Yesterday (September 10, 2009), Rob Bloomfield posted regarding a legal perspective on FASB Accounting Standards Codification Topic 450 (FAS 5 – Accounting for Contingencies) and proposed changes in the disclosure rules for litigation-related contingencies.  This post relates to both the FAS 5 issue as well as an earlier discussion regarding conducting research with the FASB.

While serving as FASB research fellow last year, I spent some time with the Loss Contingencies project team and read their Exposure Draft, “Accounting for Certain Loss Contingencies.”  While reading, I noticed in the introductory section that the FASB’s stated motivation for undertaking the project was assertions by “investors and other users of financial information” that existing guidance in FAS 5 was failing to instigate adequate disclosure of contingent legal liabilities.  In speaking with the project team, I ascertained that while they had indeed heard from multiple constituents about the nature of the problem, there was a paucity of actual, systematic empirical evidence about whether indeed there was or wasn’t inadequate disclosure.

With two co-authors, we proposed to the project team a study where we would document the extent and characteristics of disclosures, in an effort to provide some hard data to the Staff and Board of the FASB to aid them in determining the exact extent and nature of the alleged problem.  We conducted the work and provided a report to the Staff in June of this year; the report was then included in the materials provided by the Staff to the Board in preparation for the August 19, 2009 meeting where the Board re-deliberated the issue. 

Briefly, we identified a sample of lawsuit outcomes reported in firms’ financial statements where the firm suffered a loss.  We then looked back in time at previous quarters’ financial reports to read and code firms’ disclosures related to the lawsuit.  We found that there was a surprising frequency of non-disclosure of lawsuits in the quarter immediately preceding the disclosure of the loss (approximately 8% of our sample).  We also found that a  majority of firms (52.9%) did not provide an estimate of the loss or range of possible loss, or explicitly state that an estimate could not be made (as required by FAS 5) in the quarter immediately preceding resolution of the lawsuit.  On the other hand, a fairly large proportion of firms voluntarily provided disclosures that were among those new proposed requirements of the Exposure Draft. 

Taken together, our evidence is supportive of the conjecture that there is inadequate disclosure for some litigation-related contingencies and supportive of some of the new disclosures proposed in the Exposure Draft.  We believe that our work was helpful to the FASB in their re-deliberations.

Beyond its contribution to an understanding of the status quo with respect to litigation-related loss contingency disclosures, I think that this sort of study can be used as a template for other work likely to be useful to standard-setters.  That is, standard-setting is decision-making under uncertainty.  Standard setters hear claims of financial reporting issues from their constituents, but most often in the form of (unsupported) assertions and anecdotal evidence.  Academic researchers have the comparative advantages of skills, access to data, and objectivity that can enable them to provide empirical evidence to support or refute the claims on which standard setters rely in making their decisions.

A manuscript version of this report will shortly be available on SSRN; interested readers can contact me for a copy (r.pfeiffer@tcu.edu)).