Does Columbia Business School study show investors aren’t using structured corporate financial data? No, says its co-author.


Supporters of H.R. 4164, which would exempt most U.S. public companies from the obligation to file their financial statements in the XBRL structured data format, have frequently cited a landmark January 2013 study by Columbia Business School’s Center for Excellence in Accounting and Security Analysis as evidence that investors aren’t using this data set. Dr. Suzanne Morsfield, co-author of the study, responds in this guest blog post.

Thank you  for your interest in my thoughts about how our study has been used recently in connection with H.R. 4164.  I was surprised to learn that a study I co-authored was being cited by various stakeholders to H.R. 4164 (the Bill) as evidence to support its passage. To my knowledge, no one citing the study for this purpose has contacted either of the paper’s co-authors to find out our views, or more simply, to confirm with us whether the quotes from our paper are indeed accurate. The study, “An Evaluation of Current State and Future of XBRL and Interactive Data for Investors and Analysts,” was released by Columbia Business School’s Center for Excellence in Accounting and Security Analysis (CEASA) in December of 2012.  While I do not wish to enter into the politics of the bill, I would like to note a few key things from the study for consideration. Any views expressed are solely my own, unless I am quoting directly from the paper.

Dr. Suzanne Morsfield: Investors demand structured data.

1. Most investors are using structured data through data vendors, not directly. The statistic in our study doesn’t reflect indirect use. The quote I am aware of that has been used to support the Bill is a misquote, most likely by accident, and, as such, does not convey the full views of the paper with respect to the importance of machine-readable, interactive data to all types of capital markets investors and analysts. The possible misquote (presuming it is taken from our study) in a recent press release from BIO, for example, states “A recent study of investors across all sectors of the market found that 92% of investors do not consider data from XBRL reports when making investment decisions.” Our paper actually states “Of our sample, few (8%) analyst/investors are utilizing XBRL data directly from the SEC website/RSS feed or other direct sources….those using XBRL data are using it for the perceived informational advantage they have by having interactive access to certain types of data that they believe they cannot collect elsewhere with the same effectiveness and efficiency. This [8%] measures whether they are using XBRL directly, as opposed to a data vendor’s product with XBRL-supplied data in it.”  Various data vendors are using XBRL data in their respective financial data subscriptions, and the statistic in our study does not reflect the number of investors utilizing XBRL data indirectly. The study also does not necessarily reflect current usage numbers.

2. Investors want more structured data, not less.  The users of financial reports we surveyed and interviewed also conveyed to us that they wanted more, not less machine-readable, interactive SEC filings and other data.  In the paper, we summarized this input as follows: “The core finding is that there is clear demand for timely, structured, machine-readable data including information in financial reports, and that this need can be met via XBRL as long as the XBRL-tagged data can reduce the total processing costs of acquiring and proofing the data, and that the data are easily integrated (mapped) into current processes.”

3. Investors want the SEC to fully enforce XBRL data quality–and transform more of its filings from documents to structured data. Investors and analysts further told us that they also would like the SEC to enforce the XBRL rule so that they as users can rely on these data more fully, and that they would like more SEC filings data “tagged” than currently were at the time of our study.  We stated in the paper, “there is demand for interactive data that captures the information in the footnote, MD&A, or earnings release data. However, this is generally pent-up demand due to current data quality and usability concerns.” In fact, 89.4% of those in our sample who knew about XBRL-tagged filings conveyed that they would be interested in utilizing these XBRL data once their respective concerns were addressed.

On a very personal note, I would like to add that I have spoken with my family members about our study, XBRL, the SEC’s machine-readable data efforts and status, and other related topics at length (and yes, they still want to see me when I come into town). I have four siblings, and parents who celebrated their 53rd wedding anniversary in November.  Although “investors and analysts” can seem like dirty words in some circles, as opposed to “entrepreneurs and business owners,” each of my hard-working family members is an investor and analyst to some degree because of their 401k plans and other retirement savings–the stay-at-home mom, the retired quality-control engineer, the CEO of a biotech company, the engineer at a defense contractor, the high school special education teacher, the single mom business manager of a nephrology practice, and finally, the researcher/former auditor/former equity analyst/former professor/former controller of a high tech start-up preparing for an IPO. With that in mind, and although I have been a transparent critic of many aspects of the SEC’s XBRL implementation and of ineffective financial reporting requirements, I do not support any exemptions to providing public regulatory data in a machine-readable format. This is especially true both in light of the decreasing costs of doing so reported by filers themselves in a recent FEI study, the increasing consumption tools now available, and of the proportion of Accounting and Audit Enforcement Releases to date that relate to filers with less than $250m in revenue in the years leading up to the enforcement actions. I cannot currently foresee a case whereby companies should not provide machine-readable data to the SEC and investors of all types and sizes, regardless of company size or industry, especially if these same companies also consider themselves capable and responsible enough to raise capital in the U.S. public markets.

4. In the 21st century, disclosure should be digital, not documents.  Finally, the paper addresses whether machine-readable regulatory filings would be important regardless of investor use: “Even if issuer-tagged XBRL data were not being consumed en masse by investors and analysts because they had found other, equally relevant, and more reliable data in the meantime, the usefulness of these data to regulators and enforcement/compliance agencies is a likely and potentially desired outcome. That is, digital filing of regulator-required information with filer-tagging is not an unreasonable societal expectation in the 21st century.”