So now the hard work really begins. The SEC's decision today to require registrants (ie: companies with listed securities) to start filing their financial statements in XBRL format is pretty similar to the proposed rule announced in the middle of the year. As widely expected, there is a slip in the timing. It will start for company filings (quarterly or annual) as from 15 June 2009, ie: 2009-Q2 10Qs instead of the larger and harder 2008-Q4 10Ks as originally anticipated. They've also pushed back the timing for the roughly 8,000 mutual fund risk/return summaries to 1 January 2011, instead of the middle of 2009.

The Commission is not unanimous on the liability arrangements. Commissioner Aguilar is voting against the entire proposal because he wants the liability and the assurance arrangements contained in the proposals greatly strengthened.

For what it's worth, I think that the liability carve-out arrangements are an attempt to ease the transition costs - the change management involved in a complex new technology - and the SEC (particularly the Chairman) are relying on new SEC leadership to reinstate them. We need to wait for the final rule to see the details, but there appears to be a phase-in arrangement for liability (and therefore assurance) in any event. This should be a relief for a lot of auditors.

Commissioner Aguilar is pointing out that the current economic climate is the exact moment at which the regulator could simply reject the need for a carve-out. Since the requirements for audited financial statements in ASCII or HTML are not going away, and the liability and assurance arrangements for them remain, I suspect the issue is almost moot.

However, I think, from an accuracy and consistency perspective, we should be looking for XBRL filings to be in iXBRL (embedded inside web pages) format. This should happen very quickly and at that point, the liability arrangements must be uniform. Cox seems to have made this point, right at the end of the meeting.

A number of the other details will be in the rule. Final SEC rules tend to be published a week or so after these open meetings. Stay tuned.

I think this decision today is much bigger than most people realise. Thanks and congratulations to the SEC, particularly Paul Wilkinson, Jeff Naumann, David Blaszkowsky and (more recently) Walter Hamscher, all those involved in the US GAAP XBRL taxonomy project at XBRL US under Mark Bolgiano and Campbell Pryde, as well as the extraordinary international collaboration that XBRL continues to represent, especially at XBRL International.

Ever sat in a traffic jam in the evening, on a road that is narrowed by a lane filled with orange cones and silent bulldozers, as the workers have all gone home? I sit there fuming, calculating in my head the cost of all the additional fuel that is being burnt by cars moving at a snail's pace, plus the average cost of everyone's time. Surely, if government was "by the people, for the people", the cost/benefit analysis for work on major roads would include those costs. The result would be that an awful lot of road work would happen 24x7.

The reality is that not that many parts of government work in a way that takes account of people's time and money. But here's an area that does: S.B.R.

Standard Business Reporting projects are based on the unusual notion that government should change the way that it works so that things are just a little bit easier for the private sector. We've recently added the Australian Government's SBR project to our list of major clients. Their counterparts in the Netherlands have been important customers for some time.

What are they up to? They are working across government agencies, and even across levels of government, harmonising the way that business reports to government. The problem is that government is complex, and it needs lots of information from the private sector in order to be able to operate. But the provision of that information is complicated, burdensome and duplicative. How burdensome? Well, according to the SBR projects' analyses, about 2.5% of GDP in most countries is devoted to compliance costs. Two point five per cent??? That's USD48 Billion in the UK, or a gobsmacking USD333 Billion in the USA. In the Netherlands and Australia, both countries are dealing with smaller economies, with the burden in the order of USD12 Billion each, and both looking to save around 8% of that. Which is still really worth doing.

So, SBR aims to shave a bit of that burden down. Even small improvements, when you are talking international telephone numbers, really help!

How many government initiatives can say that they are working to save taxpayers hundreds of millions without cutting back on the services being provided?

As one senior official in the UK put it recently, rather than dealing with the symptoms of duplication and complexity, for instance, by trying and often failing to create cross-governmental databases, SBR aims to deal with the root of the problem, by getting different agencies to agree on definitions. This results in the construction of a single set of structured requirements that are much easier for small, medium and large businesses to comply with, and (crucially) for their accounting software to report with. The technology behind these initiatives? You guessed it: XBRL.

You can find more information about SBR on the Dutch and Australian websites. Stay tuned. These projects are worth doing.

So, I was contemplating a piece on the Assurance aspects of the SEC's interactive data proposals, now that there are a number of statements on the subject from the audit firms. But it turns out that Jim Hamilton has written a great analysis of the comment letters from the audit firms' positions on audit and XBRL. So I don't have to. Recommended reading. Some background from my perspective, here.