I spent a couple of fascinating hours Tuesday at a round table hosted by the United States Securities and Exchange Commission. The subject was Interactive Data, a term which is hardly self-explanatory but really means “Business Transparency”. This in the same week that Jonathan sent a letter on the same subject to SEC Chairman Christopher Cox, who was also around the table. Mr. Cox and the SEC are definitely on the right track; I expect bumps in the road, but there’s a chance that Accounting As We Know It could be blown up. Which would be a good thing; and not just because Open Source is creeping in.
Interactive Data? · When they say this, they really mean XBRL (Extensible Business Reporting Language), an XML dialect designed for encoding companies’ financial statements. This rebadging, I am told, is Chairman Cox’s idea, and you can see his point; “XBRL” is neither euphonious nor compelling as a rallying cry.
The SEC wants public companies to replace the current practice of financial reports being big fat globs of unstructured heavily-footnoted text by an alternate reality in which they’re clean structured detailed machine-readable edifices of transparent truth. What’s not to like?
Sidebar: Washington Surprises · I’ve done a lot of business in That Nation’s Capital over the years, and to be honest I love it. Yeah, it’s slow and process-bound and infected with cronyism and worse things too, but at the end of the day they’ve got their arms, up to the elbows, in the stuff of reality. And it can surprise you.
Chairman Cox and his SEC surprised me too. He looks like the Republican insider he is; the haircut, the suit, the jaw, the smile. A lot of West Coast geeks would be apt to blow all this off as old-school or right-wing or “Enterprisey” or whatever. Which would be silly; Cox’s SEC is taking steps to drag business reporting into the cold clear light of truth by way of what he calls Interactive Data and I call XBRL, but the name isn’t the thing; and the parallels to the Open-Source world-view are not exactly subtle.
Oh yes, Open Source; among other things, the SEC is putting its money into some. Although it’s not perfectly clear in this press release, a substantial part of that $500K worth of software will end up as Open Source. That’s my kind of Republican.
Sidebar: On Accounting and Evil · It’s fashionable, in technology circles, to diss lawyers. Not here; I’ve been doing business for 25 years and while I’ve frequently complained about my legal bills, I’ve long since come to appreciate how essential attorneys are to the process of getting things done.
My bête noire has always been the practice of accounting. The reality is that in any given industry segment, exactly half of the companies are doing worse than average. Their management teams are highly motivated to conceal the truth. All too often, they find Generally Accepted Accounting Practices helpful. A certain part of the deep complexity and abstraction in accounting theory is in fact there to reflect business reality; but I’m convinced that another part of it is there to serve as a carpet under which to sweep inconvenient truths.
Attempting to impose accounting formalisms on business reality is always going to involve compromises and various kinds of twisty little passages, but some of those passages lead to the Dark Side, and I’ve seen companies go there. At a company I co-founded, we had to fire a CEO for “revenue recognition problems”. To those who don’t know what that means: keep it that way as long as possible. To those who do: XBRL is a tool that might turn turn out to be real useful in fighting the bastards who do this kind of thing.
Back to Interactive Data ·
Anyhow, here’s the dream: right now, if you know the name of a company,
you can be pretty sure that by visiting
www.company-name.com you can find the basics:
where the offices are, who the CEO and Directors are, and so on.
I imagine a future in which you can go to xbrl.company-name.com and be
pretty sure of finding authoritative machine-readable financial data.
And in this picture,
Metcalfe’s Law
applies in more than one way: not only does the value of the financial data
increase as a strong function of how many companies are providing it, but the
pressure to join in does too, on those companies who aren’t providing it.
XBRL ain’t perfect; they made no particular effort to hit any 80/20 points, so it’s big and sprawling and taxonomist-ridden and it tries to Solve the Whole Problem. In this particular case, I claim that the information is so valuable that it’s worth fighting through all this and finding a way to make it work.
In this vision, it‘s a whole lot harder for a management team gone bad to turn a decent company into a den of thieves.
Done right, the impact of XBRL oops Interactive Data could dwarf that of RSS and Atom put together. I’m on board, and anyone who believes “truthful business” isn’t necessarily an oxymoron should be too.
Contributions
Comment feed for ongoing:
From: Tony Fisk (Oct 05 2006, at 22:58)
Have you heard of Ricardo Semler, CEO of Brazilian company Semco International?
He tells of how he opened the company books up for general scrutiny. Standard wisdom would call this suicidal. Instead, they found the benefits (in the form of helpful feedback from the staff) outweighed the disadvantages. XBRL seems to be on his wavelength.
[link]
From: des von bladet (Oct 06 2006, at 00:42)
Exactly half the companies are doing worse than average iff by "average" you mean "median".
Almost everyone has an above average number of limbs, where "average" means "arithmetic mean".
[link]
From: John Cowan (Oct 06 2006, at 05:50)
"But the worst of all GAAP's many ills" is the decision to treat employees as liabilities (because you have to pay them at the end of the fortnight). That made the RIF the preferred way of improving a company's balance sheet.
I realized this years ago when I saw a NYT article about a woman who came in on Monday morning to find herself the sole survivor of her department of ten people. (What now, start working a 400-hour week?) Her employer -- a large and well-known bank -- was merged away within two years.
[link]
From: Woozle (Oct 06 2006, at 09:24)
I came up with something I've been calling the <a href="http://issuepedia.org/Open_Business_Model">Open Business Model</a>, which seems to be closely related to ideas discussed here. I've put in a link to your Transparent Business article, but please feel free to add more related information to the page (it's a wiki).
[link]
From: Geoff Shuetrim (Oct 06 2006, at 14:01)
On the topic of XBRL open source activity, there is already a small amount of activity. For example, there is a Java implementation of an open source API for XBRL that significantly eases some of XBRL's complexity. If you are interested, take a look at http://www.xbrlapi.org/.
[link]
From: Eric Meyer (Oct 09 2006, at 20:11)
"The SEC wants public companies to replace the current practice of financial reports being big fat globs of unstructured heavily-footnoted text by an alternate reality in which they’re clean structured detailed machine-readable edifices of transparent truth."
Well, in those cases where the financial reports are truthful, yes. (*cough*Enron*cough*) Though I suppose that at least if the falsehoods are encoded into machine-readable edifices, it will be much easier for third parties to analyze and uncover fraud where it exists.
I wonder if this will spawn a whole sub-market of gray-market accountants who not only know how to cook books, but to do so in ways that slip through the seams of XBRL. All in the best interests of shareholders, of course!
[link]
From: David vun Kannon (Oct 10 2006, at 07:35)
Tim,
Speaking as one of the editors of the XBRL spec since 1.0, I have to take issue with your characterization that it did not hit (or care to hit) the 80/20 mark. In each iteration of XBRL, this design heuristic was very much on our minds. The existence of a pipeline of modular additions to XBRL is evidence of all the ideas that didn't make the cut for inclusion in the core spec. Nor is the growth in the size of the spec document related to a great change in scope - rather it is related to a desire to be precise in areas that were previously ill defined.
OTOH, the constuction of specific taxonomies by subject matter experts may have been driven by other concerns. There are calls for the next version of the US GAAP taxonomy to be incredibly granular, in the hope that the average reporting entity won't have to extend it.
[link]