APPENDIX A
WORKSHOP AGENDA
Workshop on Improving Models of Forecasting Demand and Supply for Doctoral Scientists and Engineers
MARCH 19-20, 1998
Lecture Room
National Academy of Sciences
2101 Constitution Avenue, N.W.
Washington, DC
Office of Scientific and Engineering Personnel
National Research Council
MARCH 19, 1998
1:00pm
Welcome
Michael Teitelbaum
Alfred P. Sloan Foundation
Daniel McFadden
University of California, Berkeley
Jeanne Griffith
National Science Foundation
1:30-3:00pm
Panel I
Forecasting Models: Objectives and Approaches
What would the characteristics of good models of supply and demand for scientists and engineers be? Such models would likely recognize that adjustment can occur in three dimensions: quantity, price, and quality. To date, models have focused only on quantity. What models can actually be estimated, however, depend on what data is available, not just rich specification. This session will focus on alternative approaches to models and describe what models could be estimated, given available data. It will also look at what additional data would make possible improvement in the kinds of models used to describe adjustment in these labor markets.
Presenter:
Burt S. Barnow
Johns Hopkins University
Panel:
Ronald Ehrenberg
Cornell University
George Walker
Indiana University
John A. Armstrong
IBM, retired
Chair:
Ronald Oaxaca
University of Arizona
3:15-4:45pm
Panel II
Neglected Margins: Substitution and Quality
Much modeling of labor markets for scientists and engineers neglects price, price elasticity, substitution among different types of labor and other variables (corporate restructuring, use of different kinds of academic workers, effects of immigration, etc.). What are the external drivers of substitution? What data might serve as indicators?
Presenter:
Sherwin Rosen
University of Chicago
Panel:
Michael Finn
Oak Ridge Institute for Science and Education
Eric Weinstein
Massachusetts Institute of Technology
Paula Stephan
Georgia State University
Chair:
Daniel Hamermesh
University of Texas, Austin
5:00-7:00pm
Reception
Members Room
MARCH 20, 1998
8:30am
Continental Breakfast
Lecture Anteroom
9:00-10:30am
Panel III
Models of Scientific and Engineering Supply and Demand: History and Problems
This session will focus on shortage/surplus or “gap” models that have been estimated in the past. Have they been useful? For what purposes? Given the importance of external shocks in these markets, how should the importance of inherent uncertainty be conveyed to non-expert users? Could these models be modified to take into account the simultaneous adjustments of quantity, quality, and price in either equilibrium or adjustment models?
Presenter:
George Johnson
University of Michigan
Panel:
Geoff Davis
Dartmouth College
Charles A. Goldman
RAND
Robert Dauffenbach
University of Oklahoma
Sarah E. Turner
University of Virginia
Chair:
Brett Hammond
TIAA-CREF
10:45-12:30pm
Panel IV
Presentation of Uncertainty and Use of Forecasts with Explicit Uncertainty
What is the best way to communicate the sensitivity of model outputs to assumptions and to uncertainty? How should uncertainty be presented and explained to policymakers and others who are educated but not expert users? How and why are policymakers likely to misunderstand and “misuse” forecasts and what can be done about it by modelers/forecasters?
Presenter:
Nancy Kirkendall
Office of Management and Budget
Panel:
Daniel Greenberg
Bureau of Labor Statistics
Neal Rosenthal
Science & Government Report
Johns Hopkins University
Chair:
Caroline Hoxby
Harvard University
12:30-1:30pm
Lunch
1:30-3:00pm
Discussion Panel
Should we continue forecasting S&E supply and demand?
Should forecasts be presented differently?
Should we continue forecasting supply and demand for scientists and engineers? Should forecasts be presented differently?
Moderator:
Daniel McFadden
University of California, Berkeley
Participants:
Alexander H. Flax
Institute for Defense Analysis, retired
Skip Stiles
Committee on Science
U. S. House of Representatives
Open discussion