ultimate backstop leading to questions about the accounts of individual banks while the German system favoured continued secrecy. The paper also describes the level of transparency at the EU level. Multilevel legal restrictions and bureaucratic capacity tilt EU banking union practices towards member states that treat financial supervisory data as confidential.

Keywords:

• Secrecy
• European Union politics
• Germany
• Netherlands
• European Central Bank
• bank supervision

Notes

1. This may be the case for some US taxpayers who have money in Swiss bank accounts, it is unclear that such a sweeping conclusion is applicable overall (Masciandaro and Balakina 2015).

2. Note that we do not consider transparency and confidentiality to necessarily be at opposite ends of a common spectrum. This is because governments may choose neither option in that they do not release data, not because of an effort to keep it secret, but because of some other reason, such the non-salience of the issue. We do not explore the latter cases here.

3. We argue that there may not actually be an accountability dilemma in that the concerns about transparency undermining financial stability are unfounded (Gandrud and Hallerberg 2015: 775–6). They also argue that privacy concerns can be addressed. However, there may be political demands for a much higher level of privacy than we see as appropriate. In this paper we do not advance an optimal level of transparency, but instead explore the causes of choices to be transparent or secret.

4. See https://www.sec.gov/edgar/searchedgar/companysearch.html (accessed May 2017).

5. See https://www.gov.uk/government/organisations/companies-house (accessed May 2017).

6. A good overview of what the supervisor does in the United States appears in Eisenbach et al. (2015).

7. Interview conducted in November 2016 on condition of anonymity.

8. This example demonstrates how, just because a supervisor regularly releases a set amount of information to the public, a supervisor may not be completely transparent and non-secretive. The definitions of quantities of interest and the formats with which data are released can obscure the quantities that the public may be interested in when holding supervisors to account.

9. See https://www.eba.europa.eu/regulation-and-policy/supervisory-reporting/implementing-technical-standard-on-supervisory-reporting

10. See https://cdr.ffiec.gov/public/PWS/DownloadBulkData.aspx (accessed May 2017).

11. See http://www.eba.europa.eu/about-us;jsessionid=89D83640987D9D402057FBA25D6633F1 (accessed 21 February 2017).

12. See http://www.eba.europa.eu/about-us/missions-and-tasks (accessed May 2017).

13. See http://www.eba.europa.eu/risk-analysis-and-data/eu-wide-transparency-exercise/2015/results (accessed May 2017).

14. See https://capitalissues.co/2016/10/13/eba-lowers-its-bank-data-sharing-ambitions-dramatically/ both for screen shots of the two work programmes and for commentary on the change.

15. See https://www.eba.europa.eu/risk-analysis-and-data/eu-wide-transparency-exercise/2016/results

16. Note that technically the ECB has the ability to supervise all banks in the Eurozone, not just the largest ones, though in practice the ECB focuses only on the largest.

17. See https://www.bankingsupervision.europa.eu/banking/comprehensive/html/index.en.html (accessed May 2017).

18. This decision was foreshadowed in a speech by Sabine Lautenschläger, a member of the ECB Executive Board, in a speech on 18 November 2015. Discussing transparency, she noted that

‘We aspire to a kind of transparency which not only features abstract methods but which also shows the results of applying a consistent and holistic methodology to 123 banking groups ‒ a methodology which of course will be anonymised but which will be meaningful in showing trends’.

See https://www.bankingsupervision.europa.eu/press/speeches/date/2015/html/se151118.en.html

19. See https://www.bankingsupervision.europa.eu/press/conferences/sup_rep_conf/shared/pdf/Item1_1_AurelSchubert.pdf

20. See https://www.bankingsupervision.europa.eu/ecb/pub/pdf/supervisorybankingstatistics_third_quarter_2016_201701.en.pdf

21. Lithuania, which had data available, became a banking union country in 2015. The Netherlands, an original banking union member, became more transparent.

22. The title of William Greider’s 1987 book on the Federal Reserve Bank, Secrets of the Temple, fit the mood at the time.

23. See Woodford (2005) for a summary of the arguments.

24. The first such account is available here: http://www.ecb.europa.eu/press/accounts/2015/html/mg150219.en.html See also Blinder et al. (2008) for a survey and Born et al. (2014), for an assessment of the transparency of central banks on financial stability.

25. See http://www.bis.org/list/bcbs/sac_1/tid_31/index.htm (accessed May 2017).

26. See https://www.eba.europa.eu/regulation-and-policy/transparency-and-pillar-3 (accessed May 2017).

27. Quarterly statements dating back to 1918 are available at the St Louis Federal Reserve bank at https://fraser.stlouisfed.org/title/1007#38368

28. http://www.reuters.com/article/us-financial-fdic/fdic-seen-raising-premiums-extending-backstop-idUSTRE51M7MI20090223 (accessed 10 September 2017).

29. The Dutch government initially provided €4 billion for 49% of Fortis Holding Netherlands, while the German government organised a credit line of €25‒30 billion for Hypo Real Estate (HRE). The German economy is about 4.5 times the size of the Dutch one (see https://www.destatis.de/Europa/EN/Country/Comparison/GER_EU_Compared.html), so the two bailouts had approximately the same scale in terms of the respective economies. The estimates for the size of the bailouts at the time they were decided appears in http://www.spiegel.de/international/business/finance-crisis-europe-sees-three-bank-bailouts-in-two-days-a-581068.html (accessed 11 September 2017).

30. BaFin was created in 2002 through the merger of three agencies.

31. Section 9 covers Professional Secrecy Requirements.

32. The English text is available at https://www.bafin.de/SharedDocs/Downloads/EN/Aufsichtsrecht/dl_kwg_en.html (accessed May 2017). The relevant text of the section reads as follows:

‘Persons employed by BaFin ... and persons employed by the Deutsche Bundesbank, insofar as they are acting to implement this Act, shall not disclose or use without authorisation information they receive in the course of their activities and which should be kept secret in the interests of the institution or a third party (especially business and trade secrets), including after they have left such employment or their activities have ended.’

33. See https://www.bundesbank.de/Navigation/EN/Service/Freedom_Information_Act/freedom_information_act_node.html (accessed May 2017).

34. Author interview with supervisor, 24 April 2017.

35. The text reads:

‘Confidential information which such persons, auditors or experts receive in the course of their duties may be disclosed only in summary or aggregate form, such that individual credit institutions cannot be identified, without prejudice to cases covered by criminal law.’

Exceptions include results of stress tests and when an institution is being wound down. Available at https://www.eba.europa.eu/regulation-and-policy/single-rulebook/interactive-single-rulebook/-/interactive-single-rulebook/article-id/182 (accessed May 2017).

36. Author interview, 25 April 2017.

37. Author interview, 11 May 2017.

38. See https://www.bundesbank.de/Redaktion/EN/Downloads/Publications/Monthly_Report_Articles/2015/2015_12_deposit_protection.pdf?__blob=publicationFile (accessed May 2017).

39. See http://uk.reuters.com/article/uk-eu-deposits-germany-idUKKCN0RB23X20150911 (accessed May 2017).

40. p. 16, Credit Lost. Summary in English. Report of the Parliamentary Committee Inquiry Financial System Tweede Kamer der Staten-Generaal., 10 May 2010, http://www.europarl.europa.eu/meetdocs/2009_2014/documents/cris/dv/credit_lost_/credit_lost_en.pdf (accessed 22 February2016).

41. The parties represented were the Socialist Party (SP), People’s Party for Freedom and Democracy (VVD), the Labour Party (PvdA), Christian Democratic Appeal (CDA), Democrats 66 (D66), and Greenleft (GroenLinks). The original committee had one representative from each of these parties as well as from the Party for Freedom (PVV) and the Christian Union (CU).

42. see https://www.houseofrepresentatives.nl/news/final-report-‘credit-lost-ii-–-taking-stock’-presented

43. Email from Olaf Sleijpen, 20 February 2017. Cited directly with his permission.

44. Author conversation with senior DNB official, 22 February 2017.

45. Author conversation with senior DNB official, 22 February 2017.

46. See https://www.dnb.nl/en/about-dnb/consumers-and-dnb/consumers-and-supervision/depositogarantiestelsel/ (accessed May 2017).

47. See http://eur-lex.europa.eu/legal-content/en/TXT/?uri=CELEX%3A32009R0951 (accessed May 2017).

48. E.g. there is a focus in the Bank Resolution and Recovery Directive on ‘bail-in’ rules intended to force losses on creditors rather than national governments.

49. From http://www.eba.europa.eu/about-us/careers/multimedia/eba-at-glance (accessed May 2017).

50. The ECB is in fact required to inform national regulators what it intends to do with information it requests (EC Regulation 951/2009, article 8(2)).