By examining trends in undergraduate economics degrees awarded in the United States since World War II, Robert Margo and I (1996) discovered that, despite periodic cycles, the proportion of degrees awarded in economics has remained fairly steady at an average of about 2.2 percent of total degrees. Prior to the last few years, there had been two substantial drops in economics majors since World War II--during the Korean War and in the early 1970s. In both cases, the numbers rebounded. Because of this mean-reversion phenomenon, we cautioned against alarm over the consecutive 10 percent declines in economics degrees awarded in 1993 and 1994. Subsequently available data on undergraduate degrees awarded in 1995 and 1996 revealed a third consecutive decline of almost 10 percent in 1995 (Siegfried 1997). In just three years, a sample of 172 colleges and universities had lost 32 percent of their economics majors. The 1996 data suggested that self-correcting mechanisms were taking hold, however, as the slide finally moderated, to about 2 percent that year. Although I did not conduct a full-scale survey for 1997, 111 of the 172 colleges and universities in my earlier sample responded to the American Economic Association's Universal Academic Questionnaire (UAQ) in fall 1997. That questionnaire asked for the number of men and women awarded undergraduate economics degrees between July 1, 1996, and June 30, (the 1997 year). Thus I now have a seven-year panel on undergraduate economics degrees awarded at 111 institutions (Table 1). Economics degrees at the 111 institutions reporting data track the larger sample of 172 institutions closely from 1991 though 1996, with the exception of 1993. Respondents to the UAQ experienced a more severe decline in 1993 than did nonrespondents, but they recovered the following year. The degree series from the 111 respondents for and the 61 nonrespondents track quite closely from 1994 to 1996, however, suggesting that the smaller sample is reliable. (The data on M.A. public and other bachelor's private categories are derived from relatively small samples and should therefore be used with caution.) Following declines of 14, 9, and 10 percent annually for 1992, 1994, and 1995, the number of degrees in the sample of 111 colleges and universities declined further by 3 percent in 1996 (Table 2). The numbers, however, finally show