Bob Schneider is a Partner in 
Wilson So is the Director of Hitachi Consulting Corporation Five Reasons Why Sell-Side Analysts Aren’t Demanding XBRL
Written by Bob Schneider Posted March 2, 2007
A long-time IR professional recently suggested to me that he wouldn’t be interested in XBRL until the analysts are. In his presentation at the XBRL Philadelphia conference in December, Jeffrey Diermeier, President & CEO of the CFA Institute, noted a similar lack of interest among his membership, although he ascribed it to a lack of awareness. “The good news,” Mr. Diermeier says, “is that when our members learn more about XBRL, they appear to like it.” (To read his speech, go to XBRL Philadelphia Presentations and click Jeffrey Diermeier — XBRL: An Investor’s Perspective).
Is lack of awareness the main hurdle that XBRL faces in gaining acceptance among security analysts? Or are there other reasons why analyst interest may not be overwhelming? Here are some possible reasons of why sell-side analysts at the big institutions are not banging on the table and demanding XBRL financial statement filings now:
1. You never get a second chance to make a first impression Analysts are very busy people, and they don’t look gladly upon stuff that unnecessarily complicates their lives. Any new tech thing that starts with “X” will likely be greeted with rolled eyes and much suspicion. Their first impression may well be that XBRL will deliver something similar to what the data aggregators provide, i.e., numbers that must be reviewed and cleaned up before using.
2. The financials are just some of the numbers and it isn’t only about the numbers anyway Sure the financials are important, but they are only some of the data analysts look at to evaluate stocks. There’s numbers from the government, numbers from industry associations, numbers from private research outfits, and other numbers from the companies themselves. Then there’s all the stock market data — price action, volumes that even analysts who don’t care much for technical analysis must pay attention to.
And the numbers are just part of a company’s story. The really hard work of the analyst is assessing things like management quality, product innovation, labor relations, the legal environment, etc. These are things that the numbers reflect but often don’t help you evaluate much.
3. How much transparency will XBRL provide? Take footnotes, for example. Analysts know better than anyone that the accompanying notes are indeed an integral part of the financial statements. How much help can XBRL offer analysts in unraveling their mysteries? It may be possible to tag and thereby illuminate certain data within the note. But can XBRL help analysts avoid slogging through a densely worded note to discover its true meaning?
4. XBRL creates uncertainty Let’s put aside the issue of whether XBRL is complex or not and how much that matters. XBRL raises questions for analysts beyond complexity. For example, what about company taxonomies? Will they affect the way firms report their numbers and their comparability to those of like companies? In addition, to some extent XBRL has become associated with efforts to standardize accounting principles worldwide. Is that always a good thing? Maybe some of the to-be-discarded standards provided useful info. And as global standardization occurs, will financial people be calling the shots, or do the technologists have their own agenda?
I am not saying these fears necessarily reflect reality; I am saying they are legitimate.
5. Don’t take away Jodie! XBRL proponents point to the time and expense it will save through “enter once, format many” and the elimination of data re-keying. If I’m a top analyst, it’s likely I have a “Jodie” in one job title or another research assistant, data entry clerk, intern to help me with that busywork. I like Jodie. Jodie brings me coffee. Jodie talks to me about the Knicks. Jodie wants to be an analyst, and I get to teach her how. Now tell me again, Why do I want a new data standard that makes Jodie less essential, or even superfluous?
These are just educated guesses — it’s been a long time since I worked in a big equity research department, where I provided analysts with production and compliance support. Further, all of this discussion is strictly from the analyst’s point of view. The Director of Research at a big institution would have different objectives cheaper coverage of more names in more geographic markets, for example that may well cause him to embrace XBRL whole-heartedly.
Nevertheless, I do think the lack of analyst interest is real, and it isn’t only because they simply haven’t heard of XBRL. What would I suggest to allay analyst concerns? One thing I would do to gain the acceptance of analysts (and other busy professionals like auditors, financial executives, etc.) is to create short (two pages or so), focused primers on XBRL that introduce the standard and address the particular concerns of each professional group. While good XBRL resources are available, most of them speak to a general financial audience. Financial professionals are no more monolithic than, say, doctors, and the concerns of each financial specialty need be addressed.
March 3rd, 2007 at 1:53 am
I know that blog posts containing lists of things are de rigeur, but why have five reasons when one is quite enough? That one reason is lack of awareness. You are omitting the hard facts that Jeff quoted in his presentation, namely that a recent CFA Institute survey among its members indicated that a majority of analysts never even heard about XBRL. I can tell you from personal experience that everybody in the analyst community that I tell about XBRL actually wants it. But the trouble is that currently, there are way too few instance documents available – this is where the good old chicken & egg thing comes to the fore.
As for the primers you ask for, may I suggest this.
March 4th, 2007 at 8:20 am
Awareness is not the issue. Rather it is a question of relevance. Most users of MS Office never use 90% of the fuctions in that product. The assumption that users want more, interactive analytics tools is equally suspect. Yes, some users will be very interested in XBRL functionality, but the vast majority of downstream users don’t care — and probably never will. The indirect benefits of XBRL, namely accuracy and timeliness, may be as much as most end users ever care to know — if they know. The assumption that there is a function-hungry mob of end users clamoring for XBRL is one of the chief obstacles to moving this technology forward. Install it upstream for data organization, validation and transmission, then let the downstream users figure it out for themselves. A Soviet-style, top-down push approach is doomed to failure.
March 4th, 2007 at 9:19 am
Change will come to the analyst profession from the outside, not from the inside. How many examples are there of industries that refuse to stay on the cutting edge of technological change only to find themselves struggling to adapt down the road?? In fact, much of sell-side analysis is little more than an aggregation of public data with a few formulas and some commentary mixed in. This will be a commodity product in a few years’ time. Just as blogging took the newspaper industry by storm, independent financial analysis will shape the future for sell-side analysts and XBRL will be a part of this.
March 6th, 2007 at 12:56 pm
Those of us involved with XBRL probably spend too much time with a) the idea that financial analysts MUST be key users of XBRL and b) the idea that XBRL comes an an electronic balance sheet, p&l etc.
Financial analysts spend hours building their spreadsheets; they model companies, industries and market trends. Some of these spreadsheets have matured over years. And some even are very useful tools for forecasting future performance of a corporate. Are we seriously expecting analysts to let anybody flood their spreadsheet with data they have not had the chance to check? They love their spreadsheet, it reflects their expertise and they will guard it. Why not speak with other functions within the investment chain e.g. those tasked with stock selection or quantitative research? Typically, these functions scan large chunks of data. Why not speak to fundmanagers who receive tons of research material in paper or pdf? Speak to the buy-side if you want the sell-side to move, as they say.
Also, investment professionals love to have forward-looking statements from corporates. These do not come with balance sheets and p&l’s. Why then spend so much time on balance sheets in XBRL ? Wouldn’t a simple earnings release with say 12-15 key figures both for past periods and future periods show the beauty of XBRL and at the same time show investment professionals that their needs are well understood?
March 8th, 2007 at 8:41 am
The bottomline is that the SEC needs XBRL to make THEIR jobs easier (faster) to review the vast amount of filings SOX regulations have brought to them. That is why it’s coming to the forefront after 5+ years of being on the fringe. Whether financial analysts or the reporting issuers want/need/like/hate XBRL is moot.
Financial analysts will continue to use their own models, spreadsheets and intangible research forces. At the least, perhaps XBRL will just help them TACTICALLY fill in their XLS data cells faster so they can spead better time on strategic research.
March 12th, 2007 at 2:37 pm
The reason I’m skeptical of the primacy of the awareness argument is that it’s not a reason people usually accept in fields they are vitally involved in. Instead, their reaction to being told there is a great innovation that will substantially improve their lives is “If it’s so great, what haven’t I heard about it?” You would think that XBRL would have made its way into analysts’ consciousness by now if they saw it as critical to their work.
I do agree with Chris Dreyer, however, that the overview of XBRL he and Mike Willis wrote for Professional Investor is excellent; indeed, it’s among the best I’ve read. I should have mentioned it in the post I wrote a couple of months ago about who I like to read in the XBRL field. (Actually, I think the only reason I didn’t mention it was because I included Mike Willis among our guest bloggers, and therefore linked to his post on this blog instead of external writings. But Chris deserved recognition too.)
April 19th, 2007 at 3:00 pm
Bob, I can only relay my personal experience: My very first impression of XBRL was made some time in 2002 or so, and then it certainly wasn’t mature enough for our usage. That has changed now, and awareness of that fact is slowly creeping into the collective mind. But it’s difficult to revise a first impression: XBRL may be one of the rare cases that get a second chance of making a first impression …
Incidentally, expect analysts to become more aware soon.