14 November 2012

Laughing at the High Priests

The Wall Street Journal article on November 13th (Companies Grow Weary of XBRL) contains a lovely line:

"When Cyprus asked a panel of corporate controllers at the conference whether they were getting any value out of XBRL, the hotel ballroom full of corporate finance professionals erupted into laughter." 

In context, Nick Cyprus is the controller and chief accounting officer at General Motors, and this was at the FEI (Financial Executives International) conference in New York this week.

I do not know if the High Priests of the Cult of XBRL were in the room. They usually are. These were just the right venues to peddle the tired old promises that XBRL would achieve everything from faster, better internal communication, more efficient external reporting, faster closes, and a host of other benefits - most of which could be achieved simply by actually effectively implementing the already existing ERP systems. But it is long past time High Priests heard the message - no more bullshit about how XBRL will save my business! As I have said in previous posts, the only winners from XBRL so far (in the US filing implementation) are the SEC (so we are told again and again), the consultants, and Indian outsource providers.

There were options for the SEC, options that would have cost far less and that would have had a negligible impact on American public companies. These options were not taken.

When I chaired the XBRL US Steering Committee in 2005/2006, I had a meeting at the SEC discussing what it would cost to build the XBRL US GAAP taxonomy. My response, based on the detailed planning document provided to me by the XBRL US Taxonomy Working Group, said $4.5 million.

I gave that number, thinking "too high, we're dead".

The response I got was "Dan, we're the government. We really cannot solve $4 million problems. If you'd said $400 million, we might be able to do something." All tongue in cheek, of course.

Yet, how prophetic.

Because now the SEC does have a $400 million problem that they can solve. Let's hope they have heard what they need to hear from business.

As for the High Priests of the Cult of XBRL - I'm afraid they will never hear the laughter - even when it is no longer behind their backs.


  1. It's not like there wasn't thoughtful input offered to the priesthood along this journey. The core problem with XBRL -- that too little investment has been made into adapting it to be useful to the business operation or investment management process chains -- remains an undeniable artifact of its existence. As long as the biggest payoff of XBRL is that you comply with your static reporting obligation and pass an audit that you were going to pass anyway, the systemic benefit is negligible. One cannot hide economic reality behind financial reporting Fatwah's. Can it be made to work? Yes I still believe so. But it would require the priests to be brave enough to step aside for it to fourish.

  2. I wonder why XBRL *is* working -- in other countries.

    When I got started with XBRL in the early days, I was told that we (XBRL) shouldn't bother with EU countries much. (And Asia?? Hahahaha!)

    Why? Because EVERYONE takes their cue from the U.S./SEC.

    THANK GOD the rest of the world moved forward...while US corporates have perfected only one thing -- their ability to complain.

    Notice how much growth and innovation is coming out of Asia?
    Notice how much whining about regulation comes out of the U.S.?
    Notice how the US *still* has gotten its act together on IFRS?

    Connect the three. This is not about XBRL. It's about corporate resistance to government-mandates -- even in the face of evidence of intelligently-designed, socially-beneficial mandates working elsewhere.

    Question: What's the only fitting response to the conference of Controllers and CAOs who laughed at the value-add of XBRL?

    Answer: Go to a conference of CEOs and ask what's the value-add of their Controllers.

    Grow up, America. We've wasted a decade already -- and I'm not just talking about XBRL.

  3. Dear Anonymous,

    In an online exchange following the WSJ article, in response to Philip who wrote:

    "It's perhaps worth saying that in the UK, where 1,600,000 (yes, 1.6
    million) companies are filing XBRL accounts, the entire process has become
    largely invisible. The tagging is handled by accountants and few
    companies see it as a burden.



    I responded with:


    You are quite right to point that out (the UK success). It gives us the opportunity, rightly, to compare the programmes.

    - In the UK (and much of Europe where XBRL is being or has been implemented) the programmes explicitly seek to impose minimal or no burden on the filer / reporting company.
    - There are limited or no extensions created and in most cases even allowed.
    - The cost burden has been minimal at the producer end (although some companies may strongly disagree with that)
    - There have been no grand false promises about how XBRL will accomplish business benefits that can otherwise be achieved simply by effectively implementing the existing ERP systems.

    Contrast with the SEC program.

    - Each filer needs to build (or outsource) a unique XBRL document.
    - Each filer needs to develop XBRL expertise, or outsource that expertise.
    - Only a subset of filers are actually required to file (there is still, years after promised, no SEC accepted IFRS taxonomy - or maybe I missed that)
    - Only a subset of the statutory report is tagged, and not the part that analysts read for detailed analysis of the business and environment - the MD&A (Management Discussion & Analysis).
    - While there have been 50+ million data points reported to the SEC, there have also been tens or hundreds of thousands of extension elements created (for example - if 8000 companies produce XBRL, averaging 150 elements per [pre detailed footnote tagging] and they average 10% extensions - we have 120,000 extensions elements created - and I believe that is very conservative, when some of the very biggest filers are providing 60%+ extensions).

  4. Dan, in your original posting you conflated two 'enemies' --
    - SEC (wrong strategy and deaf)
    - High Priests of the Cult of XBRL (willfully deaf)

    But, then, you accept the 'success' of the UK programme?


    Of course they're not all of the success. But part of it, no?

    So, let's look objectively now...what's different about adoption of XBRL in US Corporate world and, say, just about everywhere else?

    You would be wrong to answer that question by again pointing out differences in programs!
    - Differences in programmes are ultimately the result of differences in people (users, edesigners, managers).

    The difference is US corporate culture -- and particularly its CAO brethren -- are risk-averse, litigiously-minded and ridiculously-zealous of weak regulatory controls.

    Quit blaming XBRL. (FDIC did okay by XBRL, no?)

    Back to my original point --

    > Notice how much growth and innovation is coming out of Asia?
    > Notice how much whining about regulation comes out of the U.S.?
    > Notice how the US *still* has gotten its act together on IFRS?

    > Connect the three. This is not about XBRL. It's about corporate resistance to government-mandates -- even in the face of evidence of intelligently-designed, socially-beneficial mandates working elsewhere.

    Same reason U.S. healthcare system is 35th in the world.

  5. Actually, for many years I advocated XBRL for exchange of information - as a 'boundary standard' between separate entities. XBRL is a data exchange standard, and was a good standard in some use-cases.

    It is worth noting that the 'successful' implementations of XBRL require minimal or no learning on the part of the provider of information, and where 'closed' taxonomies are used.

    With that in mind, maybe it is a good thing that the IFRS taxonomy is/was a closed taxonomy, even though the delays have further undermined the claims of benefits from the SEC program.

    After all, if an entire sector of reporting companies is exempt due to the lack of a taxonomy, maybe we could take that as a suggestion that, deep down, the SEC really isn't getting enough benefit from XBRL...

  6. It may be useful to consider why SIFMA http://www.sifma.org/about/memberdirectory.aspx is sponsoring this conference http://xbrl.us/events/pages/corpactionsforum.aspx promoting the use of XBRL for all corporate communications.

    It may be useful to consider or why SIFMA http://www.sifma.org/about/memberdirectory.aspx wrote a letter to the SEC in 2009 requesting that the interactive mandate be expanded to all corporate communications.

    It may be useful to consider that there may be a slight disconnect here when SIFMA is pushing for the expanded use of XBRL and some (repeat some) finance professionals are pushing in a different direction.

    It may be useful to consider why many finance professionals are finding disclosure management applications very beneficial to their reporting processes including the production of their XBRL reports. Companies like UTX, SAP, eBay and others that are not mentioned in the WSJ article.

    It may be useful to consider that it is no surprise that companies with highly manual report assembly and review processes are irritated when they add additional bolt-on manual processes to tag their manually assembled and reviewed documents.

    It may be useful for these individuals to consider why others have enhanced their processes using a simply idea of standardization. Below are a few examples that may be useful in considering these options:

    - "Disclosure Management: Streamlining the Last Mile" highlights the type of process and control enhancements available via these type of applications.

    - "How to Differentiate Disclosure Management Features" useful when considering the Disclosure Management application features most relevant to the company reporting processes and controls. http://hitachidatainteractive.com/2012/05/16/how-to-differentiate-disclosure-management-features/

    - Articles that discuss company Disclosure Management implementations:
    - ROI on XBRL - United Technologies 2007 implementation of the then Clarity FSR Disclosure Management product

    - SAP's Journey: Our Case for XBRL - SAP's Disclosure Management implementation

    - eBay Files 10-K 20 Days Faster - eBay's Webfilings Disclosure Management Implementation

  7. Dan, any change that you might post my prior comment?

  8. Mike,

    Of course I'll approve your posting. I moderate, but I do not block (other than spam - the only reason I do moderate responses).

    I remember another common theme - "let the market decide". If something is good enough to take hold, it will. The fact that government mandates something does not mean that it is either effective, or that it will be successful - I remember ADA (http://en.wikipedia.org/wiki/Ada_%28programming_language%29)

    If XBRL is destined to be successful, it does not require a mandate - remove it, and lets watch companies gladly create and file XBRL - because of all the benefits.

    After all, if you insist on mandating it, you're nothing but a socialist (heh heh heh). But I know you aren't, so free the market to decide!

  9. Dan, your 'free market' response was DEEPLY disappointing. As somebody in the world of standards, you should know better.

    My God, there are thousands and thousands of valuable things the free market decides against. Because it's only criteria is profitability.

    Society -- which includes government regulation and economic data analysis among other things -- decides based on a number of other criteria. Government subsidies Big Pharma to make 'orphan drugs' -- because otherwise the 'free market' wouldn't consider 100,000 people with an ailment a significant enough market to invest in.

    You still haven't acknowledged that XBRL *IS* working in many parts of the world -- and part of the credit for that must go to the 'High Priests.'

    You still haven't addressed why U.S. corporate culture -- apart from the innovators and more enlightened ones, some of which that Mike mentions -- largely is blindly, rabidly and self-destructively against XBRL. It's not because of the techonology or the High Priests -- because others abroad have done fine with it, thankyouverymuch. There must be some OTHER, American-centric reason....wonder what that is.....hmmm (hint: see my point above about healthcare system).

    And so, I don't think you're doing your clients a service if you point at the SEC and 'High Priests' as the problem. In fact, you actually do a great *disservice* to the advancement of corporate transparency by scapgoating the issue. In the end, you're just making your clients feel good about themselves.

    Instead, for objective progress in this area, I think you should be quoting them the line from Pogo: 'We have found the enemy and he is us.'

  10. What a wonderful comment, and very glad to see some push back. There are a number of themes in there, and I'll try to be concise.

    Some assumptions:

    1. Reporting burden on listed companies costs money
    2. Money spent must be earned, through product pricing or through reduced profit (but eventually though product price). Therefore the price of compliance reporting will be carried by the purchases - the 99% (give or take).
    3. Most benefit from shareholding accrue to a limited number - the 1% - 5% (giver or take).
    4. XBRL = additional cost -> carried by the 99% (give or take).
    5. Improved investment decisions = increased wealth for the 1% - 5% (give or take).
    6. Regulatory oversight protects business owners / shareholders - protecting the assets of the 1% - 5% (give or take).

    Therefore, from a purely capitalist market driven sense, XBRL is simply, when mandated, another indirect tax on consumers, resulting in the exporting of jobs, so that profits and asset value increase can continue to accrue to a very small minority, and in fact to the minority that is not paying for the benefit.

    "My God, there are thousands and thousands of valuable things the free market decides against. Because it's only criteria is profitability." Relation to your comment, I would respond - the role of mandatory standards is in situation in which the lack of standards negatively impacts the majority. Where a minority benefit from the standard, let them implement it and pay for it. I.E. let the consumers of data pay for the data, as they will be gaining the benefit of the data. Big Gov subsidies orphan drugs because the people that need them cannot afford them, and the majority suffer accordingly.

    "You still haven't acknowledged that XBRL *IS* working in many parts of the world -- and part of the credit for that must go to the 'High Priests.'" Hmmm.... you may have missed by other articles "Spain, the quiet achiever" or "The logic of the logical cave wall" to name two. Also, I think in a response above I do mention the UK's success. Yet even in the 'successful' examples, all we have are volumes, and no (or few) overt statements on how the successful use of XBRL is driving down the cost of business, the effectiveness of government, etc. Now, in this cases, I am also suggesting that we are also not seeing that XBRL cost *less* than other alternatives, or delivered *more* benefit than other alternatives.

    "You still haven't addressed why U.S. corporate culture -- " Ummm... I might be missing something, but here I'll only reference Milton Friedman: "The business of business is business". He also postulate, quite strongly, that the decision by management to pay for programs that do not improve the value of the business to the shareholders is tantamount to theft from the owners. Like US corporate culture of not, it is what it is.

    Finally, I do not care if the "High Priests" promote a voluntary standard. I do mind if WE lie about the faux benefits of a standard, when the vast majority of the promised benefits will actually accrue through effective management - re-engineered systems, efficient practice, and not through overlaying already inadequate systems with another layer.

    Final finally - Great standards naturally expand and are adopted (HTML, XML, USB, Batamax (oops, sorry about that), DVD). Inadequate (but good for their time) standards help show the way, and provide opportunities to enhance standards, or develop better standards. Let the Priest preach, but do not expect everyone in the congregation to believe, no matter how often it keeps being repeated.

  11. > Therefore, from a purely capitalist market driven sense, XBRL is simply, when mandated, another indirect tax on consumers, resulting in the exporting of jobs...

    HAAHHAHAHAHAAHAHAHA! Wonder how many Indian companies using XBRL are exporting jobs!!

    I'm sorry, I couldn't take the rest of your comment seriously after that. I'll try to count to ten and comment on the rest later.

  12. Ahh, Anonymous, you should also wonder how many Indian companies that are required to produce XBRL would do so if optional....

  13. ZERO. Well, Infosys -- which consistently wsa rewarded with a lower cost of capital for its excellence in transparency and governance.

    But, to answer your question, ZERO would take that first step. Nor would they in the UK, which was successful, right?

    So I guess not doing its optionally doesn't prove your intended point that mandatory=must be bad.

    In fact, conversely, because few would do it optionally, IS the reason why it was mandated. And -- proving that the whole is greater than the sum of its part -- (UK/Indian) society benefits.

    For a blog post titled 'Laughing at the High Priests' I'm surprised nearly all of your replies come down to the reluctance of (U.S.) Preparers. You still seem to be wedded to the idea that if they are reluctant, it must be a problem with the 'High Priests' -- so much so they are deserving of being soorned and laughed at.

    No. The reluctance of corporates has a terrible track record: they resisted Child Labor rights, Minimum Wage, Social Security, Civil Rights, Diability Rights, Maternity Leave -- my God, I can't think of any societal progression that corporate interests were in favor of -- if it would cost them even a nickel more. Vacation time? Ha!

    So, please, again, consider just how uniquely short-sighted this U.S. corporate cultural disease is. And consider, please, just for a moment, if that might not have a teensy weensy little part to play in why U.S. preparer community is so backward in its thinking about XBRL (and in so many other ways).

  14. Let me return first to the point of this posting - at the FEI conference, "When Cyprus asked a panel of corporate controllers at the conference whether they were getting any value out of XBRL, the hotel ballroom full of corporate finance professionals erupted into laughter."

    Now, I am not saying that all regulation is bad.

    I am saying that after years of preaching crap about how XBRL will deliver a range of benefits, the very people who have been preached to for years, laughed. They know the promises of benefits were empty. Because they had to pay for the XBRL.

    Corporations purpose is to increase shareholder wealth. No more, no less.

    Governments purpose in relation to corporations is to ensure that in accomplishing that single goal, corporations deliver more benefit to society than harm. Therefore "Child Labor rights, Minimum Wage, Social Security, Civil Rights, Disability Rights, Maternity Leave" - none of which I have disagreement with.

    But I also do not see on your list - needless additional administration paperwork (or computerwork).

    So my turn for a question - Why did the room erupt into laughter?

  15. I think we're talking at cross-purposes.

    I get your point that U.S. CAOs see no value in XBRL.

    I don't think you get my point that, because XBRL has proven valuable elsewhere, the U.S. CAOs are resisting XBRL for reasons that are not technical. Put another way, they are a selfish, stubborn and individually-minded lot. Not every corporation or culture is -- thank God. It's a very American corporate thing.

    But you seem to cling to the idea that if U.S. CAOs resist, the 'High Priests' must have been lying about the 'benefits.'

    I've tried from several angles (international, historical, etc.) to show why, e.g., not wanting to do a standard voluntarily, does not mean it has no value, or not having enough perceived value for your company (individually), does not mean it has no value collectively (and therefore, later, individually). In fact, there is proven benefit. So why their hostility to the 'High Priests'?

    Yes, U.S. CAOs laughed. Just as they always do for things they see as not beneficial to their own narrow, short-term interests. You acknowledge government's role in mandating Child Labor, Disabilities Act, etc. -- but you IGNORE that U.S. corporates FOUGHT these.

    Imagine a room full of coal mining execs in the 19th Century. 'No child labourers??' And the room bursts into laughter. Hahahahaahahahaha.

    We know XBRL works elsewhere.
    And yet, the U.S. CAOs laughed.
    That laughter said more about them than about XBRL.

  16. We are talking cross-purposes.

    1. CAO see no value-add in XBRL.

    2. They would also fight any regulation.

    3. Child labor, Disabilities act, etc, were enacted to protect human rights from rapacious corporations.

    4. Regulators claim that regulation implemented = success. Yet we are not hearing the regulators say "this is great, and this is why" (except in a few cases where they are doing so to justify what they have imposed).

    5. XBRL does not delivery any benefit to the protection of human rights that cannot be otherwise achieved. We are conflating regulation for the protection of the majority with regulation that introduces waste.

    6. We know that XBRL 'works' - we just cannot prove that it adds any additional value than what is otherwise achieved by less costly methods.

  17. Ok Dan,

    I will get involved. Let me start with your quote: "It is worth noting that the 'successful' implementations of XBRL require minimal or no learning on the part of the provider of information, and where 'closed' taxonomies are used."

    XBRL should be built into software, as it is in the FDIC implementation, not added on afterwards. But you know that. There is no more need for the non-technical people to know any more about XBRL than about the code that creates word-processing docs or spreadsheets.

    Yes, successful implementations of XBRL do not allow extensions. Why should they? In my opinion, this is a mistake of the SEC, one that comes on the heels of the American idea that even when facing regulation, one needs the freedom to do what? to mislead? The IRS insists that business gives them the information they need in the format they designate.

    Effective regulation is based on getting the required information. It should be easy for business to provide that information and hard for business to deceive. The SEC and all other regulators should all be using the same standard and the data should flow from the ERP systems, without anyone knowing or caring how much or what info they are providing the regulators, just knowing that if the input is correct, the output will also be correct. And there should be no need for paper or quasi-paper documentation at all.

    I know that this will be more difficult for he accountants and consultants than finding ways to add on bolt-on XBRL solutions and maybe less lucrative.

    I see advantage in good regulation and so do you. The same XBRL (hidden in the software)could assure regulators not only that securities rules are kept, but also rules which should exist if they don't about energy consumption, child labor, etc.

    Some of "high priests" tried so hard to get some regulator in the US to adopt XBRL that they did not insist on it being done right. Which ended up meaning that it was done wrong and did not yet give the benefits to companies, to regulators and to people who struggle or have struggled to build XBRL solutions.

    What should be done? Clearly solutions are in sight if only we look to find them. "Faster, cheaper , better" need not be a slogan. It ought to be the reality.


  18. I have been following XBRL from its creation up to now from the point of view of a software editor in the semantic field.
    I have noticed that there was an inherent confusion between XBRL as standard for data tagging and the technology for its implementation. Hence the name. In the industry a standard is agnostic of the tooling use for its implementaion. I do not understand why it is not possible with XBRL.
    As a defined vocabulary for tagging terms XBRL is a metalanguage.
    The terms are the language in which they are expressed the taxonomy.
    To me what has been said here is right from all sides and until we decide that XBRL is a vocabulary and not a technology there will always be problems and the costs will follow the benefits will be kept to a minimum as it has been mentioned 'to a static document' in software jargon it is called hard coding not soft coding. This has large implications in the database.