What happens when a Big 4 firm publishes a document about IFRS and brings together "experts" to discuss the "quest for a global language" and opine on XBRL? And what does it tell us about expectations - the expectations the profession has, and - by extension - the expectations and understand that the market may have?

The following comments do not represent those of my employer; the document I am referring to is also not published by my employer. I have not provided a link to the underlying document, but I know from some of the phrases I have provided, you can easily use a search engine to find it.

By the way: I am glad to hear such esteemed members of the profession and the financial community discuss XBRL. These are only opinions being offered on both sides. But I think people are missing a much larger picture. Business reporting supply chains are being revolutionized - read up over at http://www.xbrl-ntp.nl, for example - and it isn't just because it is easier to find numbers in a "spreadsheet" (as one of the experts put it, "‘XBRL is just like a giant spreadsheet for data.")

One "leading commentator" says:
‘XBRL looks very desirable. But all it is really doing is saving me the hard graft of digging out the detail myself. It doesn’t change the way the numbers were produced in the first place’.

EEC says:

• In the short-term - isn't there value in saving you that "hard graft"? If XBRL reduces the effort to type numbers into your models (or into your audit tools) by 60-90%, isn't that an insanely great thing - especially if it is the company's numbers as originally stated, and not normalized numbers, recrafted by a data infomediary?
• In the mid-term, you aren't the only one digging out detail. Perhaps as many other people ... like ANALYSTS ... find the digging easier, it will lead them to ask more questions and force the improvement of the production of the number. Or maybe - as in the Netherlands - the underlying reporting is being changed, facilitated by XBRL. What if XBRL helped facilitate a move from national GAAP to IFRS or US GAAP? Or if it pushed financial and tax reporting to converge, simplifying the reporting process and letting a company spend more time making sure its common filings properly represent operations instead of having to file the same numbers in different ways on a multitude of forms?
  
• In the mid- to long- term, XBRL isn't just about standardizing the end reporting. It is, and has always been, about integrating the entire business reporting supply chain. XBRL's Global Ledger is about improving the production of the numbers - the data integration, consolidation, migration and archival - and perhaps laying down a path where today's internal reporting will increasingly be tomorrow's external reporting. Let's work on encouraging the accounting and ERP vendors to move towards standardized detailed data for better integration and sharing with advisors and auditors, a more powerful controls environment, and a seamless audit trail, from first transaction to end reporting.
  

A "group finance director" doubts:
that technologies like XBRL have yet found a place in the financial reporting world. ‘It is a question of what should the frequency of reporting be?’ he said. ‘It would simply increase the possibility of manipulation’. ‘Let’s fix half-yearly reporting first’.

EEC says:
Wow. Blaming XBRL with the assumption it is all about more frequent reporting? And questioning a global pressure for more frequent reporting, with or without XBRL? And saying there is more change for manipulation with XBRL than with text and graphics?

When SOx 409 is a first step in a move that the SEC and Congress have stated is a push toward real-time reporting, when similar or greater real-time disclosure efforts are already in place in Australia and Canada and on the way in Japan, this is an interesting statement.

1. It seems to say that XBRL is the push toward greater frequency, rather than a response to it. If someone has a better tool to cope with the greater periodicity, please let me know.
2. It assumes that XBRL will lead to more manipulation that today's HTML/PDF model
3. It assumes that XBRL means no one will be working to "fix" half-yearly reporting; they aren't mutually exclusive. But the market has always relied on rumors, gossip and innuendo, and "fixing" half-yearly reporting won't get better current data to the market today.
   

That same person goes on to say:
I’d rather focus on getting the narrative reporting right than follow the XBRL approach’. It is down to the culture again. ‘Will XBRL information from a company in Malaysia say the same thing as the information from a company in France?’

EEC says:
An XBRL approach will facilitate getting that narrative reporting improved. Have you ever tried to pull a particular note from a half dozen companies' annual reports and trying to compare components of a particular disclosure? I have. It takes a while on paper. Take that back to 3 years for each of the companies. How do they change from year to year for a specific company? How do they compare between companies? The NASDAQ Investor's Assistant showed five years ago that XBRL made this discovery and comparison "child's play".

Experience also seems to prove that "getting the narrative reporting right" doesn't matter an awful lot if the market isn't reading the annual reports, but doing its work based on information from infomediaries. XBRL can actually lead to more of the ability to "tell your own story" rather than less - in practice.

Of course, XBRL does not force "Sales" as reported by Automobile manufacturer F to have the same components as the "Sales" as reported by Automobile manufacturer G. But XBRL can help overcome the format differences, the language differences, and bringing together the like disclosures so the market can work toward reconciling them, and asking them to be more consistent. Being able to FIND the same disclosure across languages is the first hurdle - look at http://www.iasb.org/xbrl/translations/downloads_2006.html and you will see that XBRL tags will lower the bar in half a dozen languages TODAY.

My musings

For the profession to provide assurance on XBRL, the market needs to have appropriate expectations on what XBRL can and cannot do. When the "experts" say they are skeptical about XBRL because they miss market demands and activities - that can be a challenge.

What expectations do you have about XBRL? Theoretically, an XBRL filing is no more - and no less - a fair view than a parallel paper filing on which it is based, if such paper-paradigm filings came first. In this short term, an XBRL filing should be like a mirror, faithfully reproducing beauty and blemishes alike. Then the market will act based on those filings, which are easier to find, consume and compare.

Data Level Assurance - it is a phrase that brings fear to some and hope to others. But DLA is probably NOT what you think it is, especially if you are an auditor today.

In my study for my presentation at last November's Continuous Auditing conference (http://raw.rutgers.edu/12thwcas/Papers/Eric%20E.%20Cohen/Cohen_12thWorldCA_06_prelim.pdf), where I compared business reporting to physics (periodic reporting and audit is to continuous reporting and audit as classical physics is to quantum physics), I learned that "explaining quantum physics to someone with no physics background may be easier than explaining it to someone highly trained in classical physics).

Data Level Assurance is not the auditor taking on the risk of saying that individual reporting items can be trusted without "taken as a whole" or "materiality" in today's model of reporting. In fact, with XBRL and XML Signature, DLA can lead to a reduction in risk, rather than an increase.

Werner Heisenberg, were he an auditor, may have said, "I can assure the Balance Sheet or the Income Statement, but not both." And auditors would have always looked into the audit room to make sure that Schrödinger's cat wasn't there before closing the door. By observing the physical inventory count, the auditor changes it; some might say until the auditor observed it, there was no count going on, at least in Copenhagen.

And never forget, the cat fell off the roof because it lost its mu.

Much is happening with XBRL and assurance.

Discussions underway

US

We have had many discussions on the topic within XBRL circles. You can find some of those discussions in PowerPoint form in the past events section of the XBRL web site: http://www.xbrl.org/PastEvents/

Under the umbrella of the AICPA, an interpretation of AT 101, the Attestation standard, was released. The PCAOB used that interpretation as part of its "Staff Questions and Answers: Attest Engagements Regarding XBRL Financial Information Furnished Under the XBRL Voluntary Financial Reporting Program on the Edgar System (May 25, 2005)" found at http://www.pcaob.com/Standards/Staff_Questions_and_Answers/2005/05-25%20.pdf

Two attestation engagements on XBRL as part of the SEC Voluntary Filing Program have taken place; those attestation reports can be seen as part of the XBRL filings found at http://www.sec.gov/Archives/edgar/xbrl.html and most recently at http://www.sec.gov/Archives/edgar/data/101829/000119312506247769/0001193125-06-247769-index.htm

IFAC

For those subject to international auditing rules, the XBRL Assurance Working Group released a white paper, “INTERACTIVE DATA: THE IMPACT ON ASSURANCE, NEW CHALLENGES FOR THE AUDIT PROFESSION,” which can be found at http://www.xbrl.org/Announcements/Interactive-Data-Assurance-2006-11-10.pdf - this paper brings up ISAE 3000, which is very similar to the US Attestation standard.

Upcoming events:

The next two months will be quite active with events where XBRL assurance will be discussed -
the University of Kansas is holding a conference on the topic (www.eycarat.ku.edu/Conf2007/Program2007Conf.pdf ), and it will be on the agenda at XBRL International Conference and the Ferrara Continuous Auditing Conference.

It has been a tremendously long time since I began this "Assurance and XML" Blog (ok, and a long time since my first couple of entries). Now, with the SEC Pilot project for XBRL on course to begin soon, the topic of assurance around XBRL instance documents is on the front-burner.

When Wayne Harding (www.wayneharding.com) first told me that the project that has come to be known as XBRL (www.xbrl.org) would revolutionize accounting and audit, I believed him and have been involved from its foundation. XML moved into systems quickly. Now there are issues about assurance around XBRL, assurance around XML used in business operations leading tofinancial reporting, assurance that can be facilitated using XML derivatives like XBRL GL (www.xbrl.org/GLTaxonomy) ...

and yet assurance and XML discussions are still very limited. Regular searches on the excellent Auditnet.org web site bring up little in the area of any audit checklists related to XML. Discussions on Data assurance and Data Level Assurance are limited to a few academicians.

The conversation and collaboration in this area must move quickly forward. More to come.

Is that file actually from who it says it is? It's not as simple as a traditional annual report - they cost lots of money to make and lots to send. Today's financial information can be found on a company's web site, quoted in a press release, stored in a repository, quoted accurately or inaccurately on USENET or a chat room. What is the actual question today? Perhaps: "Are the business facts represented in this file the original assertions of the organization being reported upon, the opinion of the (second hand) publisher, or some other choice?" A digital notary is a start for some assurance that the file is indeed from the indicated source ... but was that source authorized to make the disclosure?

How can you know that the information you find on the Internet is reliable? Is the Certified Public Accountant/Chartered Accountant the trusted, independent third party whose digital signature can offer you comfort? This Blog will follow musings about information on the Internet, the goals of the U.S. Securities and Exchange Commission, XML, XBRL and other information that will lead to a reliable part down the Internet. Find out about Blogging at Blogger