A new working paper by Marc Joffe (Public Sector Credit Solutions) and Jacqueline Reck (University of South Florida) has been released by the Mercatus Center on the feasibility of PDF to XBRL migration of CAFRS so that their content can be quickly and automatically imported into spreadsheets and databases:

Ten years after the Securities and Exchange Commission mandated the conversion of corporate financial statements to machine-readable formats, there is still no analogous mandate for state and local government Comprehensive Annual Financial Reports (CAFRs). We explore the challenges and benefits of migrating from PDF CAFRs to machine-readable filings using eXtensible Business Reporting Language (XBRL). After explaining the benefits of machine-readable audited municipal financial data, we consider the challenges of creating and implementing an XBRL taxonomy for this sector and the impact a filing mandate would have on state and local governments. To better assess the challenges, we update a CAFR taxonomy previously published by Neal M. Snow and Jacqueline L. Reck and apply it to a city in Florida. While corporate XBRL filers generally use third-party filing firms, they can also use open-source software, low-cost licensed software, or both to produce the filings. Providing a variety of low-cost alternatives to state and local governments helps mitigate the challenge of providing affordable filings.

I was curious about the benefits of this for governments that might persuade them to start voluntarily using XBRL. That is, the benefits to academics and monitoring agencies from being able to harvest mass data are clear, but I wasn’t sure what would be in it for governments doing the filing. Indeed, as their review shows, many of the beneficiaries of this are regulators (the SEC requires XBRL in the 10-K reports for public companies for this reason), so it looks to me like voluntary provision will likely remain the exception rather than the rule.