Adopting XBRL in the Financial Statements Data Pool of Deutsche Bundesbank

1
Adopting XBRL in theFinancial Statements Data
Poolof Deutsche Bundesbank

• Dominik Elgg
• Deutsche BundesbankStatistics Department

2
Contents

• I. The Financial Statements Data Pool
• 1. The project
• 2. Benefits from pool
• 3. Considered data and benchmark statistics
• 4. Participation and status
• II. XBRL as data format
• 1. Information chain
• 2. Issues and problems - Solution through XBRL
• 3. Special requirements and examples
• 4. Status and outlook
• III. Advantages and challenges from a technical
  and programming perspective.

3
I. Financial Statements Data PoolThe project

• Joint project of Bundesbank and German financial
  institutes
• Idea Pool of financial statements which is
  filled by several contributors
• Project rests upon the principles
• Voluntary participation
• Reciprocity
• Data protection (anonymous data)
• Low costs for participants.

4
Benefits from pool

• Participants extensive benchmark statistics
  (ratios)
• Advantage of enlarged dataset
• Bundesbank continuation and extension of
  corporate financial statements statistics
• Dataset is basis for micro- and macroeconomic
  analyses
• Providing information (statistics) to the public
• Compensation for reduction of Bundesbanks own
  dataset.

5
Considered data

• Financial statements of German companies
• Only individual accounts (no consolidated
  accounts so far)
• Basically data on income statement and balance
  sheet
• Non-financial companies
• Only based on German Commercial Code.

6
Benchmark statistics (financial ratios and
differentiation)

• More than 180 ratios
• Preparation by sector, size, legal form, region
  and year
• Differentiation as far as data protection allows
• Quartiles as statistical parameters
• Quarterly update of the statistic
• Calculation of ratios for nearly 70.000 groups
  per year.

7
Example
8
Participation

• Banks
• Credit insurers
• Datev
• Other institutes with considerable datasets
• Bundesbank
• Current participation 9.

9
Current status

• Database started 4th quarter 2003
• 125.000 financial statements per year (gross)
• Deduction of duplicates
• Deduction of deficient data
• More than 85.000 financial statements per year
  (net)
• Shipment of 4 benchmark statistics so far.

10
II. XBRL as data format

• Two different possibilities for sending data to
  Bundesbank
• XBRL
• Alternative plain text format.

11
Information chain
Bundesbank

• Partner
• Banks
• Credit insurers
• etc.

Statistics

• Ratios
• Cash Flow
• Profitability
• Leverage
• ...
• ... by legal form
• ... by sector
• ... by size
• ... by region
• ... by year

POOL
Financial statement
Financial statement
Financial statement
Financial statement
Financial statement
Database
Financial statements of German non-financial
companies
12
Issues and problems

• Participating institutes have different recording
  systems
• Partners have different guidelines for keying
  data into the recording system
• Bias in the data
• Problems in the identification process of
  duplicates.

13
Why XBRL?

• Standardised, non-proprietary format for
  exchanging financial statements data
• Data which comes directly from companies is
  qualitatively better than data that is manually
  keyed in into the recording system
• In the long-term XBRL might lead to the same data
  scheme and guidelines, as a basis for analysing
  financial statements.
• Scheme of pool dataset (relational database) is
  based on German taxonomy.

14
The issue of data quality - examples

• Several data sources with different data quality
• Strict quality checks
• About 10 loss of data in historic data due to
  big errors, maximum of 5 in actual data
• Thousands of minor errors
• With the XBRL test data we received so far 0
  errors.

15
Specific requirements on data format

• Anonymous data (data protection)
• Separation of characteristics (e.g. legal form)
  from accounting entries (data protection)
• Add-on items not represented in taxonomy
• Creation of an extended XML/XBRL framework while
  keeping the advantages of XBRL.

16
Example XBRL Instance Document
lt?xml version"1.0" encoding"iso-8859-1"
standalone"no"?gt ltxbrl xmlns"http//www.xbrl
.org/2001/instance"
xmlnsxsi"http//www.w3.org/2001/XMLSchema-instan
ce" xmlnst"http//www.xbrl.org/g
erman/ap/ci/2002-02-15"
xsischemaLocation"http//www.xbrl.org/german/ap/
ci/2002-02-15 german_ap.xsd"gt
ltnumericContext idnc1 precision5"
cwa"false"gt ltentitygt
ltidentifier scheme"www.dax.de"gtABC
S.Alt/identifiergt
lt/entitygt ltperiodgt
ltstartDategt2000-01-01lt/startDategt
ltendDategt2000-12-31lt/endDategt lt/periodgt
ltunitgt ltmeasuregtISO4217EURlt/me
asuregt
lt/unitgt
lt/numericContextgt lttbs.ass
numericContext" nc1"gt11000000lt/tbs.assgt
lt!-- Total Assets--gt lttbs.ass.fixAss
numericContext" nc1"gt8000000lt/tbs.ass.fixAssgt
lt!-- Non current assets --gt
lttbs.ass.currAss numericContext"
nc1"gt2000000lt/tbs.ass.currAssgt lt!--
Current assets--gt ... lt!--
... --gt lttbs.eqLiab numericContext"
nc1"gt11000000lt/tbs.eqLiabgt lt!-- Total
equity liabilities --gt
lttbs.eqLiab.equity numericContext"
nc1"gt3000000lt/tbs.eqLiab.equitygt lt!-- Equity
--gt lttbs.eqLiab.liab
numericContext" nc1"gt7000000lt/tbs.eqLiab.liabgt
lt!-- Liabilities --gt ...
lt!-- ... --gt lt/xbrlgt
17
Accounting entries
lt?xml version"1.0" encoding"iso-8859-1"
standalone"no"?gt ltxbrl xmlns"http//www.xbrl
.org/2001/instance"
xmlnsxsi"http//www.w3.org/2001/XMLSchema-instan
ce" xmlnst"http//www.xbrl.org/g
erman/ap/ci/2002-02-15"
xsischemaLocation"http//www.xbrl.org/german/ap/
ci/2002-02-15 german_ap.xsd"gt
ltnumericContext idnc1 precision5"
cwa"false"gt ltentitygt
ltidentifier scheme"www.dax.de"gtABC
AGlt/identifiergt
lt/entitygt ltperiodgt
ltstartDategt2000-01-01lt/startDategt
ltendDategt2000-12-31lt/endDategt lt/periodgt
ltunitgt ltmeasuregtISO4217EURlt/me
asuregt
lt/unitgt
lt/numericContextgt lttbs.ass
numericContext" nc1"gt11000000lt/tbs.assgt
lt!-- Total Assets --gt lttbs.ass.fixAss
numericContext" nc1"gt8000000lt/tbs.ass.fixAss
gt lt!-- Non current assets --gt
lttbs.ass.currAss numericContext"
nc1"gt2000000lt/tbs.ass.currAssgt lt!--
Current assets --gt ... lt!--
... --gt lttbs.eqLiab numericContext"
nc1"gt11000000lt/tbs.eqLiabgt lt!-- Total
equity liabilities --gt
lttbs.eqLiab.equity numericContext"
nc1"gt3000000lt/tbs.eqLiab.equitygt lt!-- Equity
--gt lttbs.eqLiab.liab
numericContext" nc1"gt7000000lt/tbs.eqLiab.liabgt
lt!-- Liabilities --gt ...
lt!-- ... --gt lt/xbrlgt
18
Context
lt?xml version"1.0" encoding"iso-8859-1"
standalone"no"?gt ltxbrl xmlns"http//www.xbrl
.org/2001/instance"
xmlnsxsi"http//www.w3.org/2001/XMLSchema-instan
ce" xmlnst"http//www.xbrl.org/g
erman/ap/ci/2002-02-15"
xsischemaLocation"http//www.xbrl.org/german/ap/
ci/2002-02-15 german_ap.xsd"gt
ltnumericContext idnc1 precision5"
cwa"false"gt ltentitygt
ltidentifier scheme"www.dax.de"gtABC
AGlt/identifiergt
lt/entitygt ltperiodgt
ltstartDategt2000-01-01lt/startDategt
ltendDategt2000-12-31lt/endDategt lt/periodgt
ltunitgt ltmeasuregtISO4217EURlt/me
asuregt
lt/unitgt
lt/numericContextgt lttbs.ass
numericContext" nc1"gt11000000lt/tbs.assgt
lt!-- Total Assets --gt lttbs.ass.fixAss
numericContext" nc1"gt8000000lt/tbs.ass.fixAss
gt lt!-- Non current assets --gt
lttbs.ass.currAss numericContext"
nc1"gt2000000lt/tbs.ass.currAssgt lt!--
Current assets --gt ... lt!--
... --gt lttbs.eqLiab numericContext"
nc1"gt11000000lt/tbs.eqLiabgt lt!-- Total
equity liabilities --gt
lttbs.eqLiab.equity numericContext"
nc1"gt3000000lt/tbs.eqLiab.equitygt lt!-- Equity
--gt lttbs.eqLiab.liab
numericContext" nc1"gt7000000lt/tbs.eqLiab.liabgt
lt!-- Liabilities --gt ...
lt!-- ... --gt lt/xbrlgt
19
Extended XML/XBRL framework(accounting entries)
lt?xml version"1.0" encoding"ISO-8859-1" ?gt -
ltfinancialStatementDelivery xmlns"http//www.bund
esbank.de/spool/datenlieferung/2003-01-01"
xmlnsxsi"http//www.w3.org/2001/XMLSchema-instan
ce" xmlnsxbrli"http//www.xbrl.org/2001/instance
" xmlnst"http//www.xbrl.org/german/ap/ci/2002-0
2-15" xmlnsISO4217"http//www.iso.org/4217"
xsischemaLocation"http//www.bundesbank.de/spool
/datenlieferung/2003-01-01 spool.xsd"
partnerName"Testpartner" partnerId0000000"
creationDate"2003-01-01" deliveryId"1"gt -
ltfinancialStatement companyId"1"
yearMonth"2000-12"gt ltreportTypegthlt/reportTypegt
ltreportStatusgtelt/reportStatusgt
ltincomeStatementTypegtglt/incomeStatementTypegt
ltcurrencygtDEMlt/currencygt - ltxbrlixbrlgt -
ltxbrlinumericContext id"nc1" precision"8"
cwa"false"gt - ltxbrlientitygt ltxbrliidentifier
scheme"urnwww.partner.de/"gt3,1/3lt/xbrliidentifi
ergt lt/xbrlientitygt - ltxbrliperiodgt
ltxbrlistartDategt2000-01-01lt/xbrlistartDategt
ltxbrliendDategt2000-12-31lt/xbrliendDategt
lt/xbrliperiodgt - ltxbrliunitgt
ltxbrlimeasuregtISO4217DEMlt/xbrlimeasuregt
lt/xbrliunitgt lt/xbrlinumericContextgt
lttbs.ass numericContext"nc1"gt1338066lt/tbs.assgt
lttbs.ass.accountingConvenience
numericContext"nc1"gt0lt/tbs.ass.accountingConveni
encegt lttbs.ass.accountingConvenience.changeDem2E
ur numericContext"nc1"gt0lt/tbs.ass.accountingConv
enience.changeDem2Eurgt lttbs.ass.accountingConven
ience.startUpCost numericContext"nc1"gt0lt/tbs.ass
.accountingConvenience.startUpCostgt
lttbs.ass.currAss numericContext"nc1"gt749385lt/t
bs.ass.currAssgt lttbs.ass.currAss.cashEquiv
numericContext"nc1"gt259760lt/tbs.ass.currAss.cash
Equivgt
20
Status of XBRL project in data pool

• Implementation completed
• Successful tests with test data and authentic
  data
• As the first partner, Datev will send XBRL data
  from this quarter on.

21
III. Advantages and challenges

• Pros and Cons from a programming perspective
• Possible validation on sender side
• Flexibility of XML
• Lower effort of programming
• Minor error rate
• XBRL as non-proprietary format
• Current Hybrid format
• Some familiarisation with XBRL/XML
• Dependency on external XBRL-specification
• File size.

22
Advantages from an accounting perspective

• Structured data instead of paper or PDFs
• Non-proprietary standard
• Separation of content from presentation
• Flexibility taxonomies can be created
  individually
• Independent of information type (e.g. accounting
  standard)
• XML as basis is rather easy to understand
• Knowledge transfer taxonomies can be created
  together (e.g. 1001 working groups filled by
  auditors, data receivers, software companies).

23
Challenges from an accounting perspective

• Dependency on an independent organisation (e.g.
  enduring specification)
• Dependency on software (companies)
• Sender has to be convinced of using XBRL.

24
Outlook

• Integration of IFRS financial statements
• Relevant for individual accounts from 2005 on
  (2006).
• Potential integration of consolidated accounts
• Use of IFRS (and SME-IFRS) taxonomy
• Use of results of III WG on IFRS impact and CBSO
  databases (group of Manuel Ortega - Banco de
  España)
• Use of potential extensions of German
  jurisdiction.

25
Vision for the next 5 to 10 years

• XBRL is used by all participants sending and
  receiving
• 10.000 to 20.000 financial statements of German
  companies based on SME-IFRS that can be compared
  to their European counterpart.

26

• ???
• Questions Answers
• !!!

27
Contact information

• Dominik Elgg
• DEUTSCHE BUNDESBANK
• Financial Statements Data Pool
• Statistics Department
• Tel. [phone]
• Email dominik.elgg_at_bundesbank.de

28

• ltendgt
• Thank you very much!
• lt/endgt