We had a very interesting session yesterday to brainstorm ideas on leasing.  Thanks to another doctoral student for summarizing key parts of the discussion.  –RJB

Despite the loss of power of our campus 15 minutes prior to the beginning of the office hour and the rush to set up everything on our backup battery, I, along with other PhD students here, enjoyed attending the session of “Brainstorming on Leasing” very much. The discussions were enlightening, especially to junior researchers like myself.

I was particularly interested in the discussion of FASB/IASB’s proposal on “Measurement of contingent rentals”. The two approaches proposed by the boards are: (a) a probability-weighted estimate of the rentals payable (an expected outcome technique) and (b) the most likely rental payments. The IASB recommends the first approach, while the FASB favors the second one.

Both approaches have their pros and cons. Obviously the standard setters have not come down to a conclusion as to which one is better. I think it would be helpful if the standard setters and researchers could put the issue under the positive accounting theory framework in conjunction with conclusions from psychology research. Several office hour participants mentioned that people (managers in this case) have limited ability to assess the probability distribution of likely events.

While I am not an experimental researcher, I am concerned with the cognitive bias of managers under either approach. Under the first approach, conclusions from psychology research indicate that people are not very good at probability assessments, especially with the extremes of the distribution. Under the second approach, managers have to assess the most likely outcome. This approach may lead to confirmation bias of managers. If a number somehow strikes managers’ mind as the most likely rental payment, they will seek evidence that support their belief about the true outcome. Therefore the reported “most likely” rental payment may not be truly “most likely”.

Another thought of mine is that the bias could be intentional or unintentional. Unintentional bias could be due to people’s cognitive limitations as mentioned above. However, intentional bias could be due to managers’ self-interested purposes. Both approaches are susceptible to this bias. With the first approach, managers may discretionarily assign the highest probability to the number they want to report. Under the second approach, if managers have a personal goal in mind, they may simply report the number in their minds as the most likely rental payment in accordance with their personal goals. I think this is where findings from both experimental and archival research could inform standard setters and possibly help them with their decisions.