Donihue begins every fall semester by assigning each student a series of real-world data from sources like the U.S. Census Bureau or the Bureau of Economic Analysis. Students spend the first half of the semester learning forecasting methods, then turn in forecast reports modeled on policy briefs commonly found in government agencies, investment banks, consulting firms, and nonprofits. They also make professional presentations to the class about their findings.
"These presentations are real-world situations, where we are not a student at Colby College, but a forecasting expert presenting to either a board of executives or some other group that wants the information regarding the outlook of our series or industry," said Tanya Rosbash '08, a current member of Donihue's seminar who has been assigned to look at U.S. manufacturers' shipments of boats and ships. "This kind of simulation is not something you get in the more typical classes and will easily help all of us in anything we do post-Colby."
All this is just a precursor to the creation of the CEO, which each year includes a focus chosen by the students and often influenced by current events. Past forecasts have focused on national issues like oil prices or consumer spending; sometimes the focus is Maine-centered, such as last year, when students created a model to forecast auto sales in Maine, something Maine Revenue Services was interested in using. This year's focus hadn't been decided, but Donihue says it most likely would be a national focus, maybe the war in Iraq or something to do with the stock market. "But I could be wrong," Donihue said. "I leave that up to the students. That way they can put their personal stamp on this year's issue."
When the students wrap up the CEO in December, Donihue sends between 30 and 50 copies of the forecast to policy makers and economists in the state, as well as to the students' families and interested Colby staff. The students may have an opportunity to present the forecast to Gov. John Baldacci in person. Maine's Consensus Forecasting Commission, of which Donihue is a member, uses the Colby forecast when coming up with its own forecast of the state's economy. And depending on that year's focus, Donihue will send a few copies to colleagues in Washington, D.C., where he
was a senior economist on the Council of Economic Advisors for a year during the Clinton administration.
In general, Donihue says the forecast is well received and is usually in line with professionally produced forecasts--especially when it comes to predicting interest rates. But the students have been wrong on occassion. The most notorious faulty forecast came in 1990, when Donihue's class was putting together the CEO on the brink of the First Gulf War. There was a lot of "saber rattling" going on, Donihue recalled, and students predicted that if war broke out in the Middle East oil prices would rise to $65 a barrel and could cause a worldwide economic recession. In reality, there was a war, he said, but oil prices spiked and then went down and there was no recession. "I guess the consolation was that we weren't further off than other forecasters," Donihue said.
But Economics 473 is ultimately designed as a learning experience, so Donihue says he doesn't worry when the students get it wrong. "You're predicting the future," he said, "and by definition the future is unknown."