Research Subcommittee: FR-14 Recommendation
Published: August 1, 2013
The OFR should do whatever it takes to gain access to the data collected by the Federal Reserve, the OCC and the FDIC to execute the Dodd Frank mandated stress tests. These data are collected on forms FR-14 (Y,Q and M) and DFAST-14.
Section 154(c)(1)(D) of the Dodd-Frank Act requires OFR “to evaluate and report on stress tests or other stability-related evaluations of financial entities overseen by Council member agencies.” Our subcommittee feels this can only be done if the OFR has the underlying data that the other agencies are using. There are also other potential uses of these data that dovetail with the OFR’s mission.
Dodd Frank mandates that the primary supervisors of critical financial institutions undertake regular stress tests. Section 154(c)(1)(D) of the Dodd-Frank Act requires OFR “to evaluate and report on stress tests or other stability-related evaluations of financial entities overseen by Council member agencies.”
To this end it is essential that the OFR is able to access the data that are analyzed by the primary supervisors. Besides allowing the OFR to improve stress testing methodologies, these data have other independent potential uses that would assist the OFR in its overall mission. For example, the data could be used in analyzing counterparty risks, mapping network linkages, estimating cross-system leverage, assessing vulnerabilities in consumer credit and corporate lending, and understanding operational risks.
The OFR needs to develop a way for outside scholars to collaborate with OFR staff and to undertake individual research projects that are of core importance to the OFR. The FRAC has identified a range of candidate scholars and research designs that comprehensively cover the type of visitors and collaborators that should be covered under this program. After designing the appropriate programs, the OFR should advertise their existence and develop the infrastructure to evaluate proposals promptly.
Engaging with outsiders is a way for the OFR to bring in additional expertise to assist in the mission of the OFR. We envisage various types of researchers who might want to collaborate and suggested that the OFR develop programs to work with each of the different types of requests that might arise.
There is very high interest among academics in working on problems that are critical to the mission of the OFR. In many cases, the data that the OFR has are unique. Hence the OFR has the opportunity to attract a range of people to collaborate with or do individual research projects that will advance the OFR’s mission. In many cases, the financial commitment from the OFR for these projects would be minimal and would allow the OFR to supplement its staff expertise with outstanding scholars. But for this to take place, the OFR needs a well-designed program in place to make it easy for visitors to apply for support. Decisions on offering support also need to be made promptly.
The OFR should study the process that the Census bureau uses in allowing researchers to access their data. That program is long-established and has had numerous participants. A variant of that program could be adapted for the OFR’s needs.
In designing a program, we envision five types of requests that the OFR should be prepared to evaluate and respond to.
First, it should be possible to accommodate a PhD student who wants to spend the summer starting a project at the OFR working full time and then continuing work over the fall on occasional visits. (Note PhD students often are not employees of a University so their time cannot be bought through an agreement with the University; they also may not have a green card.) The student may or may not need any financial support to do this.
Second, it should be possible for a PhD student who wants to spend the summer and much of the next academic year working on what will be their dissertation to do so. The student will require at least financial partial support, and so would need to be hired in some capacity.
Third, it should be possible for a professor to come for a series of short visits to work on a specific project. In many of these cases, no salary support is needed, but reimbursement for travel expenses would be necessary. It is likely that these faculty members cannot spend an extended period in Washington DC (or New York), so the program must not require too much residency time.
Fourth, a faculty member could want to take a leave from his/her position for a year and work at the OFR. They would need an appointment at the OFR in some capacity.
Finally, it is possible that a professor might want to work on a theoretical project of interest to the OFR. This does not require any time in Washington to complete the project, nor does it require access to any OFR data. The professor would like salary support for the summer. The professor would also like to hire a PhD student at her university to work as a research assistant (on site at the university) and to buy a computer and some software to do some of the analysis. This kind of a team project should be possible.
A first step in getting this started would be to create a document that could be used to describe the steps (and restrictions) that would be involved for someone in each of these five scenarios to make a proposal to the OFR.
A second step would be to describe the process that the OFR would use in evaluating proposals.