Toward Better Reporting: A Talk With Robert Pozen

The SEC's Advisory Committee on Improvements to Financial Reporting is scheduled to issue a report detailing its recommendations later this month. Committee chair Robert Pozen met with Business Finance to discuss the panel's current thinking.

Business Finance: The Committee has been examining industry-specific accounting. How does that arise in the first place?

Robert Pozen: A lot of industries -- casinos, mutual funds, for example -- say, "We're really unique, so in order to capture our financials we need to have different rules than everybody else." And while this may make sense for each industry, the result from the investor's point of view is that you've got a lot of specialized rules, and it's very difficult to have comparability and to really understand what each industry is doing.

Our proposal is not to wipe out industry-specific rules, but to say there ought to be a presumption against them which can be overcome if you can show that there really is a special reason for having them. And then over time we should move in the direction of having less industry-specific exceptions -- they're essentially not rules, they're exceptions.

There are also what are called alternative accounting treatments. For example, in the oil and gas area, for the same transactions you can account for it this way or that way. We're also recommending that we phase out these treatments for the same reason, that from the point of view of the investor, to have the same transaction accounted for in different ways is pretty hard.

BF: Will that make for easier convergence with international financial reporting standards?

RP: Yes, IFRS tends to have very few industry exceptions, so that would make it easier.

BF: Is complexity a factor behind the rise in restatements?

RP: In 2006, roughly 10 percent of all public companies in the U.S. restated their financials. Now some of this increase may be warranted by people looking more closely at accounting errors and being more vigilant, but it's clear that some significant part of this is not the result of more vigilance. We're getting a lot more of what I call technical restatements, whether it's because people don't fully understand the rules or there's a very high degree of concern about litigation, or other reasons.

In any event, probably somewhere around half of the restatements in the United States produce no market response. Which suggests that, from the investor's point of view, they aren't really important.

A restatement imposes obvious costs on a preparer, but it also imposes two sorts of costs on investors. One is confusion: If a restatement used to mean to investors "this is a really big problem," and you get a lot of restatements that aren't big problems, then we haven't sorted that out.

The other thing is, when you get a restatement, you usually have what's called a dark period; it takes six months to a year to get it right. And during that period there's very little disclosure.

Our effort is to try to do three things: First, to make sure that all errors are corrected. Second, to do so without requiring a total restatement for the last three to five years, unless the error is really material to current investors. And third, we want to have more disclosure as to what's going on. We hope this will mean that on the one hand you'll have more errors corrected and explained, but on the other you'll have a lot less restatements.

BF: Why is preparer and auditor judgment a focus?

RP: Contrary to popular opinion, accounting has always involved judgments. But judgment is becoming more important because when you move from historic cost to fair value there is, obviously, more judgment involved. Or take, for example, AS5; it's filled with risk-based judgments.

Our view was that we needed to come up with a proposal for how these judgments should be reviewed. We developed what I think of as a procedural approach to reviewing judgment. We can't, obviously, tell in advance what's good or bad judgment, but we want a rigorous procedure by which, first, you identify the alternatives you're considering. Second, what does the accounting literature say about it? Third, how does it fit in with the facts? Fourth, there could be an explanation -- a reasoned conclusion. And fifth, [the process is] documented contemporaneously with making the decision.

Our view is that if the SEC would adopt this as a policy statement, it would be quite helpful. It would be a big incentive for companies and auditors to follow this process, and it should give them some comfort that if they do this their judgment will be respected.

BF: What is the Committee's position on tagging financial information electronically, for example through the use of extensible business reporting language (XBRL)?

RP: The recommendation is that we have a gradual, phased-in approach to XBRL. We think before anybody gets into XBRL we need to make sure that we have the taxonomies right and we have the technology working. Our proposal is that we first have these [largest] 500 companies [tag some information] and next another 1000 companies, and we don't get to the smaller companies until we're really sure the technology works and there's software available that's inexpensive.

We also suggest that during the phase-in, these XBRL tags be considered "furnished" to the SEC and not filed, which means that they have less legal liability attached.

BF: What other major proposals are on the table?
RP: There's one other set of other proposals, on the process of standard setting. At FASB, we propose that there be an agenda advisory committee. Right now we have lots and lots of projects; they take a long time to get done, some of them. And nothing ever gets taken off that agenda. So we recommend that we set a priority set of agendas.

The second recommendation is that the FASB be the one that issues broad-based interpretations and that we don't have the AICPA [American Institute of Certified Public Accountants] or other quasi-groups do this, because all it leads to is what I call quasi-GAAP. So the recommendation is that the only authoritative sources will be the FASB and the SEC.

Third, we recommend that the SEC take the lead on registrant-specific guidance, but if there's a broad-based issue that grows out of a particular registrant, that it be kicked back to the FASB.

We also propose that there be a periodic assessment of all existing standards and we don't just let these standards stay out there, but see if they're responsive -- test them out and see how they're actually working -- and have some feedback and review.