I am coming to the party late on this one.  Back in October 2009, Rob initiated several posts related to the Kothari, Ramanna and Skinner paper that he saw presented at last year’s JAE conference.  I read the paper with a more critical eye recently in preparing for an introductory PhD seminar.  While it did not fit into this seminar, I wondered what I might have said to the students if I had assigned the paper.  Perhaps some bloggers have used the paper in this context and can report their experience.

The old posts can be found through these links: Unabashedly Normative and
The Asset-Liability Approach: Primacy does not mean Priority.

Since my observations are somewhat like running through the potato field a second time, I will try to be brief.  But this is fascinating.

I agree with Rob, the disclaimer about the paper being positive rather than normative is pretty weak.  And I cannot figure out why the authors are so petrified of admitting to normative tendencies, other than perhaps that JAE is founded on positivism.  FYI, I assign Friedman’s essay that one of the commenters mentioned, although I soften it with a discussion of how normative preferences often dictate the hypotheses we test and the data we collect.

To me, the normative leanings are most obvious in and around the section “Why should market efficiency be the maintained assumption <in standard setting>?”  The authors point out that no well developed theory identifies the precise amount of market inefficiency and even if we knew the amount of inefficiency, the standard setter is unlikely to be able to figure out how to alter the standards to correct the inefficiency.  Yet don’t the same issues arise with conservatism?  What positivist theory tells us how much conservatism is needed to maximize contracting efficiency – are contracts more efficient if companies write off 1% or 100% of fixed assets at the first sign of a decline in demand? And since the amount of bias needed is likely to vary with the nature of the contract and the characteristics of the contracting parties, how do you write a standard to promote the optimum level of conservatism without allowing  management a large amount of discretion in taking unverifiable write downs?  In the face of this uncertainty, I think a positivist might predict that a standard setter is going to identify neutrality and representational faithfulness as its goals because the alternatives are too difficult to specify/justify.  The JAE paper comes to a rather different conclusion on what a standard setter “should” do – ignore investor inefficiency and focus on contracting efficiency.

I was also interested in the following: “One solution to the current collusive agreement between the FASB and the IASB is for the U.S. courts, the U.S. Congress, or the SEC to expressly dismantle the convergence project on antitrust grounds and allow U.S. listed firms to adopt IFRS without reconciliation to FASB standards. This arrangement will likely force the two standard setting bodies into competition.”  I was hard pressed to see this as a positivist prediction.  While FASB and IASB have chosen to work together, I observe that this is largely due to pressure from a variety of constituents, including the major stock exchanges, multinational companies, the G-20, etc.  Indeed, has anyone heard from a group that recommends the Boards compete for market share other than academics?  Are academics seeing opportunities that others are missing or missing opportunity costs that others are seeing?  I have my views, but I think they are normative.

In summary, I find the paper interesting and thought provoking.  I just can’t help believing that a number of the positivist observations are normative preferences in disguise.

For what it is worth, I did assign Waymire and Basu, (2008) “Accounting is an Evolved Economic Institution”, Foundations and Trends® in Accounting: Vol. 2: No 1–2, pp 1-174.  That appears to be a much more positivist view on some key issues in accounting.  A shorter paper discussing some of the same issues is in Accounting Horizons 2006; it could be assigned to masters or undergrad classes.