Charles Hoffman, creator of XBRL
Charlie Hoffman has been described as brash, outspoken and a rebel among accountants.
But the Tacoma, Washington-based auditor has managed the unthinkable. He may have revolutionized financial reporting. And he did it in a matter of months.
“It was a no-brainer,” Hoffman told The Bottom Line in his blunt way. “It had to be done.”
Hoffman is a certified public accountant with an MBA and virtually no training in software. But he is the guy behind XBRL – eXtensible Business Reporting Language – a new computer language that is being adopted around the world because it makes financial reporting easier, cheaper, faster and a lot more accurate.
Most people familiar with XBRL seem convinced that it will become the new global standard within a few years.
It has been adopted by regulators in the European Community, and by Japan, China, Britain and Australia.
In the United States, 8,200 financial institutions are already filing reports to the Federal Financial Institutions Examination Council (FFIEC) and the Federal Deposit Insurance Corporation (FDIC) in XBRL, while the U.S. Securities and Exchange Commission (SEC) is spending $49 million US to bring its EDGAR reporting system into the XBRL age.
“Today’s announcement demonstrates the commission’s firm commitment to interactive data,” SEC Chairman Christopher Cox said in a release in September. “The new system will make it easier both to file information with the commission, and to use it.
“For investors and analysts, it will represent a quantum leap over existing disclosure technologies. For companies, it will mean easier and less costly compliance with SEC requirements.”
It is proving popular in the private sector, where XBRL is used for internal reporting by Pepsico, Pfizer, Dow and General Electric.
It is making solid inroads in Canada. The Toronto Stock Exchange and Statistics Canada experimented with earlier versions of XBRL with some success.
And last summer, the Canadian Securities Administrators (CSA) announced it was taking a hard look at XBRL “to make it easier and more efficient for investors to receive, find, compare and analyze financial information.”
The Canadian Institute of Chartered Accountants (CICA) has been involved with XBRL for some time.
“Historically, we are where the United States was two years ago,” said Bill Swirsky, a CICA vice-president of new technology who has been active in the XBRL initiative.
He expects Canada to seek a voluntary filing system in due course. And the Canadian Securities Administrators’ SEDAR national reporting system will likely require work, he said. “We will probably find that SEDAR needs updating too.”
For the many people who have never heard of it, XBRL is variously described as a computer language, a communications tool, a global standard for financial reporting and the biggest thing since Lotus 1-2-3. It is really all of those, and none of them.
What it does is takes financial reporting out of the ledger book and brings it into the machine age.
Using XBRL, accountants can digitize their financial reports and convert them into a standardized, web-based, interactive format. They can compare and contrast; they can search; they can file regulatory reports to the authorities, and do it all cheaper, more accurately and faster than ever before with a minimum of keyboarding.
In its announcement last summer, the CSA offered one of the better explanations of XBRL.
They described XBRL as “a business-reporting language that allows financial information to be easily exchanged between organizations and the Internet.”
Its secret is to handle data in a slightly different way, using a trick adopted from the XML, the Internet’s Extensible Markup Language. Instead of treating financial information as a block of text, XBRL assigns an identifying tag to each individual item of data, whether it be revenue, cost of sales or long-term debt.
These tags remain attached to the items of data, the CSA said, “allowing information to be processed automatically in different ways for different users, eliminating labour-intensive manual data re-entry and verification.
“As a result,” the CSA noted, “XBRL can increase the speed of handling financial data, reduce errors and make it easier to analyze information.”
The CSA summarized the pros and cons of XBRL. The benefits include:
• Data only needs to be entered once, eliminating the cost of data re-entry and the potential for typing errors. That data can be used in many different ways, for calculations, printing, websites and regulatory reports.
• Financial data can be exchanged quickly and easily over the Internet, allowing investors and analysts to study the data using their own software.
• The data can be extracted and analyzed automatically, without the need for re-entry onto a new spreadsheet. Data from several companies and different time periods can be compared, making life considerably easier for regulators and financial analysts.
But XBRL isn’t perfect. The CSA offered several disadvantages, most of which will be minimized over time as the language becomes more familiar.
• It is offered free to all users. But the software needed to access XBRL is new and can be expensive.
• It is in its early stages and relatively few people are familiar with it.
• The time and effort needed to prepare information for XBRL depends on the complexity of the data and the preparer’s knowledge. That problem, the CSA said, will probably decrease over time.
The remarkable aspect of the XBRL story is that Hoffman had little understanding of software when he sat down in 1997 to read a book on XML, the computer language behind the web.
At the time, he was a certified public accountant (CPA), who had worked as an auditor and financial officer at a variety of accounting firms and small businesses in the Tacoma area. He had an MBA from Pacific Lutheran University and had taken a course in programming from nearby Microsoft.
He was, however, an innovator and rebel who had a reputation for constantly shaking up his employers, his fellow auditors and their established ideas.
He had worked at the former Price Waterhouse in the Tacoma area, but soon left. “They didn’t really get it,” he said later. “CPAs don’t like change. They thought I was a pain in the ass.”
Cary Deaton was a partner in the accounting firm of Johnson, Stone, Deaton & Pagano in Tacoma in the late 1990s, and Hoffman’s immediate supervisor. It wasn’t always easy being Hoffman’s boss, he said.
“He’s a guy who has always been frustrated that everybody couldn’t see what he saw,” said Deaton, who is now a litigation consultant at his own firm and still Hoffman’s friend. “He was probably constantly pushing them (the partners) and they were probably pushing back.”
“He was responsible for bringing us into the 21st century,” Deaton said. “For that time, for a CPA firm, we were cutting edge, pretty much due to Charlie. He was relentless. We thought he was outspoken, even a little brash. Charlie is one of those guys who didn’t suffer fools.”
Deaton said his friend Hoffman may have been “a rebel,” but he was “probably the right guy at the right time. Guys like him have to bring the rest of us along kicking and screaming.”
Frustrated with traditional accounting firms, Hoffman kept moving until he found a place at Knight, Vale & Gregory, a small, innovative accounting firm that gave him a long leash, a budget and the time to dabble with his new computer language.
The idea for XBRL came out of a project that Hoffman was doing for a client at KVG in early 1998.
The client wanted to transfer product pricing information from their accounting system to their website, and do it automatically.
Normally, that job would take a considerable amount of tedious typing, checking and double-checking. But Hoffman didn’t have the patience for that.
It occurred to him that XML could be modified to do the job, using hidden tags that would identify the parameters on the balance sheet and follow them electronically.
“I had been dickering around with the XML stuff,” he told The Bottom Line. But he admitted that he was soon in deep water.
“I’m a systems analyzer, not a computer programmer,” he said. “I’m a CPA that has taken some programming classes. I know how computers work. I can do Excel macros.”
But limited knowledge didn’t stop him. Hoffman showed some software experts what he wanted to do and taught them how to read a balance sheet. Between them, they got a version of XBRL up and running in only three months.
That was the easy part. “It took a couple of months to get it working,” he said. “Now, I have to basically pull the entire world along with me.”
Even that job moved surprisingly fast given the enormity of the task, and the traditional reluctance of CPAs to venture into unproven waters.
“I know how CPAs think,” Hoffman said. “I am a CPA. I work for CPA firms. I have had a lot of experience dealing with CPAs. I have had experience in making CPAs do what they don’t want to do.”
And, that basically boils down to packaging, he said. “If they can only say ‘yes,’ they can only move forward.”
His explanation may be simplistic, but it worked just fine.
In July, 1998, Hoffman told a friend on the High Tech Task Force at the American Institute of Certified Public Accountants (AICPA) about the potential of using XML techniques for financial reporting.
The task force was so impressed that it helped underwrite the development of the first prototype, a language dubbed XFRML for eXtensible Financial Reporting Markup Language.
The first draft of XFRML was finished by December that year, followed by a prototype of the true XBRL in October 1999.
In the meantime, the AICPA had established a self-funded, non-profit group called XBRL.org to oversee the development of a language they saw as a potential new global standard for financial reporting.
That, in turn, led to the launch of the XBRL Consortium, an organization with more than 200 member groups, including regulatory bodies, standards bodies, accounting firms, and software providers in Europe, Asia, the United States and Canada.
The European Community probably leads the world in XBRL use, Hoffman said, because it wants a uniform standard for all its member countries in 2007.
Some European countries are moving to XBRL for their domestic reports. In Britain, tax authorities will require XBRL for all company returns in 2010, while the Spanish securities regulator wants filings in XBRL by 2008.
It has also proved popular in Asia. The Shanghai Stock Exchange already gets its financial reports in XBRL, while Japanese regulators will require public companies to report in XBRL by 2008.
With XBRL, there’s one question that intrigues many, but is usually left unasked: “Is Charlie Hoffman rich?”
For once, Hoffman seems lost for words. “My goal is not to get rich,” he says haltingly.
“Making money is not important to me. I’ll never be a Bill Gates. I just make my salary. I’m comfortable.”
From what he says, Hoffman seems uninterested in making money.
“The payoff is that I get up in the morning and I love my job,” he told The Tacoma News Tribune for a recent story. “Nobody understands my job so they can’t tell me what to do.”
He started a company named XBRL Solutions Inc. in 2000 to “bring XBRL software solutions and services to the marketplace.”
But running his own company didn’t work out and he sold two years later to a local software developer named UBMatrix, which gave him a salary and the title of director of innovative solutions. Now he works from home, where he is left in peace to push the development of his pet project, naturally XBRL.
He isn’t even listed in the company telephone directory, but he is very active in the development of XBRL taxonomy, committee work and international promotions.
He decided the best way to spread the word was to follow the example of numerous other software developers and launch XBRL in the public domain, where anybody can download it for free. That doesn’t put any money into his pocket, but it does put him into position to making money at some point in his future, when he gets around to it.
“I’m well-positioned to make money,” he said. “I understand this stuff.”