Correspondence: Roger Debreceny, University of Hawaìi at Mnoa, School of Accountancy, Shidler College of Business, 2404 Maile Way, BusAd C306, Honolulu, HI 96822, USA

¹is Project Manager for International Accounting Standards Committee Foundation where he is responsible for the IFRS and XBRL-related projects. His main areas of activities include technological aspects of the IFRS Taxonomy development. His doctoral dissertation, 'XBRL Financial Reporting Supply Chain Architecture', received the Outstanding Dissertation Award from the Strategic Emerging Technologies Section of the American Accounting Association.

²is Full Professor at Technische Universität Bergakademie Freiberg, Germany. He holds the Chair of Management Information Systems and teaches Business Intelligence and IT Governance. Main topics of his research are business analytics, data warehousing, XBRL and IT maturity models. He is member of the XBRL Germany strategy advisory board and works at the Business Intelligence Research Center and TDWI Germany.

³is a PhD candidate at the Dresden University of Technology. He holds a masters degree in economics with a major in Management Information Systems. His main research activities are XBRL, XBRL Dimensions (XDT), Standardisation, semantic technologies and multidimensional Modeling'.

⁴is the Shidler College Distinguished Professor of Accounting at the Shidler College of Business, University of Hawai`i at Mnoa. His main research areas are in Internet Financial Reporting, XBRL, IT governance and accounting information systems.

Received 8 April 2009; Revised 8 April 2009.

Abstract

The eXtensible Business Reporting Language (XBRL) is an XML-based, de facto standard for the dissemination of metadata associated with business reporting information. The standard allows parties to build reporting solutions within a wide variety of information value chains. The standard provides considerable flexibility in the way those parties construct those services. The XBRL specification provides a framework upon which those that wish to develop an XBRL-based reporting environment then build. XBRL taxonomies provide metadata about reporting concepts. Instance documents, aligned with those taxonomies, report facts and provide entity-specific metadata. The XBRL specification constrains taxonomy design but within these constraints, developers have very wide discretion. This inherent flexibility creates challenges to information providers, intermediaries and consumers. Designers must balance flexibility, transparency, reliability and cost. We provide four XBRL use cases to illustrate these challenges. We further analyse these issues in a case study of the prudential supervision of financial institutions in Europe with the COREP framework. The COREP taxonomy was the first to employ the dimensional XBRL specification. The dimensional specification allows for the reporting of data in a hypercube modelled in a taxonomy. The hypercubes in the COREP taxonomy allow for breakdowns of solvency reporting into their constituent structures. The COREP hypercube structures are highly complex. Meeting objectives for flexibility, national prudential regulators are able to make national extensions to the foundational COREP taxonomy. As a result, we observe considerable national variation in calculation relationships and in dimensional structures. Radically different data models contribute to the inconsistencies among the various national taxonomies within the overall COREP framework. We show how this flexibility in multiple taxonomy design across different European countries has created challenges for information providers and difficulties for information consumers seeking to assess institutions across national borders. We make recommendations on the balancing of needs for flexibility and data compatibility within taxonomy design.