Using Market Data in the Supervisory Process
Introduction to the Symposium
Published September 1, 2001 | September 2001 issue
When banks take on too much risk, whole countries can suffer. The spate of banking crises over the last decade, most recently in Asia, as well as the increasing complexity and size of modern banks, have led observers to search for new methods for managing bank risk taking. The answer, according to some analysts, is to make additional use of market forces to encourage more socially optimal levels of risk taking by banks. Advocates for greater use of markets see benefit in the discipline that creditors could exert directly on banks. They also argue that bank supervisors should take advantage of the assessment of risk that market activity generates.
But not everyone agrees with this analysis. Some observers, including a number of bank supervisors, have expressed skepticism that market signals actually contain information on the risk taking of banks. And even if some information exists, they question whether these signals would enhance the ability of already informed bank supervisors to do their jobs.
To help analyze these legitimate concerns, the Federal Reserve Banks of Minneapolis and San Francisco held two conferences for bank supervisory staff on the use of market data. The first conference sponsored by the Federal Reserve Bank of San Francisco in the summer of 2000 primarily examined the question of information in market signals. Based on existing research, presenters at the conference concluded that market data provide information on the risk taking of banks.
This leaves the second, more specific, concern about the utility of market data for supervisors. The Minneapolis and San Francisco Federal Reserve banks sponsored a follow-up conference in May 2001 on the practical use of market data in the supervisory process, which is described in the following three articles.
First, in his conference address, Federal Reserve Chairman Alan Greenspan unequivocally called for increased review of market data with the expectation that such data would play a larger future role in the supervisory process. He described the conference as:
... part of an ongoing and crucial conversation that, I am confident, will move our banking and supervisory systems toward increased reliance on market discipline and use of market data in supervision. The alternative, I fear, would be the expansion of invasive and burdensomeand less effectivesupervision and regulation.
Federal Reserve Bank of Minneapolis President Gary Stern seconded Greenspan's expectations, arguing that blanket rejections of market signals were suspect because such signals represent the collective views of millions of investors with money at risk, can reduce the uncertainty of assessing bank risk taking and are available for analysis at a relatively low cost.
The final article more directly responds to concerns about the utility of market data in the supervisory process. Supervisors face real challenges in using market data because, in part, obtaining real-time, easy-to-interpret risk assessments from market data is not always easy. That said, the authors, Ron Feldman and Mark Levonian, argue that a number of practical uses of market data in the supervisory process appear reasonable to pursue in the short term on cost/benefit grounds (even if net benefits are not overwhelming). They conclude that additional applied research and increased use of market data offer the best way to determine its full potential.