by Ashwini Robin | December 22, 2015
It starts with a simple question.
What is the best way of using the governing taxonomy for a companyÂ’s compliance reporting in XBRL?
And while there are many responses to this question, it all boils down to the use of extensions. Prima facie, filers are required to to fit their financials under available elements suggested by the governing taxonomy. But while this provides for standardization of data across filers, there is a fear of losing the full picture that the filer wants to communicate.
HereÂ’s where extensions come in handy. Extensions allow filers to Â‘expandÂ’ the taxonomy, to fit in their reported figures in cases where standard elements of the taxonomy are not available. But while this can help communicate the complete message to the stakeholder, it takes away from standardization in the process.
This is a problem best defined by the oxymoron Â‘customization of standardized taxonomyÂ’.
The idea behind allowing extensions was to provide filers with the flexibility they needed to share their business- specific information that could not have been accommodated within the existing taxonomy.
However, this flexibility makes way for the very problem that XBRL aims to eliminate: non-comparability of data. So while machines now can read and process the XBRL data individually, Â‘extensionsÂ’ limit the possibility of instant comparison of data across companies.
Finding a Happy Middle: IRIS Data Consumption Platform (DCP)
At IRIS, we value the benefits of both data comparability as well as granularity of company-specific reporting. Here is how we have done justice to both in the IRIS Data Consumption Platform( DCP) - our rich, structured data repository of global financial and non-financial data of public and private companies, available both in Â‘as reportedÂ’ and Â‘normalizedÂ’ forms.
For normalization, we have created a normalized taxonomy: the IRIS Standard Taxonomy or IST that sits above the regulatorÂ’s taxonomy. Initially prepared for the US GAAP Taxonomy (UGT), IST aims to eventually cover all XBRL taxonomies to enable global comparison of data tagged in XBRL format.
The IST has around 3000 elements when compared to the UGT. To make IST comprehensive for US company analysis, in addition to the UGT elements, it includes a standard set of extensions that allow for data comparison. As a further refinement, we have removed duplicate elements such as alternate methods of tagging the same concepts. For example, class of assets in fair value footnote / disclosure which can be represented in two different ways in the US GAAP taxonomy Â– dimensionally and non-dimensionally is accounted for only once in the IST- dimensionally.
IRIS DCP enables mapping of US company-specific taxonomies to the IST which provides a much deeper level of comparability at the disclosure / footnote level, making it easier for analysts to consume data. Not only are analysts benefitted but corporates also get access to peer-data straight from the source (SEC filings), thus enabling competitive peer benchmarking.
LetÂ’s use an example to demonstrate this - if you want to compare the debt profiles of Intel and AMD, currently this is how their SEC filings would look like:
Company name and ticker: INTEL CORP (Filer), CIK: 0000050863
Company name and ticker: ADVANCED MICRO DEVICES INC (Filer), CIK: 0000002488
Though the senior notes data is disclosed and tagged by both the companies, they have extended it in their own unique ways. This immediately reduces the comparability and usefulness of XBRL data.
IRIS DCP offers the ability to create groups to enable better comparison of these concepts. The IST provides for debt instruments to be mapped to characteristics such as senior notes/ subordinated debt, secured / unsecured debt, fixed rate / floating rate debt.
In the same example, with IRIS DCP, all senior notes for Intel and AMD are mapped to iris_SeniorNotesMember. This allows users to easily compare the debt profile for both companies as shown:
Further, IRIS DCP retains a link at all times between the IST and the Â‘as-reportedÂ’ document filed by the company. It is, therefore, possible to trace back values to the original source, as the companies have reported it.
Thus, IRIS DCP maintains a fine balance between standardization and allowing for global comparisons while retaining the uniqueness of the data reported by each filer.