During the FASRI Roundtable this week, Tom Selling sparked a discussion about the extent to which accounting standards are unnecessarily complex.  In response I posed the question whether it was possible to have simple accounting standards for underlying transactions that are complex.  As I continue to think about it, I think that this is a very interesting question that is under-appreciated by many. I am reminded in particular of segments of the congressional hearing held last year about the role of fair value accounting in the credit crisis ["Mark-to-Market Accounting: Practices and Implications," Subcommittee on Capital Markets, Insurance, and Government Sponsored Enterprises, March 12, 2009].  One persistent theme underlying many of the arguments made by members of Congress in that hearing was that the problem was simple — all that was required was for the SEC and the FASB to get rid of fair-value accounting (or at least limit its use for financial institutions).

In some ways, “simplicity” is like “freedom”, or “justice.”  Who would argue against the virtues of simplicity?  However, is that really what we want?  Simple rules presumably would be generally applicable with very few if any exceptions.  By their nature, they would then treat a relatively wide range of transactions and items similarly.  So, for example, an accounting rule that requires that all leases must be capitalized or that all leases must not be capitalized is clearly simple.  But the contracts being accounted for — leases — are themselves quite complex because they are designed to optimize across multiple dimensions, including tax treatment, the benefits and costs of ownership, risk sharing, etc.  “Complex” lease accounting rules include flexibility to comprehend specific details about lease contracts and to look beyond the form of the contract to its economic substance, thereby enabling users of the financial statements potentially to understand subtlety surrounding the firm’s investing and financing strategy that would be lost if a simpler accounting rule were in place.

Granted, complexity in the rules has multiple impacts.  In addition to the potential to communicate subtlety, it also facilitates strategic financial reporting that can mislead rather than inform.  In addition, financial reporting has multiple audiences, each of whom likely has unique preferences for the information that is communicated about contracts, meaning that it may not be widely agreed-upon whether more or less simplicity is optimal.

Hence in choosing along the spectrum of simplicity and complexity of accounting rules, it seems to me that we need to be aware of the tradeoffs inherent in differing levels of complexity and choose as well as is possible, given the inherent uncertainty about the existence of an optimal level.

I think it is important to resist the temptation to strive for simplicity and to ignore the potential benefits of complexity.