1 Standardization through XBRL Why XBRL? XBRL = open standard for - - PDF document

1

Administrative simplification structural business statistics survey, NSI Belgium Heerlen, 23-03-2011

The structural business statistics

Structural business statistics?

• Yearly survey on the structure of enterprises (costs,

profits, investments,…)

• Asks for accounting data

Administrative burden:

– AB=((Ttime x Pprice) + OOP) x Qfrequency x Qnumber) – AB= ((5.01 x €32) + 0) x 1 x 40,000) – AB= €6,412,800 or +/- €160 per actor

SBS survey simplification

Study Federation Of Enterprises in Belgium (2007)

% burden

SBS survey simplification History

– Initially (1995): paper only – In 2002: first administrative simplification

• Deleting sections that are asked for twice (annual

accounts, NSSO,...)

• Limited number of documents could be transferred by

floppy disk to NSI – Result simplification:

• Up to 40% of sections were deleted
• +/- 92% of declarations on paper and 8% on floppy disk

SBS Survey Simplification 2008

• Administrative simplification techniques

– Gold-plating

• Checking to what extent national surveys are in line

with European statistical demands – Questioning fewer enterprises

• The threshold for exhaustive surveying has been

increased

• A rotation scheme was introduced for SME’s,

estimation model for non – surveyed enterprises

• Administrative fines for non – responding enterprises

– Avoiding double questioning

• Figures known from annual accounts are prefilled in the

questionnaire

• Checks whether sum of subheadings equals the

prefilled total from the annual account

SBS Survey Simplification 2008

– Changing survey frequency

• Decrease frequency of the annexes
• negative coordination in the rotation scheme allows

better spread across target population – E-government

• Can we simplify the B2G information transfer and can

we do so without burdening the enterprise too much?

• XBRL: Electronic reporting language based on XML

(open standard) for the exchange of company data through the internet, limiting manipulation as much as possible

2

Why XBRL? XBRL = open standard for electronic exchange of information. A lot of the information required by statistical surveys is available in the enterprises information system (eg accountancy data, data on personnel, …) We need matching between the information in the enterprises information system and the statistical information required, XBRL allows this matching XBRL in Belgium already being used by NBB, taxes administration, NSI, federal planning agency

Standardization through XBRL

SBS survey simplification

What is XBRL? XBRL works with taxonomies

– = Type of dictionary in which elements are uniquely defined and additional characteristics added – Technically a taxonomy is a combination of xml files – The schema file contains the unique element declarations and basic characteristics – Linkbases add more characteristics to the elements:

• Label linkbase: different labels in different languages
• Reference linkbase: references
• Presentation linkbase: lay - out of documents
• Calculation linkbase (formula linkbase): order of
• perations
• Definition linkbase

SBS survey simplification

XBRL taxonomy editor screenshot

Data collection based on XBRL Three answering methods:

– An application automatically constructs a websurvey from the XBRL taxonomy, enterprises can fill in the questionnaire online – An upload procedure to upload XBRL instances created from the enterprises information system – The paper questionnaire still exists

3

Results

Results: enterprise characteristics by declaration method

Median Turnover Median no employees % filled in by external accountant Mean session length Paper 495 080 6 44% / Websurvey 1 627 493 12 35% 17 mins Upload procedure 2 558 772 14 51% 8 mins

Results: response over time

Results

13.6 31.3 Staff (in FTE) 4,282,800 6,412,800 Cost for enterprises 7 11 Data treatment process (in months) 10% 92% Paper questionnaires 90% 8% Electronic data transfer 40% (95%) 40% Deletion-pre-fill (upload) 28,000 41,000 Sample size After simplification Before simplification Summary

Plans for the future Horizontal integration of the data collection process (organization per process instead of per survey) All web surveys based on 3 tools: XBRL, Blaise and Java One declaration platform for all web surveys Expand the use of XBRL to other surveys (structure

• f earnings survey, tourism survey, transport of

goods over the road survey) Upgrade the XBRL web application to include XBRL dimensions