ordinary operating expenses outweigh any costs associated with disclosure of strategic information to competitors and regulators.

These points largely supported Lev’s point that the leadership to improve the value of financial disclosure has to come from government agencies or those related to government. The key role of external forces has been evident historically. In the New York Stock Exchange of the late 19th century, there were no SEC regulations. The exchange pleaded with the CEOs of major companies to disclose their annual sales, but the response was the same as that heard today from managers with respect to other types of information: The disclosure will benefit competitors, it is costly, and it is not needed. And this is with respect to sales—nothing of importance was ever disclosed without having been required.

One participant pointed out that spending by companies on intangibles is hidden in income statements. If the information is disclosed in a separate statement of detail, as is done with other kinds of management information, it would go far in promoting an understanding of investment at the firm level. It would be powerful for companies to be able to see what they are spending on, say, training, relative to their peers in the industry. It would be in the firm’s interest and also in the interest of analysts. Lev agreed with this assessment but noted that there are two aspects of disclosure: One is getting managers to disclose new information, and the other is standardizing the information—that is, making it comparable, which is as important as the first one. With respect to most intangibles, including R&D, there is little standardization.

Bossio followed up this point by noting that the chairman of the SEC has been pushing for the use of extensible business reporting language (XBRL) in an effort to get people to standardize data reporting electronically. A common taxonomy is needed; it is of little use if one firm calls a category one thing, and another firm calls the same category something else. XBRL will be helpful in getting data at a disaggregated level to conform to common definitions. Even if information is aggregate—such as cost-to-sales figures or on-the-job training within cost-to-sales—if it is tagged with a common definition, it may be possible to get useful data on expenditures, such as whether an asset is capitalized or expensed.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement