Are Accountants the Wallflowers at the XBRL Prom?

Written by Bob Schneider     Posted on April 2, 2010

The stereotype of the timid accountant was established long before Hollywood started taking its shots. Leo Tolstoy captured this apparently timeless stereotype exquisitely in his epic War and Peace. As Napoleon’s soldiers battle with Russian forces, the accountant provides the comic relief: 

Behind Prince Bagration rode an officer of the suite, the prince's personal adjutant, and…an accountant who had asked permission to be present at the battle out of curiosity. The accountant, a stout, full-faced man, looked around him with a naive smile of satisfaction and presented a strange appearance among the hussars, Cossacks, and adjutants, in his camlet coat, as he jolted on his horse with a convoy officer's saddle.

"He wants to see a battle," said Zherkov to Bolkonski, pointing to the accountant, "but he feels a pain in the pit of his stomach already."

His reputation for timidity notwithstanding, today's accountant may well be forgiven for feeling a pang of torment when he first encounters XBRL. Taking a look at the XBRL section of the EDGAR Filer Manual, he finds most of it sounds like this:

The URI content of the xlink:href attribute, the xsi:schemaLocation attribute and the schemaLocation attribute, after XML Base resolution, must be relative and contain no forward slashes…

Told to get a feel of what an instance document looks like, he likely finds even the simplest of examples forbidding:

<HelloWorld:Land contextRef="I-2007" unitRef="U-Monetary" decimals="INF">5347000</HelloWorld:Land>

OK, the accountant rationalizes, XBRL is a markup language, so of course it’s going to have many more <’s and >’s than dr’s and cr’s.  As he navigates the new financial reporting superhighway, the accountant comforts himself that he can remain in the financial reporting driver’s seat, while the XBRL technician slaves under the hood.

At least that has been the assumption. As John Turner recently said in an interview with this blog: 

Since the early days of XBRL, there has been a mantra within the Consortium that XBRL models existing reporting processes, it doesn’t change them. XBRL doesn’t alter GAAP, or impose a standard chart of accounts on companies.

But despite the best efforts and intentions of XBRL’s leaders, I wonder if this powerful technology, like so many others, changes the game simply by being there.

When accountants first learned about XBRL several years ago, they were indeed concerned that it would entail a standard chart of accounts. They were reassured that it did not. Extensiblity is part and parcel of XBRL – it’s even part of the name itself.

But isn’t something like a standard chart of accounts where we’re headed? At the July 2009 XBRL Technology Workshop, Campbell Pryde of XBRL US said that extensions that appear often in instances will be added to the US-GAAP taxonomy – which, if I heard correctly, was reiterated by SEC staff at the SEC’s March 23 XBRL Public Education Seminar.

Fast forward five or eight years. At that point, won’t any extension be viewed with a whiff of suspicion by analysts and investors?  A company must be doing something unusual if the line item it needs to describe has not yet been incorporated in the US GAAP chart of…er, taxonomy. 

Financial analysts are already circumspect about extensions (see page 9 of the most recent CFAI XBRL survey). Perhaps they’ll become less so, but it seems more likely that, as taxonomies are refined, they’ll become more wary. When considering a new item for financial statements, won’t accountants look first to taxonomies for an existing element and try hard – very hard -- to find one?

What about XBRL’s impact on accounting processes? In the recent SEC XBRL Seminar, Tony Mealey, Senior Accountant at the Office of Interactive Data, said (beginning at 54:12):

“The third rule, to use the standard element with the narrowest definition, can be illustrated by the following example.

“In the statement of cash flows, the filer reports the payment for common stock repurchases. Instead of selecting the standard element “payment for the repurchase of equity,” the filer should select the element “payment for repurchase of common stock,” because it has a narrower definition.

“I might also point out that, even if the line item label in the traditional format financial statements reads  “payment for the repurchase of equity,” if the transaction represents only the payment for repurchasing common stock, that is the concept that needs to match the element selected.”

This surprised me. I would have thought the SEC’s instruction here would be “pick the element that most closely matches the line item on the traditional format financial statements.” The financial statements are the product of the hard work, debate, and compromise among management, company accountants, and the auditors. In this case,  that process yielded “equity,” not “common stock.”

Well, six of one, half-dozen of the other, right?  But the impact may be more significant. In her post on this blog last November, Chie Mitsui said:

When selecting sales from Japanese financial statements, users just choose the element “NetSales” from the “IncomeStatement”  and the “Current year” period context. All Japanese companies’ taxonomies have the same name for this element, hence users can easily select it.…[But] naming for US XBRL financial statements varies. Each company decides its own names for elements, presentation links, and contexts. For users unfamiliar with US company financials, such inconsistent naming causes confusion during the selection of elements for comparison.

This too surprised me. The item to be mapped and tagged is the top-line, i.e., net sales or total revenue. On traditional financials, the terminology chosen by companies varies little. How many different elements could there be for the top line?

The 3Q 2009 report of Progress Energy, a utility holding company, provides an example of what Chie was talking about. In the instance, the element selected for the top line was UtilityRevenue. The standard label for UtilityRevenue is “Electric and Gas Revenue.” The company extended the taxonomy to overwrite the label with “Operating Revenues,” which matches the top line in the traditional income statement. 

I don’t know if UtilityRevenue was chosen because it was the narrowest definition. I don’t know whether -- given the many objectives that financial reporting might have, from international comparability of financial statements to improved company analysis by SEC staff -- UtilityRevenue is the best choice or not.

What I do know is that “utility revenue” does not appear anywhere in the company’s  3Q financials (which were reviewed by the company’s auditors). Like “equity” versus “common stock,” perhaps the choice of UtilityRevenue for the top line of a utility company seems like a quibble – in fact, most definitely an improvement.

But suppose it was a forest products company whose business is 92% wood and 8% paper. The traditional P&L has “Net Sales”; the US-GAAP taxonomy offers both TimberRevenue and SalesRevenueNet elements. Which is better? At the least, such choices and the decision-making it requires seem like a non-trivial change in traditional accounting processes (not to mention what additional auditing procedures it may entail).

Perhaps that’s inevitable; perhaps that’s all to the good. But like so many other industries where new technology is introduced, it is reasonable to ask what the impact on accounting will be, and whether XBRL will merely express accounting output, or change it in important ways.