I spent the first two days of this week in Washington, D.C., at the 2010 National Association for Business Economics (NABE) Professional Development Seminar. Now in its seventh year, the NABE Professional Development Seminar (PDS) is a program designed to strengthen participants' knowledge of economic statistics and analytical techniques. I've been honored to present the Atlanta Fed's approach to regional economic analysis at several PDS events over the years.
This year I had the privilege of presenting with Dr. Charles Stiendel, a senior vice president at the New York Fed, and Dr. Keith Phillips, a senior economist and policy advisor at the Dallas Fed.
Dr. Stiendel discussed the Beige Book process in detail—a very well-timed presentation since today is when the Fed releases the latest report. He pointed out that while the Beige Book is driven by anecdotal reports from contacts in the business and labor community, one study has found that a quantitative classification of the Beige Book can be used to enhance near-term GDP forecasts. The Dallas Fed published a study in 1998 by Balke and Petersen that found that both in-sample and out-sample, the quantitative Beige Book indices, do have significant predictive content for current and next-quarter real GDP growth. Furthermore, the Beige Book has information about current-quarter real GDP growth not present in other indicators such as the Blue Chip Consensus Forecast or time series models that use real-time data.
The Atlanta Fed published a study in January 2003 by Ginther and Zavodny that examined whether the descriptive content of the Beige Book affects asset prices. The results indicate that more positive Beige Book reports on economic growth are associated with increases in interest rates, particularly long-term rates, even after controlling for other macroeconomic data releases. Stronger Beige Book reports are positively associated with changes in equity prices during expansions but negatively during recessions.
In his presentation, Dr. Philips presented some of the novel methods the Dallas Fed uses in measuring economic performance in Texas. Among them are several measures of state and metro area business cycles. In 2004, the Dallas Fed published Phillips' work on a New Monthly Index of the Texas Business Cycle and followed that up in 2005 by extending coverage to include major metro areas in the state.
In 2007, we at the Atlanta Fed developed our own measure of regional economic activity called the "D6 Factor." The paper that developed the D6 factor by Silos and Vilan outlines and estimates a model of the Sixth Federal Reserve District economy that provides a single economic indicator. The model assumes that the region's economic activity—measured by a large set of time series of employment, construction, earnings, and sales tax revenues—is driven by an unobserved common factor. The model incorporates disaggregated information for each state into a large model to derive a common component. Having a single economic indicator for the region allows for simpler and faster interpretation of various (sometimes contradictory) economic signals and makes comparisons with the nation's and other region's economies easier.
Whether it's the Beige Book or data-driven measures of economic activity, the Fed's regional banks provide a wealth of information to gauge regional economic performance.
By Michael Chriszt, assistant vice president in the Atlanta Fed's research department