Financial Accounting Standards Research Initiative

I once asked Sir David Tweedie at a presentation he made in New Zealand how accounting research contributed to his work as a standard setter and he replied “very little.” He went to add that he didn’t understand most accounting research and he preferred normative accounting research of old to the modern empirical research.

I think most Board members believe they trying to write accounting standards to reflect economic reality and that they can logically discern economic reality for a given set transactions so there is little need to consult research. In this case, research is primarily used to support an already established position. See “The Demand for and the Supply Accounting Theories: The Market for Excuses” by Watts & Zimmerman. Thus, accounting standards may be driven more by political considerations than research.

In my own view, what standard setters need most from academics are empirical tests of specific, theory-driven research questions. This is in contrast to needing academics to theorize in the absence of hypothesis testing.

As Rick points out, much of the analysis done by standard setters, and I would add their constituents, is rooted in the use of logic and/or rhetoric to reach conclusions and consensus.

Generally speaking, standard setters have not been trained to do hypothesis testing, and this is where academics can come in, with a strong comparative advantage, to provide empirical evidence, untainted by political considerations or a bias for one outcome over another.

In fact, this will be the topic of a panel that I will be on at FARS later this week with Tom Linsmeier and Bob Lipe.

Disclaimer: The views expressed here are my own and do not represent positions of the Financial Accounting Standards Board. Positions of the FASB are arrived at only after extensive due process and deliberations.

My take is if you ask IASB and FASB members if they value academic research, you will get varying answers. This is for two reasons. First, as mentioned by Rick, Board members complain that research is not written in ways they can understand (see the articles in my earlier post for some suggestions how academics can address this problem).

Second, experimental evidence suggest that people are not all that good at understanding what they do. For example, a stock trader might say he/she never uses earnings to pick stocks, but in an experimental setting, differences in P/E ratios may have a very strong correlation with their trading behavior. In this regard, as FASB Research Fellow, I attended a meeting with several Board members in which they reviewed financial statements that companies had “recast” under the preliminary guidance in Financial Statement Presentation project. Tweedie had supported one particular format for one disclosure, whereas many other Board members supported a different format. When looking at evidence of how real companies would apply the proposed guidance, Tweedie observed that the format he preferred was not useful. From that day forth, this format was removed from the project.

Now you can argue that the recasting exercise is different from the large sample empirical tests that many academics perform. But the exercise is very similar to smaller sample field studies. As such, I posit that Tweedie and other Board members are quite willing to pay attention to evidence when they can understand what the evidence means for the choices they are being asked to make.

I’m very gratified that someone responded to my comment. It is the first time I received feedback on my observation. I now feel part of the community. Thank you.